From cehunter at gb-x.org Mon Mar 1 02:13:21 2010 From: cehunter at gb-x.org (Christopher Hunter) Date: Mon, 01 Mar 2010 02:13:21 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <201002281904.25325.ccurrie@usa.net> References: <201002281904.25325.ccurrie@usa.net> Message-ID: <1267409601.10915.24.camel@localhost> On Sun, 2010-02-28 at 19:04 +0000, Christopher Currie wrote: > I was surprised that no-one on this thread mentioned the Linux emporium > http://www.linuxemporium.co.uk/products/desktops/. Maybe they're not cheap > enough, but you get the system with Ubuntu installed and they will customize > it for you. > > I had a Thinkpad X41` from them three years ago - though there were a few > problems of communication with them for some reason, I had good service > overall and was pleased with the machine. Wow! Just had a look at their site. I think the machines must be made with gold cases or something - their prices are bizarre. I'd be very surprised if they sold much at those prices. C. From general.mooney at googlemail.com Mon Mar 1 09:19:05 2010 From: general.mooney at googlemail.com (=?UTF-8?Q?Ciar=C3=A1n_Mooney?=) Date: Mon, 1 Mar 2010 09:19:05 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <1267409601.10915.24.camel@localhost> References: <201002281904.25325.ccurrie@usa.net> <1267409601.10915.24.camel@localhost> Message-ID: <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> Hi, > Wow! Just had a look at their site. I think the machines must be made > with gold cases or something - their prices are bizarre. I'd be very > surprised if they sold much at those prices. They are a small company with excellent customer service. I've known atleast 5 people that have bought from them and they are all extremely happy. Due to their size they cannot, unfortunately, rely on economies of scale to drive their prices down. However the stuff that they stock is usually very reliable and good quality (Thinkpads etc) which are not always cheap from many retailers. They have a history of directly supporting Free software, by which I mean they do not just pay it lip service. They have contributed code, and sponsored many Free software events (LugRadio, PyCon UK, Europython). Ciarán From johngeoffreywalker at yahoo.co.uk Mon Mar 1 09:36:28 2010 From: johngeoffreywalker at yahoo.co.uk (John G Walker) Date: Mon, 1 Mar 2010 09:36:28 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> References: <201002281904.25325.ccurrie@usa.net> <1267409601.10915.24.camel@localhost> <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> Message-ID: <20100301093628.13c1b3ff@Rosebud.Harmony> On Mon, 1 Mar 2010 09:19:05 +0000 Ciarán Mooney wrote: > > Wow! Just had a look at their site. I think the machines must be > > made with gold cases or something - their prices are bizarre. I'd > > be very surprised if they sold much at those prices. > > They are a small company with excellent customer service. I've known > atleast 5 people that have bought from them and they are all extremely > happy. Add one more to that list. Their customer service is truly excellent, including a couple of hours-worth of support more than a year after I bought my Thinkpad from them. Amongst the tips I got, incidentally, when I bought it, was to join this list, -- All the best, John From danthegeekman at googlemail.com Mon Mar 1 10:30:10 2010 From: danthegeekman at googlemail.com (Dan) Date: Mon, 1 Mar 2010 10:30:10 +0000 Subject: [Gllug] LUG 18th March Message-ID: <4e73d5ed1003010230w4de28e7cnbe89a8ba2e59cffe@mail.gmail.com> I shall be attending, hopefully bringing another as well. I have previously replied to the original post. So please don't count me twice. Regards. Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100301/b4830dc0/attachment.htm From general_email at technicalbloke.com Mon Mar 1 15:19:57 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Mon, 01 Mar 2010 15:19:57 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <20100301093628.13c1b3ff@Rosebud.Harmony> References: <201002281904.25325.ccurrie@usa.net> <1267409601.10915.24.camel@localhost> <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> <20100301093628.13c1b3ff@Rosebud.Harmony> Message-ID: <4B8BDB1D.2020205@technicalbloke.com> John G Walker wrote: > On Mon, 1 Mar 2010 09:19:05 +0000 Ciarán Mooney > wrote: > > >>> Wow! Just had a look at their site. I think the machines must be >>> made with gold cases or something - their prices are bizarre. I'd >>> be very surprised if they sold much at those prices. >>> >> They are a small company with excellent customer service. I've known >> atleast 5 people that have bought from them and they are all extremely >> happy. >> > > Add one more to that list. Their customer service is truly excellent, > including a couple of hours-worth of support more than a year after I > bought my Thinkpad from them. > > Amongst the tips I got, incidentally, when I bought it, was to join this > list, > > Their machines are VERY expensive though. I bought a Quad core 2.6Ghz machine a couple of days ago for £250 delivered. Admittedly it only had 2GB RAM, 250GB HD and a dubious power supply but even if you spent another £150 upgrading it to matching their top machine's spec that's only £400. Theirs costs £810. That's a lot for slotting the parts together and providing a years support. Of course I'm quite comfortable building and supporting my own machines so this may be worth a lot more to other people who aren't but wow, they're certainly not cheap! Roger. From richardlewis at fastmail.co.uk Mon Mar 1 15:23:12 2010 From: richardlewis at fastmail.co.uk (Richard Lewis) Date: Mon, 01 Mar 2010 15:23:12 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <4B8BDB1D.2020205@technicalbloke.com> References: <201002281904.25325.ccurrie@usa.net> <1267409601.10915.24.camel@localhost> <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> <20100301093628.13c1b3ff@Rosebud.Harmony> <4B8BDB1D.2020205@technicalbloke.com> Message-ID: <87eik4rrq7.wl%richard.lewis@gold.ac.uk> At Mon, 01 Mar 2010 15:19:57 +0000, general_email at technicalbloke.com wrote: > > John G Walker wrote: > > On Mon, 1 Mar 2010 09:19:05 +0000 Ciarán Mooney > > wrote: > > > > > >>> Wow! Just had a look at their site. I think the machines must be > >>> made with gold cases or something - their prices are bizarre. I'd > >>> be very surprised if they sold much at those prices. > >>> > >> They are a small company with excellent customer service. I've known > >> atleast 5 people that have bought from them and they are all extremely > >> happy. > >> > > > > Add one more to that list. Their customer service is truly excellent, > > including a couple of hours-worth of support more than a year after I > > bought my Thinkpad from them. > > > > Amongst the tips I got, incidentally, when I bought it, was to join this > > list, > > > > > > Their machines are VERY expensive though. I bought a Quad core 2.6Ghz > machine a couple of days ago for £250 delivered. Admittedly it only had > 2GB RAM, 250GB HD and a dubious power supply but even if you spent > another £150 upgrading it to matching their top machine's spec that's > only £400. Theirs costs £810. > Cool. Was that without the MS tax too? Any chance you could mention where from? -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Richard Lewis ISMS, Computing Goldsmiths, University of London Tel: +44 (0)20 7078 5134 Skype: richardjlewis JID: ironchicken at jabber.earth.li http://www.richard-lewis.me.uk/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- +----------------------------------------------+ | Support open access to scholarship | | http://freeculture.org/ http://www.doaj.org/ | +----------------------------------------------+ From john at sinodun.org.uk Mon Mar 1 16:10:21 2010 From: john at sinodun.org.uk (John Winters) Date: Mon, 01 Mar 2010 16:10:21 +0000 Subject: [Gllug] =?utf-8?q?recommendations_for_retailers_of_OS_free_or_lin?= =?utf-8?q?ux_installed_desktop_boxes=3F?= In-Reply-To: <009401cab845$b9dbbb80$2d933280$@co.uk> References: <1267277510.10915.8.camel@localhost> <4B892DB1.80503@shrdlu.com> <009401cab845$b9dbbb80$2d933280$@co.uk> Message-ID: <4e41df6bfc533edfdd6d4b4e044e5934@sinodun.org.uk> On Sun, 28 Feb 2010 07:14:51 -0000, "Paul Rayner" wrote: > I think the mark of a good retailer is a > good returns system. I've ordered many things from them in the past 10 > years, but once had to send back a new PC because of a fault. Despite > ordering a system without an OS and installing Fedora on it rather than > Windows, they agreed to a swap after two minutes on the phone explaining > the > fault (the person on the end of the phone knew what Linux was and > understood > my description of the fault), and the swap took place the next day free of > charge. I too have had good experiences on that front. A slightly customised box which I'd ordered arrived with no CD drive. (I discovered this after having connected it up. I bent down to put the Linux install CD in and found myself stymied.) A quick phone call to them and a slightly puzzled lady said, "That's odd. The invoice specifies a CD drive but the build list doesn't. If I send you one are you OK to put it in yourself?" I happily agreed, then took one of the 27 spare CD drives from under my workbench and put that in instead, adding the new one to the pile when it arrived. John From general_email at technicalbloke.com Mon Mar 1 17:36:27 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Mon, 01 Mar 2010 17:36:27 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <87eik4rrq7.wl%richard.lewis@gold.ac.uk> References: <201002281904.25325.ccurrie@usa.net> <1267409601.10915.24.camel@localhost> <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> <20100301093628.13c1b3ff@Rosebud.Harmony> <4B8BDB1D.2020205@technicalbloke.com> <87eik4rrq7.wl%richard.lewis@gold.ac.uk> Message-ID: <4B8BFB1B.6030606@technicalbloke.com> Richard Lewis wrote: > At Mon, 01 Mar 2010 15:19:57 +0000, > general_email at technicalbloke.com wrote: > >> John G Walker wrote: >> >>> On Mon, 1 Mar 2010 09:19:05 +0000 Ciarán Mooney >>> wrote: >>> >>> >>> >>>>> Wow! Just had a look at their site. I think the machines must be >>>>> made with gold cases or something - their prices are bizarre. I'd >>>>> be very surprised if they sold much at those prices. >>>>> >>>>> >>>> They are a small company with excellent customer service. I've known >>>> atleast 5 people that have bought from them and they are all extremely >>>> happy. >>>> >>>> >>> Add one more to that list. Their customer service is truly excellent, >>> including a couple of hours-worth of support more than a year after I >>> bought my Thinkpad from them. >>> >>> Amongst the tips I got, incidentally, when I bought it, was to join this >>> list, >>> >>> >>> >> Their machines are VERY expensive though. I bought a Quad core 2.6Ghz >> machine a couple of days ago for £250 delivered. Admittedly it only had >> 2GB RAM, 250GB HD and a dubious power supply but even if you spent >> another £150 upgrading it to matching their top machine's spec that's >> only £400. Theirs costs £810. >> >> > Cool. Was that without the MS tax too? Any chance you could mention > where from? > Yeah, no OS, it was from "Envisage" via ebay.co.uk http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=230406700579&ssPageName=STRK:MEWAX:IT 3 year warranty too! Roger. From john at cornerstonelinux.co.uk Mon Mar 1 17:59:58 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Mon, 1 Mar 2010 17:59:58 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <4B8BFB1B.6030606@technicalbloke.com> References: <201002281904.25325.ccurrie@usa.net> <1267409601.10915.24.camel@localhost> <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> <20100301093628.13c1b3ff@Rosebud.Harmony> <4B8BDB1D.2020205@technicalbloke.com> <87eik4rrq7.wl%richard.lewis@gold.ac.uk> <4B8BFB1B.6030606@technicalbloke.com> Message-ID: <20100301175958.GK28166@cornerstonelinux.co.uk> On Mon, Mar 01, 2010 at 05:36:27PM +0000, general_email at technicalbloke.com wrote: > Yeah, no OS, it was from "Envisage" via ebay.co.uk > > http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=230406700579&ssPageName=STRK:MEWAX:IT > > 3 year warranty too! That is not a full 3 year warranty: "[3] Years 2 and 3 - return to base. Labour is free but any parts that are required are chargeable." No mention of Linux anywhere, so it would be hard to return if you find that a component does not working Linux drivers (beyond the standard Distance Selling Regulations). ps. Ebay says "History: 5 sold" -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100301/e48f5c9d/attachment.pgp From lesleyb at herlug.org.uk Mon Mar 1 17:58:18 2010 From: lesleyb at herlug.org.uk ('LesleyB ') Date: Mon, 1 Mar 2010 17:58:18 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - University of Westminster In-Reply-To: <732076a81002250416o256f091as5ebea656c3ab3a5f@mail.gmail.com> References: <3e4e51a81002190140i75703effm81d6ed8f7305e0e8@mail.gmail.com> <732076a81002190307o4547539fk5cfa3530c5daec9b@mail.gmail.com> <3e4e51a81002250032hecd9c4ep6e1d0643b7041f8e@mail.gmail.com> <732076a81002250416o256f091as5ebea656c3ab3a5f@mail.gmail.com> Message-ID: <20100301175818.GA31174@pgcroft.net> On Thu, Feb 25, 2010 at 12:16:34PM +0000, Benjamin Donnachie wrote: > On 25 February 2010 08:32, Ciarán Mooney wrote: > > This is a reminder that we are planning a meeting on the 18th Match > > and UoW. Please add you name to the list so I can keep Sean Tohill up > > to date with numbers to expect. > > Can you put me down as a possible please? > ditto Lesley From andrew at osmosoft.com Mon Mar 1 19:26:35 2010 From: andrew at osmosoft.com (Andrew Back) Date: Mon, 1 Mar 2010 19:26:35 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - University of Westminster In-Reply-To: <1267244826.18913.2.camel@localhost> References: <3e4e51a81002190140i75703effm81d6ed8f7305e0e8@mail.gmail.com> <732076a81002190307o4547539fk5cfa3530c5daec9b@mail.gmail.com> <3e4e51a81002250032hecd9c4ep6e1d0643b7041f8e@mail.gmail.com> <20100225113123.GA28277@kanga.frasco> <1267244826.18913.2.camel@localhost> Message-ID: <20100301192635.GB6200@rhys.osmosoft.com> On (04:27 27/02/10), Justin Perreault wrote: > On Thu, 2010-02-25 at 11:31 +0000, Frank Scott wrote: > > On Thu, Feb 25, 2010 at 08:32:46AM +0000, Ciarán Mooney wrote: > > > > > > This is a reminder that we are planning a meeting on the 18th Match > > > and UoW. Please add you name to the list so I can keep Sean Tohill up > > > to date with numbers to expect. > > > > > > > +1 > > Me too. +1. Cheers, Andrew -- Andrew Back mailto:andrew at osmosoft.com http://carrierdetect.com From general_email at technicalbloke.com Mon Mar 1 20:33:34 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Mon, 01 Mar 2010 20:33:34 +0000 Subject: [Gllug] recommendations for retailers of OS free or linux > installed desktop boxes? In-Reply-To: <20100301175958.GK28166@cornerstonelinux.co.uk> References: <201002281904.25325.ccurrie@usa.net> <1267409601.10915.24.camel@localhost> <3e4e51a81003010119w65c3a1aei730247c23d684cfb@mail.gmail.com> <20100301093628.13c1b3ff@Rosebud.Harmony> <4B8BDB1D.2020205@technicalbloke.com> <87eik4rrq7.wl%richard.lewis@gold.ac.uk> <4B8BFB1B.6030606@technicalbloke.com> <20100301175958.GK28166@cornerstonelinux.co.uk> Message-ID: <4B8C249E.5040105@technicalbloke.com> John Edwards wrote: > On Mon, Mar 01, 2010 at 05:36:27PM +0000, general_email at technicalbloke.com wrote: > > >> Yeah, no OS, it was from "Envisage" via ebay.co.uk >> >> http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=230406700579&ssPageName=STRK:MEWAX:IT >> >> 3 year warranty too! >> > > That is not a full 3 year warranty: > "[3] Years 2 and 3 - return to base. Labour is free but any parts > that are required are chargeable." > > No mention of Linux anywhere, so it would be hard to return if you > find that a component does not working Linux drivers (beyond the > standard Distance Selling Regulations). > > > ps. Ebay says "History: 5 sold" > > > I take your point about the warranty but that's still better than you get from most retailers. Also you are technically right about returns too however... There's not much for linux to be incompatible with in a bare bones desktop anyway. The onboard graphics are nVidia which I've never had any trouble with. I have run Ubuntu live CDs on dozens if not hundreds of arbitrary machines that come to me for repair and the only thing it ever struggles with are onboard Intel graphics sometimes, I have never seen it fail to pickup LAN or audio correctly, even on laptops. Besides, if there is an issue you can always look at it as an opportunity to file a bug report and help the community fix it ;) Roger. ps: Actually the retailer has the best part of 70,000 transactions & 99.9% positive feedback. The number you're looking at is sales for that listing only. From andy at strugglers.net Tue Mar 2 01:31:18 2010 From: andy at strugglers.net (Andy Smith) Date: Tue, 2 Mar 2010 01:31:18 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - University of Westminster In-Reply-To: <3e4e51a81002190140i75703effm81d6ed8f7305e0e8@mail.gmail.com> References: <3e4e51a81002190140i75703effm81d6ed8f7305e0e8@mail.gmail.com> Message-ID: <20100302013118.GU15325@bitfolk.com> Hi, On Fri, Feb 19, 2010 at 09:40:02AM +0000, Ciarán Mooney wrote: > Could you please add you name to the attendees list to give Sean > Tohill an idea of the numbers we should be expecting. I probably will be attending. Cheers, Andy -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100302/03b338f5/attachment.pgp From phil-gllug at tinsleyviaduct.com Tue Mar 2 09:14:20 2010 From: phil-gllug at tinsleyviaduct.com (Phil Reynolds) Date: Tue, 02 Mar 2010 09:14:20 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - University of Westminster In-Reply-To: <3e4e51a81002190140i75703effm81d6ed8f7305e0e8@mail.gmail.com> References: <3e4e51a81002190140i75703effm81d6ed8f7305e0e8@mail.gmail.com> Message-ID: <20100302091420.168973bqdx1p2c4s@topdeck.tinsleyviaduct.com> Quoting "Ciarán Mooney" : > Morning, > > There will be a talk for the GLLUG by Richard Jones on libguestfs[1] > on Thursday 18th March at 1930. It will be held at the Cavandish > campus of the University of Westminster[2], in the Large Lecture > Theatre. I will try and get a more accurate description of the lecture > theatre's location and post it to the mailing list and website > If people could arrive at 1900 to make sure we are ready to begin at > 1930. After the talk I suggest we go to a decent local pub for a > social. > > Could you please add you name to the attendees list to give Sean > Tohill an idea of the numbers we should be expecting. I intend to attend. -- Phil Reynolds mail: phil-gllug at tinsleyviaduct.com Web: http://www.tinsleyviaduct.com/phil/ Waltham 66, Emley Moor 69, Droitwich 79, Windows 95 ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From troybie at gmail.com Tue Mar 2 10:17:26 2010 From: troybie at gmail.com (Troy Jendra) Date: Tue, 2 Mar 2010 10:17:26 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - University of Westminster In-Reply-To: <20100302091420.168973bqdx1p2c4s@topdeck.tinsleyviaduct.com> References: <3e4e51a81002190140i75703effm81d6ed8f7305e0e8@mail.gmail.com> <20100302091420.168973bqdx1p2c4s@topdeck.tinsleyviaduct.com> Message-ID: me too On 2 March 2010 09:14, Phil Reynolds wrote: > Quoting "Ciarán Mooney" : > >> Morning, >> >> There will be a talk for the GLLUG by Richard Jones on libguestfs[1] >> on Thursday 18th March at 1930. It will be held at the Cavandish >> campus of the University of Westminster[2], in the Large Lecture >> Theatre. I will try and get a more accurate description of the lecture >> theatre's location and post it to the mailing list and website >> If people could arrive at 1900 to make sure we are ready to begin at >> 1930. After the talk I suggest we go to a decent local pub for a >> social. >> >> Could you please add you name to the attendees list to give Sean >> Tohill an idea of the numbers we should be expecting. > > I intend to attend. > > -- > Phil Reynolds > mail: phil-gllug at tinsleyviaduct.com > Web: http://www.tinsleyviaduct.com/phil/ > Waltham 66, Emley Moor 69, Droitwich 79, Windows 95 > > > ---------------------------------------------------------------- > This message was sent using IMP, the Internet Messaging Program. > -- > Gllug mailing list  -  Gllug at gllug.org.uk > http://lists.gllug.org.uk/mailman/listinfo/gllug > From rich at annexia.org Tue Mar 2 21:52:45 2010 From: rich at annexia.org (Richard Jones) Date: Tue, 2 Mar 2010 21:52:45 +0000 Subject: [Gllug] File On 4 on waste in Govt computing projects Message-ID: <20100302215244.GA12176@annexia.org> No surprise to anyone on this list ... http://downloads.bbc.co.uk/podcasts/radio4/fileon4/fileon4_20100302-2038a.mp3 Rich. -- Richard Jones Red Hat From GLLUG at GetAroundToIt.co.uk Tue Mar 2 23:47:12 2010 From: GLLUG at GetAroundToIt.co.uk (David L Neil) Date: Wed, 03 Mar 2010 12:47:12 +1300 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <20100302215244.GA12176@annexia.org> References: <20100302215244.GA12176@annexia.org> Message-ID: <4B8DA380.7010808@GetAroundToIt.co.uk> On 03/03/10 10:52, Richard Jones wrote: > > No surprise to anyone on this list ... > > http://downloads.bbc.co.uk/podcasts/radio4/fileon4/fileon4_20100302-2038a.mp3 > > Rich. > Rich, (and GLLUGers) Many thanks. Much to my surprise I was able to hear this all the way down-under, and using the MPlayer Plug-in to Firefox under CentOS - having heard that the BBC has recently taken negative steps towards our FOSS community, to say nothing of being too shy to support expatriates or 'educate the natives' (ie imitate the grand vision of the "World Service" - the WWW ^H technology that came before the Internet...) You may be amused to hear, that whilst working on 'smoothing-out some wrinkles' for a friend I had an amusing conversation with Telstra-Clear about her Internet connection. It turns out that her connection is rated (and billed) at 256Kbps. 256K??? I chortled and questioned if that met any international standard for what is 'broadband'? This is where New Zealand was (at least) ten years ago and the accounts still exist. To 'upgrade' would only be possible by literally doubling the monthly fee (thank you for sticking with us all these years!) - or, my recommendation, moving to another supplier (BUT... one of those package deals: land-line, Internet, tolls, cells,...) Another point: this is what happens when another country's 'experts' come in and start telling 'you' what to do (see also suppliers to British gas clients, Spanish electricity companies, US helos and Euro-fighters, etc, etc) On the plus side, something I hadn't come across in the UK excepting some cellular rates, (her) traffic is paid for in 'packs' (with no monthly 'allowance' or 'maximum') - and the 'first' pack costs the same as the n-th. Thus if one downloads a new distribution .iso this month it will probably cost an extra pack/few bucks, but next month using (say) 1GB less, one pays for fewer packs. Thus one neither pays a premium rate during a 'heavy month', not for unused traffic the next! My brother-in-law has a slight variation on this. He is a professional photographer and on the days when he completes a commission at home he needs to deliver tens or even hundreds of MB, but on other days he barely makes the meter move. So he is on a plan which involves throughput-throttling once the paid-for traffic level has been exceeded, but his 'level' applies to a twelve hour period - from six to six (IIRC). Thus he rushes the first few images, and fires them off to his client (before 1800), then settles down to work on the rest, likely extending beyond six and into the new traffic period. Thereby, even during a heavy-usage month, he can still do quite well, in all aspects: financially, throughput, and traffic-wise; even when compared against (say) my (in London) commercially-oriented account with Andrews and Arnold* where I had something silly* like 10GB during peak time and 40GB more off-peak (again IIRC - and hoping I'm not confusing their plan with A.N.Other I've used in the UK). *free plug: good outfit - no connection other than as client (pardon pun!) There was NO WAY I'd ever meet those numbers during the last few years, and probably into the next couple of years... Over here the populist (current) Internet plans are essentially similar to each other, just as is in UK and particularly amongst the incumbent's re-sellers. However there are a few of these intriguingly tailored plans, particularly from the boutique outfits*, who follow the smaller-scale line of paying attention to the needs of their customers. * food for thought Jason? Whilst I'm rambling: I'm seriously thinking of not returning for a while (perhaps I'm 'cheating' more than a little, given that last week I completed a web project for my client in Durham???), but when I read: The economic forecast is simple: the next 10 years are going to be a drag | Business | The Guardian (2 March 2010) http://www.guardian.co.uk/business/2010/mar/01/drag-deficit-reduction-anaemic-growth and many similar, I can but shake my head... By comparison:- Australia (particularly Western Australia with its mineral wealth - that's Perth for those of you with a burr in your voice) is booming again, on the back of its natural resources and selling to clients in the Asian economies, and the country has just raised interest rates for the third time, so it's even worth thinking about saving some of your hard-earned... The West Australian (newspaper) carried an article last week: Another boom, another skills crisis*. *apologies can't quote source/temporarily unable to reach web site to search (see above. Telstra-Clear also seem to be practicing for the Australian National Firewall/censorship by blocking this, and also the Times and the Telegraph's sites, amongst others... as they say in New Zealand, with an exasperated tone of voice, "Aussies!") Today's reports in New Zealand are already predicting the same boom/need, for IT in particular: Computerworld > Skills shortage to return, but with variations (3 March 2010) http://computerworld.co.nz/news.nsf/careers/skills-shortage-to-return-but-with-variations I've often heard Brits comment that they can't afford to move down-under because they convert expected salaries to pounds-sterling and worry about a "drop" (change in lifestyle/better quality of life notwithstanding). I have checked-around informally, and basically the rule-of-thumb that I have offered over the years still applies: take the pound sign off and replace it with a dollar sign - thus the salary seems lower in pound terms, but then so is the cost of your grocery bill, by the same proportion! Apart from heavy (weight) imported goods, it pretty much all comes out in the wash... (the last less of a concern in Australia, but they don't have an NHS system so living costs are higher) Both countries have requirements for immigration. Again anecdotally (because I qualify for residence for other reasons) I laid hands on a write-up of the 'points system' for New Zealand immigration and kept a mental count as I read through. Simply having a degree and years of 'experience' piled up most of the points necessary to 'qualify', so by the time I added previous residence and various other attributes (my astonishingly good looks?) had no trouble meeting the requirements - YMMV, but for fellow professionals I would suggest, not by much! Should I set up an in-bound residents' or travel service for GLLUG? Keep warm! Regards, =dn PS glad to hear that a meeting has been set up. Sorry I can't be there/contribute. Well done Ciarán! From jjllmmss at googlemail.com Wed Mar 3 02:45:07 2010 From: jjllmmss at googlemail.com (JLMS) Date: Wed, 3 Mar 2010 02:45:07 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <4B8DA380.7010808@GetAroundToIt.co.uk> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> Message-ID: On Tue, Mar 2, 2010 at 11:47 PM, David L Neil wrote: > On 03/03/10 10:52, Richard Jones wrote: >> >> No surprise to anyone on this list ... >> >> http://downloads.bbc.co.uk/podcasts/radio4/fileon4/fileon4_20100302-2038a.mp3 >> >> Rich. >> > > > Rich, (and GLLUGers) > > Many thanks. > > Much to my surprise I was able to hear this all the way down-under, and > using the MPlayer Plug-in to Firefox under CentOS - having heard that > the BBC has recently taken negative steps towards our FOSS community, to > say nothing of being too shy to support expatriates or 'educate the > natives' (ie imitate the grand vision of the "World Service" - the WWW > ^H technology that came before the Internet...) > > I have a Kiwi colleague, working in London, he could not find a job down under and had to come back to the UK for work, he said the situation down under for IT professionals is dire (or do you think they are immune to outsourcing to India and other economic pressures? They even are in a similar time zone as the Philippines, Singapore and other "next generation" outsourcing hotspots which makes things particularly challenging). This friend is no dumber, has worked in top companies in the UK for several years. As for economists predicting how the UK (or Aussie or Kiwi) economy will do in the next 10 years, phwa!, like if they really knew what they are talking about (have we forgotten how little foresight about their own field of "expertise" these fellows have in general terms?) From cehunter at gb-x.org Wed Mar 3 08:10:13 2010 From: cehunter at gb-x.org (Christopher Hunter) Date: Wed, 03 Mar 2010 08:10:13 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> Message-ID: <1267603813.10915.73.camel@localhost> On Wed, 2010-03-03 at 02:45 +0000, JLMS wrote: > On Tue, Mar 2, 2010 at 11:47 PM, David L Neil wrote: > > On 03/03/10 10:52, Richard Jones wrote: > >> > >> No surprise to anyone on this list ... > >> > >> http://downloads.bbc.co.uk/podcasts/radio4/fileon4/fileon4_20100302-2038a.mp3 > >> > >> Rich. > >> > > > > > > Rich, (and GLLUGers) > > > > Many thanks. > > > > Much to my surprise I was able to hear this all the way down-under, and > > using the MPlayer Plug-in to Firefox under CentOS - having heard that > > the BBC has recently taken negative steps towards our FOSS community, to > > say nothing of being too shy to support expatriates or 'educate the > > natives' (ie imitate the grand vision of the "World Service" - the WWW > > ^H technology that came before the Internet...) > > > > > > > I have a Kiwi colleague, working in London, he could not find a job > down under and had to come back to the UK for work, he said the > situation down under for IT professionals is dire (or do you think > they are immune to outsourcing to India and other economic pressures? > They even are in a similar time zone as the Philippines, Singapore and > other "next generation" outsourcing hotspots which makes things > particularly challenging). This friend is no dumber, has worked in > top companies in the UK for several years. That's pretty much the story I've heard too. Friends of mine (a married couple from Melbourne) who are both coders, cannot find properly paid work anywhere south of the Equator, and are both taking crash courses in French, German and Spanish so they can work in Europe. They have a project for a French car maker at the moment, but that finishes this summer, so their language learning is now taking up all their spare time! > As for economists predicting how the UK (or Aussie or Kiwi) economy > will do in the next 10 years, phwa!, like if they really knew what > they are talking about (have we forgotten how little foresight about > their own field of "expertise" these fellows have in general terms?) It's pretty much a given that the current crowd of criminals have pi**ed away any wealth this country ever had and run the economy so far into the red that we will never recover, unless the debts are written off - like they have been for so many third world countries. Financially, we're worse off than much of the third world - it's just that we have the "financial acumen" to hide the fact. Gordon Brown's ONLY ability is being able to lie - about the economy, about his own snout in the trough, about his personal dealings with the staff forced to work for him, and even about his own sexuality... This country no longer has any kind of manufacturing base worth mentioning - it was killed by the massive inflation through the 60s and 70s and by trades unions making British companies uncompetitive against even our immediate neighbours, let alone the rest of the world. Government support for industry became non-existent, so all the major manufacturers moved abroad. We used to lead the world in science, technology and engineering, but nowadays we even fall way below the Italians! Our education system is a poor joke, with ever falling standards, teachers not actually allowed to teach properly, "dumbed-down" examinations, worthless "degrees" from bogus "Universities" and a significant proportion (roughly 40%) of the population who are functionally illiterate and innumerate. The last thing a government like this wants is an educated populace who are able to see through the lies and evasions.... Now - here's the problem: I've lived and worked in many parts of the world (I'm 51, and have been a professional electronic engineer for 28 years). I can't find anywhere else I'd rather live. I live in a pleasant part of North London, in a house I own, with my wife. I have a challenging job that's reasonably well paid and that I really enjoy (despite all the frustrations of dealing with Government departments). Life's pretty good (although I'm not in the best of health because of a dodgy heart). England's still a pretty good place to live - despite the lousy weather, inept government and rapidly rising crime rate. I do have the choice to leave - Chartered Engineer status gets you into any country in the world - but I'm going to stick it out. The next Governmental crowd of criminals can't be worse than the current lot, and the economy will always recover somewhat. I'll probably lose money on my house in the long run, but it's not meant as an investment, it's to keep the rain off my head and my belongings off the street. Trust me - from one who's been there frequently -"down under" isn't much better than up here. There aren't all that many reasonably well paid jobs, and prices are rising markedly as their economy suffers too. NZ is in a better state than Australia, but I don't want to move to "Wales with a funny accent" (my daughter's description on her first visit there when she was 7!). Chris From walter.stanish at saffrondigital.com Wed Mar 3 10:02:43 2010 From: walter.stanish at saffrondigital.com (Walter Stanish) Date: Wed, 3 Mar 2010 10:02:43 -0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste inGovt computing projects In-Reply-To: References: <20100302215244.GA12176@annexia.org><4B8DA380.7010808@GetAroundToIt.co.uk> Message-ID: > I have a Kiwi colleague, working in London, he could not find a job > down under and had to come back to the UK for work, he said the Couldn't resist chiming-in on this one, and the post grew... Quick work background: Australia (3 years), China (8 years), UK (1 year). Being from down under and have always felt that both the number of opportunities and the salaries are much lower there than the UK. I think it's fair to say that NZ is generally a further (small) step down in salary compared to Australia, though I don't know about demand over there. London is simply ridiculous for salaries, though it's largely offset by the crap weather and living conditions. Australia and New Zealand have brilliant living conditions, preferable weather and enough space so that you don't have to see/hear your neighbours (cramped London 'studio' apartment, anyone?) As a new arrival to London in early 2009, I was offered literally 2x the salary in GBP for work here than comparable opportunities on the mainland (Holland, Spain), which was my main motivator to choose London. Having said that, the last year I worked in Oz was 2001 and even as a youngster (19 at the time) I was able to travel extensively across the region (India, Philippines, Singapore, Taiwan, etc.) on business, which was very lucky and interesting way start a career. The Australian IT professionals I know tend to travel or relocate extensively, to places like the US, France, Turkey, Japan, etc. About competition... I don't think there is necessarily that much direct competition Between NZ/Australia and Asia at the moment. Places like Singapore and Hong Kong are very expensive to maintain a physical presence in (otherwise everyone would do it - Hong Kong has great income tax laws, and Singapore has great corporate tax laws). Secondly, Singapore is not particularly sought after as a location for living. (Fact: many of the locals hate it and find it hideously expensive to live in - which has led the government to directly to the interesting policy of paying people who have children!) In terms of software 'offshoring', I think the hype around this trend has started to die out as businesses begin to recognise that cost savings come with strings attached, and that dependable and technically competent management are an absolute must. By way of example, last year in London I interviewed for a role at a very significant web-based business that pretty much owned a particular vertical market. Their systems had been developed via 'offshoring' in the Philippines over the course of years. During this time, the codebase had gone from ample and rapidly delivered to unmaintainable, unresponsive mess on which they were inextricably dependant. Having established positive cashflow and made a fair amount of money, the management now wanted someone to build and manage a team to rewrite the whole thing from scratch in London. Looking back, I'm glad I didn't take that job. Why had it failed? Disconnection of technically competent management from the real business, and perhaps failure to adequately communicate between the technical and management teams. In this very real case, it is important to note that the standard of English for programmers in the Philippines, relatively speaking within the region's dominant offshoring locations, is *extremely* high, with competition perhaps only from India. In China I managed some small-scale offshore development for desktop software aimed at the European market, which would have been literally impossible to do with Chinese developers if I had not spoken decent Chinese and actually lived there awhile. That's not to say that there are no Chinese developers with brilliant English (far from it, though to generalise I think it's fair to say that the bulk younger educated Chinese have limited fluency) just that the cultural tendency towards working in groups with close management (vs. working more independently) does not lend itself to remote management of software projects. This can be overcome, but it's a real factor. Interestingly, in that case the end client was French! The point of this example is that many 'offshoring' Locations tend to have both language and culture barriers. Then of course, there's infrastructure reliability concerns, timezone issues, etc., etc. Back to Australia to end my rant: We have a very high representation in FOSS as well as software, computing and science in general relative to our population. Some qiuck examples: Samba + rsync (Andrew Tridgell), and netfilter (Rusty Russel), enlightenment (rasterman), heartbeats/clustering/HA Linux (Simon Horman), some of the early core VMWare developers, and I have often pondered why this is. I think it may be that the availability of high quality education, decent internet connectivity, an economy that provides realistic career options for those entering computing, enough physical space to maintain peace of mind, and finally to risk a bit of a political statement: social security (Tridge initially wrote Samba on social benefits, my taxes well spent!!!!). There's probably a lot of other factors, too. One might even say "all hail Kevin Rudd, great helmsman!" In closing to demonstrate the sincerity of my tongue-in-cheek politicking, the next stop for me (leaving in two weeks) is the GOVERNATOR's great socialist heartland of Los Angeles! Social security for the rich and famous. - Walter PS: Sorry I can't make the meeting on the 18th, would have been great to meet some more of you than the last time, but my flight's on the 17th. PPS: The above is full of rough generalisations from my own Experience ... please take them as such! From general_email at technicalbloke.com Wed Mar 3 13:40:44 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Wed, 03 Mar 2010 13:40:44 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <1267603813.10915.73.camel@localhost> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> Message-ID: <4B8E66DC.9010701@technicalbloke.com> Christopher Hunter wrote: > On Wed, 2010-03-03 at 02:45 +0000, JLMS wrote: > >> On Tue, Mar 2, 2010 at 11:47 PM, David L Neil wrote: >> >>> On 03/03/10 10:52, Richard Jones wrote: >>> > It's pretty much a given that the current crowd of criminals have pi**ed > away any wealth this country ever had and run the economy so far into > the red that we will never recover, unless the debts are written off - > like they have been for so many third world countries. Financially, > we're worse off than much of the third world - it's just that we have > the "financial acumen" to hide the fact. Gordon Brown's ONLY ability is > being able to lie - about the economy, about his own snout in the > trough, about his personal dealings with the staff forced to work for > him, and even about his own sexuality... > Rubbish, you're being utterly melodramatic. We've been far worse off than this before and recovered and we will again. Making a comparison between us and third world countries (itself a bit of a misnomer these days) in need of debt relief is totally OTT. The countries the G7 granted debt relief to had yearly interest payments in excess of their GDP, our payments are about 13% and this is a bad year. I don't see how you can claim with a straight face that "we're worse off than much of the third world" when we so objectively aren't, seriously how many children have you lost to malaria and typhoid eh? And what the hell does Gordon Brown's sexuality have to do with anything?! > This country no longer has any kind of manufacturing base worth > mentioning - it was killed by the massive inflation through the 60s and > 70s and by trades unions making British companies uncompetitive against > even our immediate neighbours, let alone the rest of the world. > Government support for industry became non-existent, so all the major > manufacturers moved abroad. We used to lead the world in science, > technology and engineering, but nowadays we even fall way below the > Italians! > Well if British manufacturing had survived the 70s completely unscathed it still would have been royally screwed by Deng Xiaoping a few years later anyway so it matters little, it's not like Britain is unique in having lost manufacturing capacity. While there is such an enormous disparity in labour costs across the world you will always see basic manufacturing moving to where it is cheapest, it has little or nothing to do with unions, they are even more powerless than the government to stop that and really, why should they?... There's a school of though that runs: If manufacturing needs supporting then sod it, it's costing us more money than it makes. Really Chris where is the difference between government "support" (aka subsidy) of the failing manufacturing industries and the welfare state you so despise? Government support of an industry means taking tax money from individuals and spending it propping up businesses that cannot sustain themselves within an open market. Personally I can see a case for this from time to time, but not often and I am VERY surprised to see you are in favour of it, given our previous conversations. What you're basically saying is "My countrymen and I are unwilling to pay more for British made products therefore we would like to be taxed more and that money given to British manufacturers." - you see how that kind of breaks the virtuous circle? > years). I can't find anywhere else I'd rather live. > > So despite being in a terrible, sub Italian, mess it's still the best country in the world? Phew! ;) Roger. From iain at shihad.org Wed Mar 3 13:48:14 2010 From: iain at shihad.org (Iain M Conochie) Date: Wed, 03 Mar 2010 13:48:14 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <4B8E66DC.9010701@technicalbloke.com> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <4B8E66DC.9010701@technicalbloke.com> Message-ID: <4B8E689E.7010801@shihad.org> > > Really Chris where is the difference between government "support" (aka > subsidy) of the failing manufacturing industries and the welfare state > you so despise? A: 1 is corporate socialism which is perfectly acceptable. The other is socialism which is so obviously wrong ;) Cheers Iain > Government support of an industry means taking tax money > from individuals and spending it propping up businesses that cannot > sustain themselves within an open market. Personally I can see a case > for this from time to time, but not often and I am VERY surprised to see > you are in favour of it, given our previous conversations. What you're > basically saying is "My countrymen and I are unwilling to pay more for > British made products therefore we would like to be taxed more and that > money given to British manufacturers." - you see how that kind of breaks > the virtuous circle? > > >> years). I can't find anywhere else I'd rather live. >> >> >> > So despite being in a terrible, sub Italian, mess it's still the best > country in the world? Phew! ;) > > > Roger. > From sanelson at gmail.com Wed Mar 3 16:26:08 2010 From: sanelson at gmail.com (Stephen Nelson-Smith) Date: Wed, 3 Mar 2010 16:26:08 +0000 Subject: [Gllug] [OT] Disk Encryption Message-ID: Hi, I have a number of Macbooks which contain commercially sensitive information, and which are sometimes taken off site. I've been asked to provide disk-based encryption to protect the data in the event of loss or theft. Have any of you done this before? I'm currently looking at: * http://www.pgp.com/products/wholediskencryption/ * FileVault * Knox Anything else I should look at? Any success or war stories to share? TIA, S. -- Stephen Nelson-Smith Technical Director Atalanta Systems Ltd www.atalanta-systems.com From andy at andymillar.co.uk Wed Mar 3 16:27:36 2010 From: andy at andymillar.co.uk (Andy Millar) Date: Wed, 3 Mar 2010 16:27:36 -0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: References: Message-ID: <1267633656.26161.25.camel@millaralpt.co.tradefair> On Wed, 2010-03-03 at 16:26 +0000, Stephen Nelson-Smith wrote: > > Anything else I should look at? Any success or war stories to share? > We've found Truecrypt ( http://www.truecrypt.org/ ) to be good for Windows, and they also have a Mac build. Andy -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 5162 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100303/5665a7ad/attachment.bin From steve.harris at garlik.com Wed Mar 3 16:34:13 2010 From: steve.harris at garlik.com (Steve Harris) Date: Wed, 3 Mar 2010 16:34:13 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: References: Message-ID: <9D82341D-86A8-42CF-B50C-9068121A591E@garlik.com> On 3 Mar 2010, at 16:26, Stephen Nelson-Smith wrote: > Hi, > > I have a number of Macbooks which contain commercially sensitive > information, and which are sometimes taken off site. > > I've been asked to provide disk-based encryption to protect the data > in the event of loss or theft. > > Have any of you done this before? I'm currently looking at: We handle a lot of /very/ sensitive data in my company and we use FileVault. Users have to be disciplined not to keep data outside of their homedir (but note that random bits of the FS are mapped there anyway), but the UI encourages that anyway. It works very well, and works securely with TimeMachine, which is critical. - Steve -- Steve Harris, Garlik Limited 2 Sheen Road, Richmond, TW9 1AE, UK +44 20 8973 2465 http://www.garlik.com/ Registered in England and Wales 535 7233 VAT # 849 0517 11 Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10 9AD From sanelson at gmail.com Wed Mar 3 16:39:43 2010 From: sanelson at gmail.com (Stephen Nelson-Smith) Date: Wed, 3 Mar 2010 16:39:43 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <9D82341D-86A8-42CF-B50C-9068121A591E@garlik.com> References: <9D82341D-86A8-42CF-B50C-9068121A591E@garlik.com> Message-ID: Hi Steve, >> I have a number of Macbooks which contain commercially sensitive >> information, and which are sometimes taken off site. >> >> I've been asked to provide disk-based encryption to protect the data >> in the event of loss or theft. >> >> Have any of you done this before?  I'm currently looking at: > > We handle a lot of /very/ sensitive data in my company and we use > FileVault. Users have to be disciplined not to keep data outside of > their homedir (but note that random bits of the FS are mapped there > anyway), but the UI encourages that anyway. > > It works very well, and works securely with TimeMachine, which is > critical. Thanks. This was my first thought too. I'm going to do some tests shortly with Knox and PGP too. My biggest concern is performance. Could you explain the issues around working with TimeMachine? Do some products not play nicely? S. From steve.harris at garlik.com Wed Mar 3 16:44:40 2010 From: steve.harris at garlik.com (Steve Harris) Date: Wed, 3 Mar 2010 16:44:40 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: References: <9D82341D-86A8-42CF-B50C-9068121A591E@garlik.com> Message-ID: <522FFD15-02C4-4639-B3C5-6576C90ADAB9@garlik.com> On 3 Mar 2010, at 16:39, Stephen Nelson-Smith wrote: > Hi Steve, > >>> I have a number of Macbooks which contain commercially sensitive >>> information, and which are sometimes taken off site. >>> >>> I've been asked to provide disk-based encryption to protect the data >>> in the event of loss or theft. >>> >>> Have any of you done this before? I'm currently looking at: >> >> We handle a lot of /very/ sensitive data in my company and we use >> FileVault. Users have to be disciplined not to keep data outside of >> their homedir (but note that random bits of the FS are mapped there >> anyway), but the UI encourages that anyway. >> >> It works very well, and works securely with TimeMachine, which is >> critical. > > Thanks. This was my first thought too. I'm going to do some tests > shortly with Knox and PGP too. My biggest concern is performance. > > Could you explain the issues around working with TimeMachine? Do some > products not play nicely? Yes, some products will backup the data on the user's machine unencrypted, just through the loopback mount device, or will not be able to do incremental backups. What FileVault does is chunk up the crypted data into 10MB (IIRC) chunks, and back up those than have changed. Since I looked other systems might have started to do this too though. - Steve -- Steve Harris, Garlik Limited 2 Sheen Road, Richmond, TW9 1AE, UK +44 20 8973 2465 http://www.garlik.com/ Registered in England and Wales 535 7233 VAT # 849 0517 11 Registered office: Thames House, Portsmouth Road, Esher, Surrey, KT10 9AD From martin at hinterlands.org Wed Mar 3 19:59:07 2010 From: martin at hinterlands.org (Martin A. Brooks) Date: Wed, 3 Mar 2010 19:59:07 -0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: References: Message-ID: <7120e0ee4bdd18c9f1740108b924693f.squirrel@olga.hinterlands.org> On Wed, March 3, 2010 4:26 pm, Stephen Nelson-Smith wrote: > Anything else I should look at? Any success or war stories to share? TrueCrypt. From nix at esperi.org.uk Wed Mar 3 22:28:10 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 03 Mar 2010 22:28:10 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <1267603813.10915.73.camel@localhost> (Christopher Hunter's message of "Wed, 03 Mar 2010 08:10:13 +0000") References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> Message-ID: <87635df3b9.fsf@spindle.srvr.nix> On 3 Mar 2010, Christopher Hunter stated: > away any wealth this country ever had and run the economy so far into > the red that we will never recover, unless the debts are written off - > like they have been for so many third world countries. Financially, Rubbish. The debts will inflate away and be grown away. That's what sovereign debts *do*. They never *are* paid off in toto -- the financial system *depends* on their never being paid off in toto. (Also, we will recover: look at the maturity dates on those bonds and see you in fourteen years.) > the "financial acumen" to hide the fact. Gordon Brown's ONLY ability is > being able to lie - about the economy, about his own snout in the > trough, about his personal dealings with the staff forced to work for > him, and even about his own sexuality... Er, you *do* know politicians have access to libel lawyers, right? You sound more like a paranoid schizophrenic than anything else at this point. You could at least have quoted him coming out with something utterly unbelievable: he's a politician so it's not rare or hard to find. Generally his factual utterances aren't untruthful, just carefully partial: the so-wrong-it's-hilarious stuff has to come from his occasional self-assessments. *Nobody* at the top of politics can *ever* give a candid self-assessment, and this is as true of Brown as of anyone else. ('I?ve got no personal animosity towards anybody in politics' was a particularly good one, from a recent Economist interview. Amazingly the interviewer did not explode in laughter the instant he said this.) > This country no longer has any kind of manufacturing base worth > mentioning - it was killed by the massive inflation through the 60s and > 70s and by trades unions making British companies uncompetitive against > even our immediate neighbours, let alone the rest of the world. Why not just say 'I support the Tories' and not bother saying anything else? (Only actual Tories would probably run screaming from what you just wrote above because it's so politically insensitive.) [snip more overdone Tory propaganda: I know people in Conservative Central Office and *they* don't go this far over the top; they certainly don't think the last thirteen years have been unadulterated disaster the way you seem to] > rapidly rising > crime rate. FAIL. Let's see what a well-known commie pinko rag mentioned above has to say about this 'rapidly rising crime rate', shall we? The British Crime Survey agrees (but of course you don't believe that because it's run by the government, and you don't believe anything else because it could have been bought: who *do* you believe? Conservative Central Office, yes, but anyone else?) -- N., Tory voter in the last election but you're making me think again From cehunter at gb-x.org Thu Mar 4 00:12:14 2010 From: cehunter at gb-x.org (Christopher Hunter) Date: Thu, 04 Mar 2010 00:12:14 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <87635df3b9.fsf@spindle.srvr.nix> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> Message-ID: <1267661534.1783.99.camel@localhost> On Wed, 2010-03-03 at 22:28 +0000, Nix wrote: > Er, you *do* know politicians have access to libel lawyers, right? Let him press on. There's MUCH worse being said about him and even TO him at the moment. He's just as much of a criminal as Blair, and both should be in the Scrubs (as should many of the rest of them). C. From cehunter at gb-x.org Thu Mar 4 00:27:42 2010 From: cehunter at gb-x.org (Christopher Hunter) Date: Thu, 04 Mar 2010 00:27:42 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <87635df3b9.fsf@spindle.srvr.nix> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> Message-ID: <1267662462.1783.114.camel@localhost> On Wed, 2010-03-03 at 22:28 +0000, Nix wrote: > Rubbish. The debts will inflate away and be grown away. Not this time. There's nowhere for the hole to disappear to, this time. > Why not just say 'I support the Tories' and not bother saying anything > else? Actually, I don't. The current crop of Tories seem as bereft of notion as the other lot. > I know people in Conservative Central Office and *they* don't go this > far over the top; they certainly don't think the last thirteen years > have been unadulterated disaster the way you seem to. The people in CCO go WAY beyond anything I've said... and the last 13 years HAVE been an almost entirely unmitigated disaster. NOTHING has improved, and we all had our Pensions stolen and the gold reserve given away. Wasn't he a "prudent" Chancellor? > > rapidly rising > > crime rate. > > FAIL. Let's see what a well-known commie pinko rag mentioned above has > to say about this 'rapidly rising crime rate', shall we? > > > > The British Crime Survey agrees (but of course you don't believe that > because it's run by the government, and you don't believe anything else > because it could have been bought: who *do* you believe? Conservative > Central Office, yes, but anyone else?) I actually believe the unmodified reported crime figures I saw recently. The Police dare not actually reveal anything like the real figures or their top brass will all be looking for jobs - there is an endemic culture of manipulated and frankly fabricated figures supplied to government. This is true of other government departments too. Most are scared to reveal the truth of their situations because they'll be seen as failing. Clue: Why do you think Ofsted are required to give notice of inspection of schools? This is just one example of the stupid, bureaucratic nonsenses foisted upon us by these bogus champagne socialists (like the "man of the people" Michael Foot - died today in his £2m hovel in Hampstead). > -- N., Tory voter in the last election but you're making me think again You're going to get them this time, most likely. C. From nix at esperi.org.uk Thu Mar 4 07:43:58 2010 From: nix at esperi.org.uk (Nix) Date: Thu, 04 Mar 2010 07:43:58 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <1267662462.1783.114.camel@localhost> (Christopher Hunter's message of "Thu, 04 Mar 2010 00:27:42 +0000") References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> Message-ID: <87wrxsedkx.fsf@spindle.srvr.nix> On 4 Mar 2010, Christopher Hunter uttered the following: > On Wed, 2010-03-03 at 22:28 +0000, Nix wrote: > >> Rubbish. The debts will inflate away and be grown away. > > Not this time. There's nowhere for the hole to disappear to, this time. What? Inflation and growth can dispose of *any* debt, given time, though if it goes too far your currency implodes into hyperinflation (which also disposes of the debt, although as it also disposes of your entire nation's savings it tends to do your political prospects no good at all). We are very, very far from hyperinflation right now. From peterachilds at gmail.com Thu Mar 4 10:36:44 2010 From: peterachilds at gmail.com (Peter Childs) Date: Thu, 4 Mar 2010 10:36:44 +0000 Subject: [Gllug] [OT] Re: [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <87wrxsedkx.fsf@spindle.srvr.nix> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> Message-ID: On 4 March 2010 07:43, Nix wrote: > On 4 Mar 2010, Christopher Hunter uttered the following: > >> On Wed, 2010-03-03 at 22:28 +0000, Nix wrote: >> >>> Rubbish. The debts will inflate away and be grown away. >> >> Not this time.  There's nowhere for the hole to disappear to, this time. > > What? Inflation and growth can dispose of *any* debt, given time, though > if it goes too far your currency implodes into hyperinflation (which > also disposes of the debt, although as it also disposes of your entire > nation's savings it tends to do your political prospects no good at > all). > > We are very, very far from hyperinflation right now. > -- Does that work with personal debt too? Peter From john at cornerstonelinux.co.uk Thu Mar 4 11:19:06 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Thu, 4 Mar 2010 11:19:06 +0000 Subject: [Gllug] OT: Debt (was: [Long] Down-under response) In-Reply-To: References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> Message-ID: <20100304111906.GQ28166@cornerstonelinux.co.uk> On Thu, Mar 04, 2010 at 10:36:44AM +0000, Peter Childs wrote: > On 4 March 2010 07:43, Nix wrote: >> On 4 Mar 2010, Christopher Hunter uttered the following: >> >>> On Wed, 2010-03-03 at 22:28 +0000, Nix wrote: >>> >>>> Rubbish. The debts will inflate away and be grown away. >>> >>> Not this time.  There's nowhere for the hole to disappear to, this time. >> >> What? Inflation and growth can dispose of *any* debt, given time, though >> if it goes too far your currency implodes into hyperinflation (which >> also disposes of the debt, although as it also disposes of your entire >> nation's savings it tends to do your political prospects no good at >> all). >> >> We are very, very far from hyperinflation right now. > > Does that work with personal debt too? Yes. Inflation reduces the value of the currency and hence the value of the debt. But it erodes both debt and savings, and has strong feedback effects. Growth on the personal level would be an increase in income. There are other ways of reducing debt, both national and personal (or even football club), such as the sale of assets (eg banks) and reduction of costs. A debt larger than income is sustainable as long as there is sufficient assets and the debt is able to be steadily paid off. A new mortage is many times higher than your salary. -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100304/86d62f04/attachment.pgp From benjamin at py-soft.co.uk Thu Mar 4 11:43:40 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Thu, 4 Mar 2010 11:43:40 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <87635df3b9.fsf@spindle.srvr.nix> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> Message-ID: <732076a81003040343w664a8705l730077a6f5b09bad@mail.gmail.com> 2010/3/3 Nix : >  -- N., Tory voter in the last election but you're making me think again Remind me again, is that on par with top posting? ;-) Ben From td at bloogaloo.co.uk Thu Mar 4 12:48:24 2010 From: td at bloogaloo.co.uk (tid) Date: Thu, 4 Mar 2010 12:48:24 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <7120e0ee4bdd18c9f1740108b924693f.squirrel@olga.hinterlands.org> References: <7120e0ee4bdd18c9f1740108b924693f.squirrel@olga.hinterlands.org> Message-ID: <1c5b64b81003040448g6396c98cxf3caf374fae9b748@mail.gmail.com> On 3 March 2010 19:59, Martin A. Brooks wrote: > On Wed, March 3, 2010 4:26 pm, Stephen Nelson-Smith wrote: >> Anything else I should look at?  Any success or war stories to share? > > TrueCrypt. +1 for Truecrypt: a good product. Simple enough for non-technical people, but secure enough for most of my clients. Unless you stick a dymo label with your password on the laptop, as has happened. The rather lame excuse was that "They'd have to know my windows file *and* find the truecypt file!". Sigh. Tid From td at bloogaloo.co.uk Thu Mar 4 12:49:37 2010 From: td at bloogaloo.co.uk (tid) Date: Thu, 4 Mar 2010 12:49:37 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <1c5b64b81003040448g6396c98cxf3caf374fae9b748@mail.gmail.com> References: <7120e0ee4bdd18c9f1740108b924693f.squirrel@olga.hinterlands.org> <1c5b64b81003040448g6396c98cxf3caf374fae9b748@mail.gmail.com> Message-ID: <1c5b64b81003040449y7a9cc7e1m7beaadc94010180b@mail.gmail.com> > rather lame excuse was that "They'd have to know my windows file *and* s/windows file/windows password/ From benjamin at py-soft.co.uk Thu Mar 4 13:01:50 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Thu, 4 Mar 2010 13:01:50 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: References: Message-ID: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> On 3 March 2010 16:26, Stephen Nelson-Smith wrote: > Anything else I should look at?  Any success or war stories to share? I have never had any problems with FileVault but a friend of mine who runs an independent Mac shop has been asked to repair broken FileVault volumes so many times that he refuses to use it and has gone with my recommendation to use TrueCrypt instead. I also understand that FileVault is (a lot) easier to crack than Truecrypt. Ben From walter.stanish at saffrondigital.com Thu Mar 4 13:12:20 2010 From: walter.stanish at saffrondigital.com (Walter Stanish) Date: Thu, 4 Mar 2010 13:12:20 -0000 Subject: [Gllug] Code Generation Conference - CG2010 Message-ID: Is anyone planning to go to the Code Generation conference in Cambridge - http://www.codegeneration.net/cg2010/ ? Has anyone been in previous years? Seems historically quite academically oriented but they're keen to get more industry in this year. Reason for asking is that I proposed a talk a few months back and just had it accepted, but will be based in the states by the time it comes around so am wondering whether to cancel or fly... - Walter -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100304/5421651c/attachment.htm From bap at shrdlu.com Thu Mar 4 15:20:04 2010 From: bap at shrdlu.com (Bernard Peek) Date: Thu, 04 Mar 2010 15:20:04 +0000 Subject: [Gllug] OT: Debt In-Reply-To: <20100304111906.GQ28166@cornerstonelinux.co.uk> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> <20100304111906.GQ28166@cornerstonelinux.co.uk> Message-ID: <4B8FCFA4.8070609@shrdlu.com> On 04/03/10 11:19, John Edwards wrote: > On Thu, Mar 04, 2010 at 10:36:44AM +0000, Peter Childs wrote: > >> >> Does that work with personal debt too? >> > Yes. > > Inflation reduces the value of the currency and hence the value > of the debt. But it erodes both debt and savings, and has strong > feedback effects. > > Surely that's only if the debt value remains static? Personal loans are usually made at interest rates greater than inflation, so the value of a debt increases faster than inflation. Are sovereign debts made at lower rates? -- Bernard Peek bap at shrdlu.com From salsaman at xs4all.nl Thu Mar 4 15:42:26 2010 From: salsaman at xs4all.nl (salsaman at xs4all.nl) Date: Thu, 4 Mar 2010 16:42:26 +0100 Subject: [Gllug] OT: Debt In-Reply-To: <4B8FCFA4.8070609@shrdlu.com> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> <20100304111906.GQ28166@cornerstonelinux.co.uk> <4B8FCFA4.8070609@shrdlu.com> Message-ID: <4c2c4d4790d10e33d8e8c4a24428a5f6.squirrel@webmail.xs4all.nl> On Thu, March 4, 2010 16:20, Bernard Peek wrote: > On 04/03/10 11:19, John Edwards wrote: >> On Thu, Mar 04, 2010 at 10:36:44AM +0000, Peter Childs wrote: >> >>> >>> Does that work with personal debt too? >>> >> Yes. >> >> Inflation reduces the value of the currency and hence the value >> of the debt. But it erodes both debt and savings, and has strong >> feedback effects. >> >> > > Surely that's only if the debt value remains static? Personal loans are > usually made at interest rates greater than inflation, so the value of a > debt increases faster than inflation. Are sovereign debts made at lower > rates? > That is true. Also the value of debt only decreases relatively if wages increase in line with inflation. This worked in the '70s, but is unlikely to work now as the UK is competing with wage levels in the likes of India and China. Salsaman. http://lives.sourceforge.net From john at cornerstonelinux.co.uk Thu Mar 4 16:06:57 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Thu, 4 Mar 2010 16:06:57 +0000 Subject: [Gllug] OT: Debt In-Reply-To: <4B8FCFA4.8070609@shrdlu.com> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> <20100304111906.GQ28166@cornerstonelinux.co.uk> <4B8FCFA4.8070609@shrdlu.com> Message-ID: <20100304160657.GS28166@cornerstonelinux.co.uk> On Thu, Mar 04, 2010 at 03:20:04PM +0000, Bernard Peek wrote: > On 04/03/10 11:19, John Edwards wrote: >> On Thu, Mar 04, 2010 at 10:36:44AM +0000, Peter Childs wrote: >> >>> Does that work with personal debt too? >> >> Yes. >> >> Inflation reduces the value of the currency and hence the value >> of the debt. But it erodes both debt and savings, and has strong >> feedback effects. > > Surely that's only if the debt value remains static? Personal loans are > usually made at interest rates greater than inflation, so the value of a > debt increases faster than inflation. During normal times, yes. But when inflation starts to spiral out of control then lenders can not renegotiate the loans quickly enough to make up the difference. > Are sovereign debts made at lower rates? Basic Gilts pay a fixed amount at fixed dates over a fixed term (from 6 months to 50 years) regardless of inflation or interest rates, so they are very susceptible to inflation. There are also variable rate bonds linked to inflation or inflation rates, but these are less common. The Bank of England is both a borrower and lender. In the case of "quantitative easing" it gives bonds to banks and then buys them back, producing a short term increase in asset flow. This is effectively "printing money" electronically. Of course the real value of the economy is not increased by this, and so it can lead to currency devaluation and inflation. -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100304/45e0ea7a/attachment.pgp From walter.stanish at saffrondigital.com Thu Mar 4 16:19:36 2010 From: walter.stanish at saffrondigital.com (Walter Stanish) Date: Thu, 4 Mar 2010 16:19:36 -0000 Subject: [Gllug] OT: Debt In-Reply-To: <20100304160657.GS28166@cornerstonelinux.co.uk> References: <20100302215244.GA12176@annexia.org><4B8DA380.7010808@GetAroundToIt.co.uk><1267603813.10915.73.camel@localhost><87635df3b9.fsf@spindle.srvr.nix><1267662462.1783.114.camel@localhost><87wrxsedkx.fsf@spindle.srvr.nix><20100304111906.GQ28166@cornerstonelinux.co.uk><4B8FCFA4.8070609@shrdlu.com> <20100304160657.GS28166@cornerstonelinux.co.uk> Message-ID: > Of course the real value of the economy is not increased > by this, and so it can lead to currency devaluation and > inflation. A very interesting character I have been in touch with writes a lot about the nature of modern financial systems and what kinds of alternatives exist. People on this list may be interested... http://www.reinventingmoney.com/ "The mission of this site is to demystify money by presenting the best leading-edge ideas on monetary and non-monetary exchange. It is a resource devoted to the advancement of economic democracy, self-determination, and global harmony." http://beyondmoney.net/ Money related blog. - Walter From john at cornerstonelinux.co.uk Thu Mar 4 16:30:39 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Thu, 4 Mar 2010 16:30:39 +0000 Subject: [Gllug] OT: Debt In-Reply-To: References: <20100304160657.GS28166@cornerstonelinux.co.uk> Message-ID: <20100304163039.GT28166@cornerstonelinux.co.uk> On Thu, Mar 04, 2010 at 04:19:36PM -0000, Walter Stanish wrote: >> Of course the real value of the economy is not increased >> by this, and so it can lead to currency devaluation and >> inflation. > > A very interesting character I have been in touch with writes a > lot about the nature of modern financial systems and what kinds > of alternatives exist. > > People on this list may be interested... > > http://www.reinventingmoney.com/ > "The mission of this site is to demystify money by presenting > the best leading-edge ideas on monetary and non-monetary > exchange. It is a resource devoted to the advancement of > economic democracy, self-determination, and global harmony." They can start advancing "global harmony" by not using fixed pixel height/width/position. The text is an unreadable overlapping mess. -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100304/900acd06/attachment.pgp From dennis at basis.uklinux.net Thu Mar 4 18:05:18 2010 From: dennis at basis.uklinux.net (Dennis Furey) Date: Thu, 4 Mar 2010 18:05:18 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> Message-ID: <20100304180517.GA4136@basis.uklinux.net> Any reason no one has mentioned LUKS? It's well supported, based on AES encryption, and I've used it for years without any issues. Dennis From john at cornerstonelinux.co.uk Thu Mar 4 18:07:07 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Thu, 4 Mar 2010 18:07:07 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100304180517.GA4136@basis.uklinux.net> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> Message-ID: <20100304180707.GU28166@cornerstonelinux.co.uk> On Thu, Mar 04, 2010 at 06:05:18PM +0000, Dennis Furey wrote: > Any reason no one has mentioned LUKS? It's well supported, based on > AES encryption, and I've used it for years without any issues. Does it work on MacOS? -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100304/5260d7ed/attachment.pgp From benjamin at py-soft.co.uk Thu Mar 4 18:07:43 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Thu, 4 Mar 2010 18:07:43 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100304180517.GA4136@basis.uklinux.net> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> Message-ID: <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> On 4 March 2010 18:05, Dennis Furey wrote: > Any reason no one has mentioned LUKS? It's well supported, based on > AES encryption, and I've used it for years without any issues. I'm happy to be corrected, but I don't believe LUKS is available for MacOS. Ben From dennis at basis.uklinux.net Thu Mar 4 18:27:56 2010 From: dennis at basis.uklinux.net (Dennis Furey) Date: Thu, 4 Mar 2010 18:27:56 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> Message-ID: <20100304182756.GA4253@basis.uklinux.net> On Thu, Mar 04, 2010 at 06:07:43PM +0000, Benjamin Donnachie wrote: > On 4 March 2010 18:05, Dennis Furey wrote: > > Any reason no one has mentioned LUKS? It's well supported, based on > > AES encryption, and I've used it for years without any issues. > > I'm happy to be corrected, but I don't believe LUKS is available for MacOS. > True, it's only for Linux. I didn't notice MacOS in the original post. Back to lurking. Dennis From benjamin at py-soft.co.uk Thu Mar 4 18:36:06 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Thu, 4 Mar 2010 18:36:06 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100304182756.GA4253@basis.uklinux.net> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> Message-ID: <2678634048094966190@unknownmsgid> On 4 Mar 2010, at 18:27, Dennis Furey wrote: > True, it's only for Linux. I didn't notice MacOS in the original > post. Reasonable assumption given this is a Linux list! Wish full disk TrueCrypt encryption was available for the Mac. May try to hack it one day if I ever have sufficient time and motivation! Ben From john at cornerstonelinux.co.uk Thu Mar 4 18:36:09 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Thu, 4 Mar 2010 18:36:09 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100304182756.GA4253@basis.uklinux.net> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> Message-ID: <20100304183609.GV28166@cornerstonelinux.co.uk> On Thu, Mar 04, 2010 at 06:27:56PM +0000, Dennis Furey wrote: > On Thu, Mar 04, 2010 at 06:07:43PM +0000, Benjamin Donnachie wrote: >> On 4 March 2010 18:05, Dennis Furey wrote: >>> Any reason no one has mentioned LUKS? It's well supported, based on >>> AES encryption, and I've used it for years without any issues. >> >> I'm happy to be corrected, but I don't believe LUKS is available for MacOS. > > True, it's only for Linux. I didn't notice MacOS in the original > post. Back to lurking. No worries. How do you use LUKS? Certain files, whole filesystem, root filesystem? -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100304/cbfb7c3c/attachment.pgp From nix at esperi.org.uk Thu Mar 4 23:44:32 2010 From: nix at esperi.org.uk (Nix) Date: Thu, 04 Mar 2010 23:44:32 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <732076a81003040343w664a8705l730077a6f5b09bad@mail.gmail.com> (Benjamin Donnachie's message of "Thu, 4 Mar 2010 11:43:40 +0000") References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <732076a81003040343w664a8705l730077a6f5b09bad@mail.gmail.com> Message-ID: <87sk8fejof.fsf@spindle.srvr.nix> On 4 Mar 2010, Benjamin Donnachie uttered the following: > 2010/3/3 Nix : >>  -- N., Tory voter in the last election but you're making me think again > > Remind me again, is that on par with top posting? Being a Tory voter? *Definitely* worse. ;P From GLLUG at GetAroundToIt.co.uk Fri Mar 5 00:48:04 2010 From: GLLUG at GetAroundToIt.co.uk (David L Neil) Date: Fri, 05 Mar 2010 13:48:04 +1300 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> Message-ID: <4B9054C4.70103@GetAroundToIt.co.uk> Hi, On 03/03/10 15:45, JLMS wrote: > I have a Kiwi colleague, working in London, he could not find a job > down under and had to come back to the UK for work, he said the > situation down under for IT professionals is dire (or do you think > they are immune to outsourcing to India and other economic pressures? > They even are in a similar time zone as the Philippines, Singapore and > other "next generation" outsourcing hotspots which makes things > particularly challenging). This friend is no dumber, has worked in > top companies in the UK for several years. > > As for economists predicting how the UK (or Aussie or Kiwi) economy > will do in the next 10 years, phwa!, like if they really knew what > they are talking about (have we forgotten how little foresight about > their own field of "expertise" these fellows have in general terms?) I think the last is the first point I would like to tackle. What may not be apparent, is that the financial crisis that has been the center of attention in the UK and US, has had an effect on, but was not replicated in other parts of the world. This is why it is worth paying attention to the Asian economies, because they have been growing this entire time (the last one~two years) whilst those parts of America that were not previously owned by the Arabs are now held as financial security by the Chinese... (you know the picture in the UK better than I, as my absence increases my out-of-datedness) New Zealand, specifically, has always run with conservative mortgage regulations, so the idea of a 100% (or more) mortgage, is almost unknown and the risk of negative equity (a la 'Maggie' and now 'Gordo') a rare consequence. Thus personal finances are (perhaps) in better shape (personal savings rates are still nothing to boast about). Economic slow-downs and suchlike have had a major impact on jobs, but to my observation, much less than in the UK. Australia had a much better idea to keep the money flowing (per money multiplier effect, and 'consumer confidence'): their government simply handed out chunks of cash to all?various members of the general population (out of which New Zealand also did rather well, when young Aussies decided to visit Kiwi-land on vacation or for better skiing). Compare this with the UK and US governments giving wads of cash to the merchant bankers, who instead of passing it on (even to their oft-sacrificed retail brethren), simply held on to it and used it as fuel for the bonuses they are paying themselves today. I agree with another correspondent here, that the business of offshore and even out-sourcing generally is being re-examined. The general problems of contract definition and contractor control (refer back to Rich's original msg and podcast) when one shifts work 'outside' are hugely magnified by differences in language, culture, value systems, education, work ethic, modes of doing business, ... (How many people do you know who grind their teeth/have given up dealing with Dell, simply because every call requires one to endure a phone conversation with some script-jockey in Bangalore?) This impact has been largely under-estimated and glossed-over even more than the necessary effort of explaining to folk outside the organisation how the insides work... Aside: over here, the major cell phone provider 'lost' half the country last week. They have recently changed to a new 'it works perfectly everywhere else' GSM set-up (American style - no comment necessary from those who know that the very words "cell phone" are transliterated from the original scandiwegian. Perhaps Chris Carter (the UK MP) shouldn't have taken a role at Alcatel/Lucent?) This fiasco was sadly, not the first. The Consumers' Association weighed-in on national TV news with the (?legal) opinion that if one has suffered three or four losses of service in recent times (and many have) such is a valid means of claiming breach of contract/failure to supply, and thus trumping the telco's contract term (as in, duration: 12, 18, 24 months) to claim a (rapid) escape clause. I can't imagine that happening in the ?consumer-friendly? UK environment. Interestingly the company had to fly staff from Auckland (where the system was working) south, because so many customers 'invaded' retail stores having refused to call the 'help line' because it lands in Manila (Philippines) - a message that has not passed unnoticed! (so did its competitors who needed extra staff to assist clients to transfer-in to their service!) Back to the topic: Your colleague may have been correct. Certain skills seem to 'get the chop' ahead of others, and contractors before in-house staff when it is imagined that the work can be 'brought back'. (etc) Therefore he may well have done the right thing - for him - and when my family considerations are settled (don't ask me how bad a plumber I have been! I trust I'll do better as a realtor...) I may yet do the same! The other point is that the situation here is v.fluid - whereas the news that I have from London is that change/improvement is not so noticeable (mind you, some didn't noticed anything 'wrong' - so back to the previous paragraph). Thus the situation a few months ago, and the state-of-play today are quite different - I can see that just from talking to people and keeping a speculative eye on the adverts. Another point is that as the economy picks-up, things will change for IT staff rather rapidly, and quite literally from one week of nothing to the next week of plenty. Such is happening here, and in (at least parts of) Australia. Thus I suspect (and may well test more deliberately) that Project Managers and others who are 'in' at the start of projects, are becoming sought-after, indeed are in rather short supply. Certainly with folk coming back to work after their summer holidays, schools and unis re-commencing, the number of ad pages has certainly increased - and presumably the online services accordingly. Coders and programmers may find that their turn comes a little further down the track... My fear is that there is more of an age-related bias over here (spoken like a true grey-beard and fully-fledged curmudgeon), in comparison to Europe. For relevant roles there is the same issue that people have experience with the given software, application, language, whatever... rather than taking on fresh graduates or training-up others. However there is also (generally) a more adventurous/entrepreneurial attitude when it comes to stepping into the future. Unfortunately whilst there are solid moves being made to utilise open source, eg government/civil service policies, and I have spoken to people involved in projects using FOSS in some form, I can't say that I have seem more than one advert that specifically mentioned Linux - to be fair, I probably don't do more than glance at technically-oriented adverts, (eg those typically seen on GLLUG) so will not pretend to have sensible comment. Contrarily I have found the two FOSS/LUG email lists to which I have subscribed, rather technical in content (with more of the 'how do I firewall/wireless/netbook...' type topics than seen on GLLUG, both as percentage and in absolute terms), yet at the same time there have been plenty of comments about ACTA (and success in persuading the New Zealand government that the 'rule of law' (presumption of innocence) comes before the avaricious interests of Hollywood and record studios, et al) and injecting FOSS into schools either by replicating a brand new school that is running 100% 'free' and/or by working on a package to fulfill the needs of schools in reporting to the Department of Education... I haven't seen many conversations about HPC, clusters, or server farms; so another (?crass) generalisation is that I see technical discussions/individuals dominating the list, rather than perhaps those stalwarts of GLLUG who moved from there to become larger scale systems/network administrators... OTOH I know that the university-types have their own and similar lists (the air there is definitely too rarefied for me!) and thus research projects and their needs are likely discussed there. There are fewer ISPs and web hosting operations, which would also shape the employment (and list) landscape. etc. If I were off-shoring from the UK or N.America, I would consider New Zealand over the others mentioned. The language and sociological arguments apply, as above. Many kiwis have done their "OE" (overseas experience) in the UK and returned, and thus have a v.good understanding of how things are (expected to be) done. I mentioned that last week I was working for a client in Durham. She thought it was great that she do 'real work' during her day, and 'only' have to worry about web-dev first and last thing - we were both putting in some long hours to maximise our real-time,though. What proved to be the clincher was that she was thrilled by the idea that we could agree a list of tasks one evening, and the dev site would be updated 'overnight' and ready for review while she munched her breakfast - and nary a mention of "agile" or "continuous integration" - nor "scrum" or "sprint", unless the All Blacks are playing the Lions or Wales... Meantime back at the coal face (or was that Newcastle?) I would work on the site during our day, and then sit outside enjoying the sunshine until our next 'window'. I could get used to this! I think the India, Singapore... opportunity is for coding tasks and self-contained projects* where things can be carefully and 'completely' specified in-advance. In other words one is pretty much stuck with the water-fall model of development. Rapid/Agile/XP is becoming 'available' but so much of the (limited) success that I've seen in such practices requires teams who can reach out and (physically) grab each other, ie co-operate in real time. So on that note, I'll shut-up. *hmm hadn't given much thought to the idea of combining this with SoA module development... Lastly (and your parting comment), please don't fall into the 'trap' or take the bankers' bait! During the 'financial crisis' the financial community has been very coyly hoping that people will continue to use the term 'banker'. The 'great British public' sees a 'banker' as someone working at the local branch of Barclays, HSBC, Lloyds... Wrong! The villains of the piece are investment or merchant bankers. They are the ones who have built a house of cards (on (tissue) paper), taken their fees up-front, and dragged entire economies down - and thoroughly deserving of your scorn. Similarly economists operate at a different level from merchant bankers, and operate from data made available. Many have sounded alarms about the way things have been going, eg levels of credit (both individual/corporate and inter-national). However the previous Republican US Administration listened to no-one except their own 'pets' who eschewed traditional economics/economic wisdom in favor of continuing to ignore today what could be loaded for payment tomorrow... Your average journo isn't going to penetrate either of those thickets, so the labels will still be used fatuously (and erroneously) and the average politician is seduced by associating with fame, power, and/or money, so... With apologies to those who might feel we're way-away from LUG topics, but trust this is helpful (to someone), =dn From dennis at basis.uklinux.net Fri Mar 5 04:59:46 2010 From: dennis at basis.uklinux.net (Dennis Furey) Date: Fri, 5 Mar 2010 04:59:46 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100304183609.GV28166@cornerstonelinux.co.uk> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> <20100304183609.GV28166@cornerstonelinux.co.uk> Message-ID: <20100305045945.GA6422@basis.uklinux.net> On Thu, Mar 04, 2010 at 06:36:09PM +0000, John Edwards wrote: > > How do you use LUKS? > > Certain files, whole filesystem, root filesystem? > It's full disk encryption, and can be used for the root filesystem. I have encrypted roots on a half dozen boxes and encrypted filesystems on some external usb drives. I also store a backup of my home directory in a loopback mounted LUKS filesystem image. Once in a while I unmount that and write the backing file to a DVD or split it into small fixed sized pieces that I sync with a bucket on Amazon S3. (The backing file is unintelligible without the key.) For a non-root filesystem, you have to set up a physical device (like /dev/sdb1) using the cryptsetup command first with a made up name and passphrase, and then mount it as /dev/mapper/madeupname instead of mounting /dev/sdb1. You'll also have to format it before mounting it the first time, as ext3 or reiserfs or whatever filesystem you want. For an encrypted root, you need a modified initramfs to prompt for the passphrase and set it up automatically when booting. The Debian cryptsetup package is helpful but migrating an existing system to an encrypted root isn't fully automated. There are some good web tutorials hanging around. There's also a way to get an encrypted swap partition with a random password generated on each boot, but I prefer having enough RAM to do without a swap. From gllug at getaroundtoit.co.uk Fri Mar 5 07:33:15 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Fri, 05 Mar 2010 20:33:15 +1300 Subject: [Gllug] Small PC, resource constraints, improving efficiencies Message-ID: <4B90B3BB.5070405@getaroundtoit.co.uk> Should I trial LXDE and/or XFCE by attempting to run multiple Desktop Environments/Window Managers at the X level, or by dual-booting whole distributions with the appropriate choice as the sole GUI? While I'm traveling around the world I'm using a very old, slow, and small Thinkpad, mainly for Thunderbird and Firefox* (netbook mode) but periodically using OpenOffice Writer or Calc, reading PDFs, VNC and SSH to servers, wired and wireless connections, etc. It took me quite a bit of time to iron-out the kinks (partly my own ignorance/learning process) but it now runs a trimmed-back, basic CentOS 5.2 - I tried and failed to persuade the likes of Ubuntu and Fedora to fire-up as a live system or to load to the hdd. I suspect an issue in X/Gnome/APCI/mouse (red nipple, stick) lies at the root of things, but also the several Windows-only h/w components. Likely it is the conservative (ok, old) nature of CentOS, that facilitates its success... (please stop laughing at the combination - I 'know' CentOS because I've been using it, more appropriately, on numbers of machines, elsewhere) *have thought to try Opera 10 (and maybe its Dragonfly system for web-dev) and compare load factors with FF...? XFCE seems to be available from the CentOS?rpmforge repositories but LXDE is not found by search. I have downloaded Vector Linux's Light Live edition (and it runs). More recently I have noted that the recent Knoppix 6.2 release offers an LXDE Lightweight option, similarly. I'd like to trial and compare the two or three choices until I've been able to gain confidence in whichever choice (including my existing CentOS set-up) runs most efficiently... Would I be better-off learning how to multi-boot a couple of Linux systems and building two systems side-by-side? (the online tutorials and the books I've read detail multi-booting Linux and Windows, but not multiple Linux-es) - and thus, to use GRUB or advantageous to upgrade to and learn GRUB2? Alternatively, should I simply load the other DktopEnv/WinMgrs on top of the existing CentOS X11 (and hopefully rewire things with the Desktop switcher)? - but I'm concerned that removal after failure might be messy/risk the current system-config? NB because I'm traveling and have limited access to the web, I don't have many resources to teach myself/learn how to do all this. My web servers (etc) all run without GUIs, so I'm not scared of the command line, but I'm no SysAdmin! Any and all advice, and web refs will be welcomed, =dn From general.mooney at googlemail.com Fri Mar 5 09:08:13 2010 From: general.mooney at googlemail.com (=?UTF-8?Q?Ciar=C3=A1n_Mooney?=) Date: Fri, 5 Mar 2010 09:08:13 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder Message-ID: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> Hi, Friendly reminder of this months meeting, details about the talk can be found here: http://lists.gllug.org.uk/pipermail/gllug/2010-February/082543.html Seán Tohill has asked that I keep track of the number of attendees. Currently 15 people have said that they will be attending. Please email the list if you plan to attend. Regards, Ciarán Attending ================== Chris Bell [maybe] Tom Westcott Dan + guest Matthew King Benjamin Donnachie [maybe] Frank Scott Justin Perreault LesleyB Andrew Back Andy Smith Ciarán Mooney Sean Tohil Nahuel Marisi Phil Reynolds Troy Jendra From sta296 at astradyne.co.uk Fri Mar 5 09:22:59 2010 From: sta296 at astradyne.co.uk (Tethys) Date: Fri, 05 Mar 2010 09:22:59 +0000 Subject: [Gllug] Small PC, resource constraints, improving efficiencies In-Reply-To: Your message of "Fri, 05 Mar 2010 20:33:15 +1300." <4B90B3BB.5070405@getaroundtoit.co.uk> References: <4B90B3BB.5070405@getaroundtoit.co.uk> Message-ID: <201003050922.o259MxVS013877@leto.astradyne.corp> -------- David L Neil writes: >Should I trial LXDE and/or XFCE by attempting to run multiple Desktop >Environments/Window Managers at the X level, or by dual-booting whole >distributions with the appropriate choice as the sole GUI? I don't really think it makes much difference, although multiple desktop environments is a less invasive solution. >*have thought to try Opera 10 (and maybe its Dragonfly system for >web-dev) and compare load factors with FF...? I'd definitely recommend Opera. I used to hate it with a passion, but recent versions have been very, very good, and having it start up in a second is a wonderfully refreshing change from Firefox's bloat. >XFCE seems to be available from the CentOS?rpmforge repositories but >LXDE is not found by search. How much do you want from a desktop environment? If you're looking for a lightweight solution, then you might want to look at FVWM. It's in EPEL, so installing it on CentOS is a breeze. It's configured through a config file, rather than dragging and dropping things onto a task bar, for example, but it's insanely powerful, and light on resources. I've tried KDE and GNOME, but found them to be limiting, and have always returned to FVWM. It does what I want with a minimum of fuss. >Alternatively, should I simply load the other DktopEnv/WinMgrs on top of >the existing CentOS X11 (and hopefully rewire things with the Desktop >switcher)? >- but I'm concerned that removal after failure might be messy/risk the >current system-config? I think that's an ungrounded fear. I habitually try things out on a regular basis, and when I decide I'm done, a quick "yum erase" gets the box back to its previous state without any problems. Tet From sta296 at astradyne.co.uk Fri Mar 5 09:26:22 2010 From: sta296 at astradyne.co.uk (Tethys) Date: Fri, 05 Mar 2010 09:26:22 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: Your message of "Fri, 05 Mar 2010 09:08:13 GMT." <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> Message-ID: <201003050926.o259QMa8013922@leto.astradyne.corp> -------- >Seán Tohill has asked that I keep track of the number of attendees. >Currently 15 people have said that they will be attending. Please >email the list if you plan to attend. The best I can give is a maybe. In future, if you need to know numbers, you might want to set up a different mechanism for that, rather than having people spam the list with AOL style "me too" replies. Tet From general.mooney at googlemail.com Fri Mar 5 09:33:51 2010 From: general.mooney at googlemail.com (=?UTF-8?Q?Ciar=C3=A1n_Mooney?=) Date: Fri, 5 Mar 2010 09:33:51 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: <201003050926.o259QMa8013922@leto.astradyne.corp> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> <201003050926.o259QMa8013922@leto.astradyne.corp> Message-ID: <3e4e51a81003050133u4e3def69t4caae054162565df@mail.gmail.com> Hi, > The best I can give is a maybe. Thanks, > In future, if you need to know numbers, you might want to set up > a different mechanism for that, rather than having people spam the > list with AOL style "me too" replies. Yup. Just kinda got stuck this way. It comes from my previous experience of mailing lists and other LUGs and they seem happy with the method. Obviously a slightly different culture at GLLUG, but that's ok. I'd certainly prefer an EventWax page or something, but didn't have a capacity for the venue when I started. Now I do have a capacity and would not have felt comfortable asking people to "re-register". Ciarán From mail-lists at karan.org Fri Mar 5 10:00:51 2010 From: mail-lists at karan.org (Karanbir Singh) Date: Fri, 05 Mar 2010 10:00:51 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: <3e4e51a81003050133u4e3def69t4caae054162565df@mail.gmail.com> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> <201003050926.o259QMa8013922@leto.astradyne.corp> <3e4e51a81003050133u4e3def69t4caae054162565df@mail.gmail.com> Message-ID: <4B90D653.1090109@karan.org> On 03/05/2010 09:33 AM, Ciarán Mooney wrote: > the method. Obviously a slightly different culture at GLLUG, but > that's ok. I'd certainly prefer an EventWax page or something, but Have you looked at 'www.doodle.com' ? its an easy way to get the me-too type schedule going ( and no registration required by responders) - KB From john at technolalia.org Fri Mar 5 10:43:44 2010 From: john at technolalia.org (John Levin) Date: Fri, 05 Mar 2010 10:43:44 +0000 Subject: [Gllug] Open source scheduling apps (was: GLLUG Talk) In-Reply-To: <4B90D653.1090109@karan.org> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> <201003050926.o259QMa8013922@leto.astradyne.corp> <3e4e51a81003050133u4e3def69t4caae054162565df@mail.gmail.com> <4B90D653.1090109@karan.org> Message-ID: <4B90E060.3080206@technolalia.org> On 05/03/2010 10:00, Karanbir Singh wrote: > On 03/05/2010 09:33 AM, Ciarán Mooney wrote: >> the method. Obviously a slightly different culture at GLLUG, but >> that's ok. I'd certainly prefer an EventWax page or something, but > > Have you looked at 'www.doodle.com' ? its an easy way to get the me-too > type schedule going ( and no registration required by responders) > > - KB Does anyone know if there's a free, open source scheduling application comparable to doodle and suchlike? Thanks, John -- John Levin http://www.anterotesis.com http://www.facebook.com/john.levin http://twitter.com/anterotesis From DomainAdmin at GetAroundToIt.co.uk Thu Mar 4 21:16:23 2010 From: DomainAdmin at GetAroundToIt.co.uk (David L Neil) Date: Fri, 05 Mar 2010 10:16:23 +1300 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> Message-ID: <4B902327.70707@GetAroundToIt.co.uk> Hi, On 03/03/10 15:45, JLMS wrote: > I have a Kiwi colleague, working in London, he could not find a job > down under and had to come back to the UK for work, he said the > situation down under for IT professionals is dire (or do you think > they are immune to outsourcing to India and other economic pressures? > They even are in a similar time zone as the Philippines, Singapore and > other "next generation" outsourcing hotspots which makes things > particularly challenging). This friend is no dumber, has worked in > top companies in the UK for several years. > > As for economists predicting how the UK (or Aussie or Kiwi) economy > will do in the next 10 years, phwa!, like if they really knew what > they are talking about (have we forgotten how little foresight about > their own field of "expertise" these fellows have in general terms?) I think the last is the first point I would like to tackle. What may not be apparent, is that the financial crisis that has been the center of attention in the UK and US, has had an effect on, but was not replicated in other parts of the world. This is why it is worth paying attention to the Asian economies, because they have been growing this entire time (the last one~two years) whilst those parts of America that were not previously owned by the Arabs are now held as financial security by the Chinese... (you know the picture in the UK better than I, as my absence increases my out-of-datedness) New Zealand, specifically, has always run with conservative mortgage regulations, so the idea of a 100% (or more) mortgage, is almost unknown and the risk of negative equity (a la 'Maggie' and now 'Gordo') a rare consequence. Thus personal finances are (perhaps) in better shape (personal savings rates are still nothing to boast about). Economic slow-downs and suchlike have had a major impact on jobs, but to my observation, much less than in the UK. Australia had a much better idea to keep the money flowing (per money multiplier effect, and 'consumer confidence'): their government simply handed out chunks of cash to all?various members of the general population (out of which New Zealand also did rather well, when young Aussies decided to visit Kiwi-land on vacation or for better skiing). Compare this with the UK and US governments giving wads of cash to the merchant bankers, who instead of passing it on (even to their oft-sacrificed retail brethren), simply held on to it and used it as fuel for the bonuses they are paying themselves today. I agree with another correspondent here, that the business of offshore and even out-sourcing generally is being re-examined. The general problems of contract definition and contractor control (refer back to Rich's original msg and podcast) when one shifts work 'outside' are hugely magnified by differences in language, culture, value systems, education, work ethic, modes of doing business, ... (How many people do you know who grind their teeth/have given up dealing with Dell, simply because every call requires one to endure a phone conversation with some script-jockey in Bangalore?) This impact has been largely under-estimated and glossed-over even more than the necessary effort of explaining to folk outside the organisation how the insides work... Aside: over here, the major cell phone provider 'lost' half the country last week. They have recently changed to a new 'it works perfectly everywhere else' GSM set-up (American style - no comment necessary from those who know that the very words "cell phone" are transliterated from the original scandiwegian. Perhaps Chris Carter (the UK MP) shouldn't have taken a role at Alcatel/Lucent?) This fiasco was sadly, not the first. The Consumers' Association weighed-in on national TV news with the (?legal) opinion that if one has suffered three or four losses of service in recent times (and many have) such is a valid means of claiming breach of contract/failure to supply, and thus trumping the telco's contract term (as in, duration: 12, 18, 24 months) to claim a (rapid) escape clause. I can't imagine that happening in the ?consumer-friendly? UK environment. Interestingly the company had to fly staff from Auckland (where the system was working) south, because so many customers 'invaded' retail stores having refused to call the 'help line' because it lands in Manila (Philippines) - a message that has not passed unnoticed! (so did its competitors who needed extra staff to assist clients to transfer-in to their service!) Back to the topic: Your colleague may have been correct. Certain skills seem to 'get the chop' ahead of others, and contractors before in-house staff when it is imagined that the work can be 'brought back'. (etc) Therefore he may well have done the right thing - for him - and when my family considerations are settled (don't ask me how bad a plumber I have been! I trust I'll do better as a realtor...) I may yet do the same! The other point is that the situation here is v.fluid - whereas the news that I have from London is that change/improvement is not so noticeable (mind you, some didn't noticed anything 'wrong' - so back to the previous paragraph). Thus the situation a few months ago, and the state-of-play today are quite different - I can see that just from talking to people and keeping a speculative eye on the adverts. Another point is that as the economy picks-up, things will change for IT staff rather rapidly, and quite literally from one week of nothing to the next week of plenty. Such is happening here, and in (at least parts of) Australia. Thus I suspect (and may well test more deliberately) that Project Managers and others who are 'in' at the start of projects, are becoming sought-after, indeed are in rather short supply. Certainly with folk coming back to work after their summer holidays, schools and unis re-commencing, the number of ad pages has certainly increased - and presumably the online services accordingly. Coders and programmers may find that their turn comes a little further down the track... My fear is that there is more of an age-related bias over here (spoken like a true grey-beard and fully-fledged curmudgeon), in comparison to Europe. For relevant roles there is the same issue that people have experience with the given software, application, language, whatever... rather than taking on fresh graduates or training-up others. However there is also (generally) a more adventurous/entrepreneurial attitude when it comes to stepping into the future. Unfortunately whilst there are solid moves being made to utilise open source, eg government/civil service policies, and I have spoken to people involved in projects using FOSS in some form, I can't say that I have seem more than one advert that specifically mentioned Linux - to be fair, I probably don't do more than glance at technically-oriented adverts, (eg those typically seen on GLLUG) so will not pretend to have sensible comment. Contrarily I have found the two FOSS/LUG email lists to which I have subscribed, rather technical in content (with more of the 'how do I firewall/wireless/netbook...' type topics than seen on GLLUG, both as percentage and in absolute terms), yet at the same time there have been plenty of comments about ACTA (and success in persuading the New Zealand government that the 'rule of law' (presumption of innocence) comes before the avaricious interests of Hollywood and record studios, et al) and injecting FOSS into schools either by replicating a brand new school that is running 100% 'free' and/or by working on a package to fulfill the needs of schools in reporting to the Department of Education... I haven't seen many conversations about HPC, clusters, or server farms; so another (?crass) generalisation is that I see technical discussions/individuals dominating the list, rather than perhaps those stalwarts of GLLUG who moved from there to become larger scale systems/network administrators... OTOH I know that the university-types have their own and similar lists (the air there is definitely too rarefied for me!) and thus research projects and their needs are likely discussed there. There are fewer ISPs and web hosting operations, which would also shape the employment (and list) landscape. etc. If I were off-shoring from the UK or N.America, I would consider New Zealand over the others mentioned. The language and sociological arguments apply, as above. Many kiwis have done their "OE" (overseas experience) in the UK and returned, and thus have a v.good understanding of how things are (expected to be) done. I mentioned that last week I was working for a client in Durham. She thought it was great that she do 'real work' during her day, and 'only' have to worry about web-dev first and last thing - we were both putting in some long hours to maximise our real-time,though. What proved to be the clincher was that she was thrilled by the idea that we could agree a list of tasks one evening, and the dev site would be updated 'overnight' and ready for review while she munched her breakfast - and nary a mention of "agile" or "continuous integration" - nor "scrum" or "sprint", unless the All Blacks are playing the Lions or Wales... Meantime back at the coal face (or was that Newcastle?) I would work on the site during our day, and then sit outside enjoying the sunshine until our next 'window'. I could get used to this! I think the India, Singapore... opportunity is for coding tasks and self-contained projects* where things can be carefully and 'completely' specified in-advance. In other words one is pretty much stuck with the water-fall model of development. Rapid/Agile/XP is becoming 'available' but so much of the (limited) success that I've seen in such practices requires teams who can reach out and (physically) grab each other, ie co-operate in real time. So on that note, I'll shut-up. *hmm hadn't given much thought to the idea of combining this with SoA module development... Lastly (and your parting comment), please don't fall into the 'trap' or take the bankers' bait! During the 'financial crisis' the financial community has been very coyly hoping that people will continue to use the term 'banker'. The 'great British public' sees a 'banker' as someone working at the local branch of Barclays, HSBC, Lloyds... Wrong! The villains of the piece are investment or merchant bankers. They are the ones who have built a house of cards (on (tissue) paper), taken their fees up-front, and dragged entire economies down - and thoroughly deserving of your scorn. Similarly economists operate at a different level from merchant bankers, and operate from data made available. Many have sounded alarms about the way things have been going, eg levels of credit (both individual/corporate and inter-national). However the previous Republican US Administration listened to no-one except their own 'pets' who eschewed traditional economics/economic wisdom in favor of continuing to ignore today what could be loaded for payment tomorrow... Your average journo isn't going to penetrate either of those thickets, so the labels will still be used fatuously (and erroneously) and the average politician is seduced by associating with fame, power, and/or money, so... With apologies to those who might feel we're way-away from LUG topics, but trust this is helpful (to someone), =dn From S.Tohill at westminster.ac.uk Fri Mar 5 11:42:09 2010 From: S.Tohill at westminster.ac.uk (sean) Date: Fri, 05 Mar 2010 11:42:09 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> Message-ID: <4B90EE11.6010803@wmin.ac.uk> Ciarán Mooney wrote: > Hi, > > Friendly reminder of this months meeting, details about the talk can > be found here: > > http://lists.gllug.org.uk/pipermail/gllug/2010-February/082543.html > > Seán Tohill has asked that I keep track of the number of attendees. > Currently 15 people have said that they will be attending. Please > email the list if you plan to attend. > > > basically the reception desk want to know how many will be attending. i have used sign in sheets at reception on the day of the talk, but some gllug attendees feel it is some sort big brother process. i will revert to using these again, it is easier, so no need to reply with a 'me too'. regards sean -- The University of Westminster is a charity and a company limited by guarantee. Registration number: 977818 England. Registered Office: 309 Regent Street, London W1B 2UW, UK. From andy at andymillar.co.uk Fri Mar 5 11:53:28 2010 From: andy at andymillar.co.uk (Andy Millar) Date: Fri, 5 Mar 2010 11:53:28 -0000 Subject: [Gllug] Open source scheduling apps (was: GLLUG Talk) In-Reply-To: <4B90E060.3080206@technolalia.org> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> <201003050926.o259QMa8013922@leto.astradyne.corp> <3e4e51a81003050133u4e3def69t4caae054162565df@mail.gmail.com> <4B90D653.1090109@karan.org> <4B90E060.3080206@technolalia.org> Message-ID: <1267790008.26839.0.camel@millaralpt.co.tradefair> On Fri, 2010-03-05 at 10:43 +0000, John Levin wrote: > Does anyone know if there's a free, open source scheduling application > comparable to doodle and suchlike? Would a wiki be sufficient? Andy -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 5162 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100305/ad24a102/attachment.bin From martin at martindengler.com Fri Mar 5 12:44:41 2010 From: martin at martindengler.com (Martin Dengler) Date: Fri, 5 Mar 2010 12:44:41 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> Message-ID: <20100305124440.GC12730@ops-13.xades.com> On Fri, Mar 05, 2010 at 09:08:13AM +0000, Ciarán Mooney wrote: > Hi, > > Friendly reminder of this months meeting, details about the talk can > be found here: > > http://lists.gllug.org.uk/pipermail/gllug/2010-February/082543.html > > Seán Tohill has asked that I keep track of the number of attendees. > Currently 15 people have said that they will be attending. Please > email the list if you plan to attend. I would like to attend - please put me as a maybe. > Regards, > > Ciarán Martin -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 190 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100305/50c111d3/attachment.pgp From GLLUG at GetAroundToIt.co.uk Fri Mar 5 18:26:54 2010 From: GLLUG at GetAroundToIt.co.uk (David L Neil) Date: Sat, 06 Mar 2010 07:26:54 +1300 Subject: [Gllug] Small PC, resource constraints, improving efficiencies In-Reply-To: <201003050922.o259MxVS013877@leto.astradyne.corp> References: <4B90B3BB.5070405@getaroundtoit.co.uk> <201003050922.o259MxVS013877@leto.astradyne.corp> Message-ID: <4B914CEE.3090701@GetAroundToIt.co.uk> Tet, On 05/03/10 22:22, Tethys wrote: > David L Neil writes: >> Should I trial LXDE and/or XFCE by attempting to run multiple Desktop >> Environments/Window Managers at the X level, or by dual-booting whole >> distributions with the appropriate choice as the sole GUI? > I don't really think it makes much difference, although multiple > desktop environments is a less invasive solution. >> *have thought to try Opera 10 (and maybe its Dragonfly system for >> web-dev) and compare load factors with FF...? > I'd definitely recommend Opera. I used to hate it with a passion, > but recent versions have been very, very good, and having it start > up in a second is a wonderfully refreshing change from Firefox's > bloat. =exactly my aim. My problem was that Firefox without the Web Developer, Firebug, etc, add-ons was like eggs without bacon. I've only recently come to my senses and employed multiple profiles (eg web-surf and web-dev) but even so, slimmed-down FF is still a beast - and non-JS browsers are too restrictive on today's web even when one avoids video as much as possible, Flash-happy sites... >> XFCE seems to be available from the CentOS?rpmforge repositories but >> LXDE is not found by search. > How much do you want from a desktop environment? =not much (at the appearance level, cf the underlying mechanics ?windows/applications management). =Analysed right down: at the desktop I rarely touch more than the five panel icons for the listed apps (FF, Tb, OOo-Writer, Calc, VNC), plus the power button (although now I've configured the hardware button, reliably), plus Nautilus and Terminal. Other stuff like GEdit and the PDF reader tend to be started by clicking on downloaded files sitting on the desktop, which I use as my staging area. Date-time widget. Don't use multiple desktops. Don't care for screen-savers and fancy backgrounds... The rest is probably all 'mechanics'. =basically, as long as the apps look-and-feel roughly the same, most of the eye-candy is unnecessary and I'd rather put the (v.scarce machine) resources into another FF tab or similar. Yes, I'm boring! =being used to running servers without GUIs, I'd be happy to run Thunderbird & from the X11 cmdLN - is such possible??? I live in the app not on the desktop!? =on my travels, it is configured as a single user machine, and connecting to/through public libraries and 'strange' networks makes security more of a concern. If you're looking for > a lightweight solution, then you might want to look at FVWM. It's in > EPEL, so installing it on CentOS is a breeze. It's configured through > a config file, rather than dragging and dropping things onto a task > bar, for example, but it's insanely powerful, and light on resources. > I've tried KDE and GNOME, but found them to be limiting, and have > always returned to FVWM. It does what I want with a minimum of fuss. =I have seen this, but never investigated - will do so. =From first glance it seems not to replace Gnome* but only Metacity. Will it result in significant resource savings? *perhaps my ignorance shows: are the others only/much the same? I figured that having trimmed-down as many of the CentOS services as I dare, I could maximise the resources available to the apps. Gnome seems to be over-kill for my 'five buttons' (as above), hence my ponderous ponderings... >> Alternatively, should I simply load the other DktopEnv/WinMgrs on top of >> the existing CentOS X11 (and hopefully rewire things with the Desktop >> switcher)? >> - but I'm concerned that removal after failure might be messy/risk the >> current system-config? > I think that's an ungrounded fear. I habitually try things out on a > regular basis, and when I decide I'm done, a quick "yum erase" gets > the box back to its previous state without any problems. =I have tried various things out, but then killed all sorts of other stuff when yum erase-ing afterwards. A little bit of knowledge is a dangerous thing (at least in my case)... =Having spent sooooooo long getting this little machine working (again - after a HDD crash shortly AFTER leaving London), I'm frankly wary of mucking it up in a (ignorant) bid to make it better! > Tet =Many thanks, =dn From tethys at gmail.com Fri Mar 5 18:45:18 2010 From: tethys at gmail.com (- Tethys) Date: Fri, 5 Mar 2010 18:45:18 +0000 Subject: [Gllug] Small PC, resource constraints, improving efficiencies In-Reply-To: <4B914CEE.3090701@GetAroundToIt.co.uk> References: <4B90B3BB.5070405@getaroundtoit.co.uk> <201003050922.o259MxVS013877@leto.astradyne.corp> <4B914CEE.3090701@GetAroundToIt.co.uk> Message-ID: <264d99b01003051045y2daee791g93c235de0f5f62b0@mail.gmail.com> On Fri, Mar 5, 2010 at 6:26 PM, David L Neil wrote: > =From first glance it seems not to replace Gnome* but only Metacity. > Will it result in significant resource savings? > > *perhaps my ignorance shows: are the others only/much the same? I > figured that having trimmed-down as many of the CentOS services as I > dare, I could maximise the resources available to the apps. Gnome seems > to be over-kill for my 'five buttons' (as above), hence my ponderous > ponderings... Exactly. Since I've still yet to find anyone that can give me a concrete definition of what GNOME actually is, it's hard to say. But FVWM is just a window manager. You could use it to replace metacity, but like you say, that won't buy you much other than a nicer user experience. The question is, do you need the rest of whatever it is that GNOME gives you? FvwmButtons can give you an application launcher, should you want one, which removes part of the reason for a GNOME panel[1]. If you want a file manager, should should be able to run nautilus as a standalone app in conjunction with FVWM. There are no shortage of X11 clocks to replace a date/time widget. Tet [1] The panel acts as both an application launcher and a home for iconified applications. FvwmButtons definitely does the first part, and you may be able to configure it to do the latter as well. I've always just stuck to launching things from desktop menus, rather than wasting screen real estate with a separate launcher, and my iconified apps go to designated places on the desktop, so I'm not best placed to talk authoritatively about FvwmButtons. -- ?It seems intuitively obvious to me, which means that it might be wrong.? -- Chris Torek From rich at annexia.org Sat Mar 6 10:37:56 2010 From: rich at annexia.org (Richard Jones) Date: Sat, 6 Mar 2010 10:37:56 +0000 Subject: [Gllug] [OT] Re: [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> Message-ID: <20100306103756.GB13486@annexia.org> On Thu, Mar 04, 2010 at 10:36:44AM +0000, Peter Childs wrote: > On 4 March 2010 07:43, Nix wrote: > > On 4 Mar 2010, Christopher Hunter uttered the following: > > > >> On Wed, 2010-03-03 at 22:28 +0000, Nix wrote: > >> > >>> Rubbish. The debts will inflate away and be grown away. > >> > >> Not this time.  There's nowhere for the hole to disappear to, this time. > > > > What? Inflation and growth can dispose of *any* debt, given time, though > > if it goes too far your currency implodes into hyperinflation (which > > also disposes of the debt, although as it also disposes of your entire > > nation's savings it tends to do your political prospects no good at > > all). > > > > We are very, very far from hyperinflation right now. > > -- > > Does that work with personal debt too? Yes. It's one reason why my parent's generation all live in much bigger houses than I will ever do. Mortgages they took out in the early 70s inflated away. Look at the bulge in this list of figures: http://safalra.com/other/historical-uk-inflation-price-conversion/ Rich. -- Richard Jones Red Hat From addw at phcomp.co.uk Sat Mar 6 11:19:05 2010 From: addw at phcomp.co.uk (Alain Williams) Date: Sat, 6 Mar 2010 11:19:05 +0000 Subject: [Gllug] [OT] Re: [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <20100306103756.GB13486@annexia.org> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> <20100306103756.GB13486@annexia.org> Message-ID: <20100306111905.GU13800@phcomp.co.uk> On Sat, Mar 06, 2010 at 10:37:56AM +0000, Richard Jones wrote: > Yes. It's one reason why my parent's generation all live in much > bigger houses than I will ever do. Mortgages they took out in the > early 70s inflated away. Look at the bulge in this list of figures: > > http://safalra.com/other/historical-uk-inflation-price-conversion/ I was going to say that house prices have risen far faster than RPI, but decided to check (I know that this is not playing the game :-) ), there are some contradictions, but the best seems to be: http://www.pricedout.org.uk/Articles/StatisticsHousepriceinflationdata/tabid/99/Default.aspx it shows that house prices have been between 3-5 times salary, but 6 times for the last few years. There is a lot of regional variation. Part of the problem is the deregulation of the finance industry, you used to be able to get 3-3.5 times income as a mortgage + a large deposit, then 5-6 times & no deposit was the norm; the result was house price inflation and we all end up paying more interest to the banks. I think that we need reguation to be brought in, to make house prices more affordable. -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php Past chairman of UKUUG: http://www.ukuug.org/ #include From martin at hinterlands.org Sat Mar 6 11:20:03 2010 From: martin at hinterlands.org (Martin A. Brooks) Date: Sat, 6 Mar 2010 11:20:03 -0000 Subject: [Gllug] Automatic mail archiving Message-ID: If you're on a lot of mailing lists, you might find this handy: http://bit.ly/axamTe From rich at annexia.org Sat Mar 6 11:20:45 2010 From: rich at annexia.org (Richard Jones) Date: Sat, 6 Mar 2010 11:20:45 +0000 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <4B902327.70707@GetAroundToIt.co.uk> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <4B902327.70707@GetAroundToIt.co.uk> Message-ID: <20100306112045.GC13486@annexia.org> On Fri, Mar 05, 2010 at 10:16:23AM +1300, David L Neil wrote: > Compare this with the UK and US governments giving wads of cash to the > merchant bankers, who instead of passing it on (even to their > oft-sacrificed retail brethren), simply held on to it and used it as > fuel for the bonuses they are paying themselves today. The US also distributed cash to the population, in the form of "rebate checks" (cheques) in the $300-$600 range to most taxpayers. It's not really clear how much effect that had. Wouldn't sensible people have mainly held on to this money because of their uncertainly over future prospects? They would have saved it in the bank, and the banks would still have had to increase their minimum capital ratios, so would not have been lending it to anyone. Rich. -- Richard Jones Red Hat From rich at annexia.org Sat Mar 6 11:49:27 2010 From: rich at annexia.org (Richard Jones) Date: Sat, 6 Mar 2010 11:49:27 +0000 Subject: [Gllug] [OT] Re: [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <20100306111905.GU13800@phcomp.co.uk> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> <20100306103756.GB13486@annexia.org> <20100306111905.GU13800@phcomp.co.uk> Message-ID: <20100306114927.GD13486@annexia.org> On Sat, Mar 06, 2010 at 11:19:05AM +0000, Alain Williams wrote: > On Sat, Mar 06, 2010 at 10:37:56AM +0000, Richard Jones wrote: > > > Yes. It's one reason why my parent's generation all live in much > > bigger houses than I will ever do. Mortgages they took out in the > > early 70s inflated away. Look at the bulge in this list of figures: > > > > http://safalra.com/other/historical-uk-inflation-price-conversion/ > > I was going to say that house prices have risen far faster than RPI, but > decided to check (I know that this is not playing the game :-) ), there > are some contradictions, but the best seems to be: > > http://www.pricedout.org.uk/Articles/StatisticsHousepriceinflationdata/tabid/99/Default.aspx > > it shows that house prices have been between 3-5 times salary, but 6 times for the > last few years. There is a lot of regional variation. Doesn't matter. The point is that if you entered into a mortgage for £ 100 at the end of 1968 (the year my parents got their first mortgage), by the end of 1969 the original loan has effectively gone down to 100 - (100 * 0.054) = £ 94.60. By 1978, ten years later, the loaned amount is down to £ 36.63 and by 1988, it's down to £ 16.67. Effectively the original debt has (mostly) disappeared. This is what we mean when we say that inflation gets rid of the loan. If you knew that inflation was going to happen ahead of time[1], then you would pay the minimum possible amount back each year to save the loan from repossession (if possible, you'd even allow a little interest to pile up, because that debt gets eroded by inflation over time too). (Note: This only applies if the interest rate is fixed, or lower than the inflation rate, which was true for much of the 70s: http://www.bankofengland.co.uk/statistics/rates/baserate.pdf ) This has nothing to do with whether house prices rise or fall over the period, but if they rise faster than inflation (which they did), you're doing doubly well. You got a free house which is now worth a fortune. Now all you need is a younger generation of suckers who will rack up huge debts which they'll be paying back for decades, and give you that money. Spend your retirement playing golf and going on cruises ... Rich. [1] Of course, inflation has a lot of downsides too, like your cash savings disappear at the same rate. Turning them into a tangible commodity (property, gold, oil, whatever) makes sense if you expect inflation to go up. Land, for example, is unlikely to go down in value unless either the world population falls or we open up new tracts of land on the moon, or start building artificial islands in the ocean ... (Is this getting off-topic :-?) -- Richard Jones Red Hat From rich at annexia.org Sat Mar 6 11:57:42 2010 From: rich at annexia.org (Richard Jones) Date: Sat, 6 Mar 2010 11:57:42 +0000 Subject: [Gllug] Automatic mail archiving In-Reply-To: References: Message-ID: <20100306115742.GE13486@annexia.org> On Sat, Mar 06, 2010 at 11:20:03AM -0000, Martin A. Brooks wrote: > > If you're on a lot of mailing lists, you might find this handy: > http://bit.ly/axamTe Was it necessary to shorten that URL? The URL itself isn't really so long: http://hinterlands.org/wiki/index.php/MailArchiveScript Rich. -- Richard Jones Red Hat From martin at hinterlands.org Sat Mar 6 11:59:08 2010 From: martin at hinterlands.org (Martin A. Brooks) Date: Sat, 6 Mar 2010 11:59:08 -0000 Subject: [Gllug] Automatic mail archiving In-Reply-To: <20100306115742.GE13486@annexia.org> References: <20100306115742.GE13486@annexia.org> Message-ID: On Sat, March 6, 2010 11:57, Richard Jones wrote: > Was it necessary to shorten that URL? The URL itself isn't > really so long: Force of Twitter habit. From andy at andymillar.co.uk Sat Mar 6 12:02:05 2010 From: andy at andymillar.co.uk (Andy Millar) Date: Sat, 6 Mar 2010 12:02:05 -0000 Subject: [Gllug] Automatic mail archiving In-Reply-To: References: Message-ID: <1267876925.26839.17.camel@millaralpt.co.tradefair> On Sat, 2010-03-06 at 11:20 +0000, Martin A. Brooks wrote: > If you're on a lot of mailing lists, you might find this handy: > http://bit.ly/axamTe That's quite a nice idea. I do exactly the same thing, manually, when I've read/dealt with an email, that way my Inbox really is for things that still need dealing with. I *hate* the Gmail idea of everything being in your Inbox (that said, I've never actually used Gmail) and people who use their Deleted Items folder as an archive for their old mail (and they really don't see why this is a bad idea). That said, I actually organise mine by month instead of year. Andy -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 5162 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100306/36792b5a/attachment.bin From rich at annexia.org Sat Mar 6 12:16:48 2010 From: rich at annexia.org (Richard Jones) Date: Sat, 6 Mar 2010 12:16:48 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100304183609.GV28166@cornerstonelinux.co.uk> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> <20100304183609.GV28166@cornerstonelinux.co.uk> Message-ID: <20100306121648.GF13486@annexia.org> On Thu, Mar 04, 2010 at 06:36:09PM +0000, John Edwards wrote: > On Thu, Mar 04, 2010 at 06:27:56PM +0000, Dennis Furey wrote: > > On Thu, Mar 04, 2010 at 06:07:43PM +0000, Benjamin Donnachie wrote: > >> On 4 March 2010 18:05, Dennis Furey wrote: > >>> Any reason no one has mentioned LUKS? It's well supported, based on > >>> AES encryption, and I've used it for years without any issues. > >> > >> I'm happy to be corrected, but I don't believe LUKS is available for MacOS. > > > > True, it's only for Linux. I didn't notice MacOS in the original > > post. Back to lurking. > > No worries. > > How do you use LUKS? I click the "encrypt the disk" box when installing Fedora :-) Other distros also support this. Configuring LUKS manually is a bit of a pain in the derriere, although I've done that too. This seems to cover the basics: http://feraga.com/library/howto_use_cryptsetup_with_luks_support_0 > Certain files, whole filesystem, root filesystem? This depends a lot on your threat model. If you just want to ensure that thieves can't browse through the contents of your laptop if it is stolen, then I would just use whole disk encryption (ie. check the "encrypt the disk" box), and make sure you turn the laptop off instead of suspending it when it's not under your control. There's some performance penalty for encrypting data, so there is an argument for not encrypting the OS volume (only /home). However I wouldn't bother with that: (a) the penalty is really small because CPUs are very fast compared to hard drives, and (b) personal data can be stored in parts of the OS volume (eg. /tmp or /var) unless you are very careful. Encrypting the whole lot is just much easier. You should be aware that if you use whole-disk encryption on a server, then you must be physically present to type in a passphrase when the server boots. For servers this is usually not so convenient! For VMs, libvirt now supports storing encryption keys. I've not actually tried this feature, but I believe the idea is that libvirt manages the secret passphrase and both keeps it secure and provides it at boot time so no manual intervention is necessary. (http://libvirt.org/formatstorageencryption.html) Finally, if you want to hide personal / sensitive / illegal information from authorities, then you would want a two-level system, where you use another layer of encryption that cannot be detected or you can plausibly deny knowledge of. TrueCrypt can allegedly do this (it's called a "hidden volume" -- I've never tried). LUKS can't do this, except by just obfuscating the file so it will still be quite obvious to anyone looking that there is some encrypted data they cannot read. There's a BZ open for a long time for supporting hidden volumes in Fedora: https://bugzilla.redhat.com/show_bug.cgi?id=454855 Rich. -- Richard Jones Red Hat From tethys at gmail.com Sat Mar 6 12:28:09 2010 From: tethys at gmail.com (- Tethys) Date: Sat, 6 Mar 2010 12:28:09 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100306121648.GF13486@annexia.org> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> <20100304183609.GV28166@cornerstonelinux.co.uk> <20100306121648.GF13486@annexia.org> Message-ID: <264d99b01003060428k146e21belffdba09ef2c210de@mail.gmail.com> On Sat, Mar 6, 2010 at 12:16 PM, Richard Jones wrote: > You should be aware that if you use whole-disk encryption on a server, > then you must be physically present to type in a passphrase when the > server boots.  For servers this is usually not so convenient! Well, no. You need access to the console, which isn't quite the same thing. I have zero sympathy for anyone that installs a server in a remote datacentre without also adding a serial console. Of course, you still need to enter the password, which means no unattended reboots. But if the machine is sufficiently important, you'll have an alert sent to your phone/pager to let you know that it needs attention anyway. Tet -- ?It seems intuitively obvious to me, which means that it might be wrong.? -- Chris Torek From andy at andymillar.co.uk Sat Mar 6 12:46:52 2010 From: andy at andymillar.co.uk (Andy Millar) Date: Sat, 6 Mar 2010 12:46:52 -0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <264d99b01003060428k146e21belffdba09ef2c210de@mail.gmail.com> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> <20100304183609.GV28166@cornerstonelinux.co.uk> <20100306121648.GF13486@annexia.org> <264d99b01003060428k146e21belffdba09ef2c210de@mail.gmail.com> Message-ID: <1267879612.26839.21.camel@millaralpt.co.tradefair> On Sat, 2010-03-06 at 12:28 +0000, - Tethys wrote: > Well, no. You need access to the console, which isn't quite the same > thing. I have zero sympathy for anyone that installs a server in a > remote datacentre without also adding a serial console. And all modern decent servers come with some form of OOB management, a la HP's iLO or the Dell DRAC cards. Given the price at which these servers are being given away at, there's really no excuse for getting anything without enterprise management features. Andy -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 5162 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100306/f40a5eee/attachment-0001.bin From rich at annexia.org Sat Mar 6 12:52:56 2010 From: rich at annexia.org (Richard Jones) Date: Sat, 6 Mar 2010 12:52:56 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <264d99b01003060428k146e21belffdba09ef2c210de@mail.gmail.com> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> <20100304183609.GV28166@cornerstonelinux.co.uk> <20100306121648.GF13486@annexia.org> <264d99b01003060428k146e21belffdba09ef2c210de@mail.gmail.com> Message-ID: <20100306125256.GA21489@annexia.org> On Sat, Mar 06, 2010 at 12:28:09PM +0000, - Tethys wrote: > On Sat, Mar 6, 2010 at 12:16 PM, Richard Jones wrote: > > > You should be aware that if you use whole-disk encryption on a server, > > then you must be physically present to type in a passphrase when the > > server boots.  For servers this is usually not so convenient! > > Well, no. You need access to the console, which isn't quite the same > thing. I have zero sympathy for anyone that installs a server in a > remote datacentre without also adding a serial console. Of course, > you still need to enter the password, which means no unattended > reboots. But if the machine is sufficiently important, you'll have > an alert sent to your phone/pager to let you know that it needs > attention anyway. True true. However when I temporarily had my desktop machine located in my old house just after moving to my new house, I cursed whole disk encryption about 3 minutes after I rebooted the desktop machine remotely ... Rich. -- Richard Jones Red Hat From white.gavin at gmail.com Sat Mar 6 14:11:31 2010 From: white.gavin at gmail.com (Gavin White) Date: Sat, 6 Mar 2010 14:11:31 +0000 Subject: [Gllug] VACANCY - Senior sysadmin in Aldgate for Financial company Message-ID: <451eaded1003060611y7a9f594cu571a920d2783ba09@mail.gmail.com> Location: Aldgate Type of employer: Brokerage Salary: 50-70k Skills: Mostly RHEL, some networking, some opensolaris We are looking for a senior sysadmin to look after a rhel/centos/opensolaris estate. The position is available now. You will be administering 250 servers using a mix of xen and esx virtualisation on HP hardware, with fibre channel and iscsi storage. The servers run a mix of java and oracle services. Please send me your CV if you would like further details. Regards, Gavin From john at sinodun.org.uk Sat Mar 6 18:01:38 2010 From: john at sinodun.org.uk (John Winters) Date: Sat, 06 Mar 2010 18:01:38 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films Message-ID: <4B929882.9030608@sinodun.org.uk> I've just bought a 2 TB Seagate USB drive (130 squids from Maplin) to store my film collection on, replacing an older 750G one. I was pleased to see that at last it didn't come formatted with one large FAT partition. Instead it had one large NTFS partition, so that had to go. Is there any reason for me not to just make one large ext3 partition? Is there anything else more suited to a large filesystem? Note that there will be a relatively small number of largish (2 to 4 gig) files on it. TIA, John From nix at esperi.org.uk Sat Mar 6 18:48:01 2010 From: nix at esperi.org.uk (Nix) Date: Sat, 06 Mar 2010 18:48:01 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <4B929882.9030608@sinodun.org.uk> (John Winters's message of "Sat, 06 Mar 2010 18:01:38 +0000") References: <4B929882.9030608@sinodun.org.uk> Message-ID: <87k4tpthge.fsf@spindle.srvr.nix> On 6 Mar 2010, John Winters uttered the following: > I've just bought a 2 TB Seagate USB drive (130 squids from Maplin) to > store my film collection on, replacing an older 750G one. > > I was pleased to see that at last it didn't come formatted with one > large FAT partition. Instead it had one large NTFS partition, so that > had to go. :) > Is there any reason for me not to just make one large ext3 partition? > Is there anything else more suited to a large filesystem? Even if you use -T largefile4, you will curse the day you chose to use ext3 the day it fscks for the first time. Natively-created ext4 filesystems are much much faster, especially if built with -G 64 or something like that (which also speeds up metadata-heavy writes, but you're not doing that with this fs: still, the effect on fsck speeds is notable). My 200Gb /usr (with 50Gb in it) fscks in about twenty seconds. Over USB it'll be a good bit slower, but still enormously faster than ext3 was. > Note that > there will be a relatively small number of largish (2 to 4 gig) files on it. Extents will lead to slightly less disk activity as reads happen, and preallocation will possibly lead to a *lot* less fragmentation of those files. From john at sinodun.org.uk Sat Mar 6 18:53:16 2010 From: john at sinodun.org.uk (John Winters) Date: Sat, 06 Mar 2010 18:53:16 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <87k4tpthge.fsf@spindle.srvr.nix> References: <4B929882.9030608@sinodun.org.uk> <87k4tpthge.fsf@spindle.srvr.nix> Message-ID: <4B92A49C.4050602@sinodun.org.uk> Nix wrote: > Even if you use -T largefile4, you will curse the day you chose to use ext3 > the day it fscks for the first time. Natively-created ext4 filesystems > are much much faster, especially if built with -G 64 or something like that > (which also speeds up metadata-heavy writes, but you're not doing that > with this fs: still, the effect on fsck speeds is notable). I have yet to venture into the area of ext4. I notice my kernel (2.6.26) has support but lists it as "experimental". What kernel do I need to go to before I can confidently use ext4? John From nix at esperi.org.uk Sat Mar 6 20:37:45 2010 From: nix at esperi.org.uk (Nix) Date: Sat, 06 Mar 2010 20:37:45 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <4B92A49C.4050602@sinodun.org.uk> (John Winters's message of "Sat, 06 Mar 2010 18:53:16 +0000") References: <4B929882.9030608@sinodun.org.uk> <87k4tpthge.fsf@spindle.srvr.nix> <4B92A49C.4050602@sinodun.org.uk> Message-ID: <87zl2lrxt2.fsf@spindle.srvr.nix> On 6 Mar 2010, John Winters verbalised: > Nix wrote: >> Even if you use -T largefile4, you will curse the day you chose to use ext3 >> the day it fscks for the first time. Natively-created ext4 filesystems >> are much much faster, especially if built with -G 64 or something like that >> (which also speeds up metadata-heavy writes, but you're not doing that >> with this fs: still, the effect on fsck speeds is notable). > > I have yet to venture into the area of ext4. I notice my kernel > (2.6.26) has support but lists it as "experimental". What kernel do I > need to go to before I can confidently use ext4? I started using it in 2.6.30. The only changes since 2.6.30 that I've noticed are changes to prevent data damage due to lack of journal replay after an abrupt power loss during a journal wraparound (that was 2.6.32, a fix to a bug exposed by a change in 2.6.31, but the fix will have been backported to affected distro kernels), fairly small block allocator improvements, and changes to fix the ioctl used by the yet-unwritten ext4 online defragmenter. Any halfway recent kernel should be fine. The class of bugs that are now being squashed are terribly rare and hard to trip over. I'd call it reliable. (I'm afraid I can't remember much about 2.6.26, it was so long ago. I'm astonished that half the systems at work are still running a deformed RHEL-patched-to-hell 2.6.9!) From caroline.ford.work at googlemail.com Sat Mar 6 20:56:41 2010 From: caroline.ford.work at googlemail.com (Caroline Ford) Date: Sat, 6 Mar 2010 20:56:41 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <4B929882.9030608@sinodun.org.uk> References: <4B929882.9030608@sinodun.org.uk> Message-ID: You'll regret ext3 if you ever need to use the drive on a non-Linux machine. I regretted making my music drive ext3 when Ubuntu had problems with permissions it just didn't have on NTFS. Caroline Sent from a mobile device. On 6 Mar 2010, at 18:01, John Winters wrote: > I've just bought a 2 TB Seagate USB drive (130 squids from Maplin) to > store my film collection on, replacing an older 750G one. > > I was pleased to see that at last it didn't come formatted with one > large FAT partition. Instead it had one large NTFS partition, so that > had to go. > > Is there any reason for me not to just make one large ext3 partition? > Is there anything else more suited to a large filesystem? Note that > there will be a relatively small number of largish (2 to 4 gig) > files on it. > > TIA, > John > -- > Gllug mailing list - Gllug at gllug.org.uk > http://lists.gllug.org.uk/mailman/listinfo/gllug From hearnsj at googlemail.com Sat Mar 6 21:07:10 2010 From: hearnsj at googlemail.com (John Hearns) Date: Sat, 6 Mar 2010 21:07:10 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <4B929882.9030608@sinodun.org.uk> References: <4B929882.9030608@sinodun.org.uk> Message-ID: <9f8092cc1003061307p2ec83508o74befbb39db60ca0@mail.gmail.com> I would use XFS for a large filesystem. I'm speaking from experience - I've been using xfs since the days of Irix, and still manage SGI equipment for multu-Tbyte data stores. From john at sinodun.org.uk Sun Mar 7 07:34:33 2010 From: john at sinodun.org.uk (John Winters) Date: Sun, 07 Mar 2010 07:34:33 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: References: <4B929882.9030608@sinodun.org.uk> Message-ID: <4B935709.8070809@sinodun.org.uk> Caroline Ford wrote: > You'll regret ext3 if you ever need to use the drive on a non-Linux > machine. "Non-Linux machine"? Sorry, not a phrase I'm familiar with. :-) > I regretted making my music drive ext3 when Ubuntu had > problems with permissions it just didn't have on NTFS. I'm not sure what you're saying there. I've never encountered any *bugs* in ext2/ext3 permission handling in over 15 years of using them. Of course, the whole idea of permissions sometimes surprises people. Ext3 also has the advantage over NTFS that it's stable and documented, which means that you can if you want use it on either Windows or Mac OS. Nix has suggested ext4 as an improvement on ext3 and I'm currently looking into it. Cheers, John From cehunter at gb-x.org Sun Mar 7 08:08:10 2010 From: cehunter at gb-x.org (Christopher Hunter) Date: Sun, 07 Mar 2010 08:08:10 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <87k4tpthge.fsf@spindle.srvr.nix> References: <4B929882.9030608@sinodun.org.uk> <87k4tpthge.fsf@spindle.srvr.nix> Message-ID: <1267949290.32115.5.camel@localhost> On Sat, 2010-03-06 at 18:48 +0000, Nix banged the rocks together and said: > Even if you use -T largefile4, you will curse the day you chose to use ext3 > the day it fscks for the first time. That's a lesson I learned just yesterday, alas. > Natively-created ext4 filesystems > are much much faster, especially if built with -G 64 or something like that > (which also speeds up metadata-heavy writes, but you're not doing that > with this fs: still, the effect on fsck speeds is notable). Thanks for the advice. I've backed everything up to another drive (ext 4) and will repartition the original one in the same way. Once again, drives are getting big enough that housekeeping is taking significant time - we went through a period with ever faster drives, but without much increase in capacity. I received five 2 Tb drives yesterday, and I'm going to build a UPS backed-up NAS using a spare machine I have here.... C. From rich at annexia.org Sun Mar 7 11:26:40 2010 From: rich at annexia.org (Richard Jones) Date: Sun, 7 Mar 2010 11:26:40 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <4B929882.9030608@sinodun.org.uk> References: <4B929882.9030608@sinodun.org.uk> Message-ID: <20100307112640.GA3917@annexia.org> On Sat, Mar 06, 2010 at 06:01:38PM +0000, John Winters wrote: > I've just bought a 2 TB Seagate USB drive (130 squids from Maplin) to > store my film collection on, replacing an older 750G one. > > I was pleased to see that at last it didn't come formatted with one > large FAT partition. Instead it had one large NTFS partition, so that > had to go. > > Is there any reason for me not to just make one large ext3 partition? > Is there anything else more suited to a large filesystem? Note that > there will be a relatively small number of largish (2 to 4 gig) files on it. FWIW: http://rwmj.wordpress.com/2009/11/08/filesystem-metadata-overhead/ If you decide to use more than one partition, you may need to partition the disk using EFI instead of MBR (depending on whether the disk is actually larger than the 2 TB limit for MBR-style partition tables). Rich. -- Richard Jones Red Hat From nix at esperi.org.uk Sun Mar 7 13:32:17 2010 From: nix at esperi.org.uk (Nix) Date: Sun, 07 Mar 2010 13:32:17 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <4B935709.8070809@sinodun.org.uk> (John Winters's message of "Sun, 07 Mar 2010 07:34:33 +0000") References: <4B929882.9030608@sinodun.org.uk> <4B935709.8070809@sinodun.org.uk> Message-ID: <87pr3gs1em.fsf@spindle.srvr.nix> On 7 Mar 2010, John Winters uttered the following: > Caroline Ford wrote: >> You'll regret ext3 if you ever need to use the drive on a non-Linux >> machine. > > "Non-Linux machine"? Sorry, not a phrase I'm familiar with. :-) > >> I regretted making my music drive ext3 when Ubuntu had >> problems with permissions it just didn't have on NTFS. > > I'm not sure what you're saying there. I've never encountered any > *bugs* in ext2/ext3 permission handling in over 15 years of using them. Given that all ext[234] does with permissions is store them, I'm not surprised :) any bugs would be in the VFS if they were anywhere. > Of course, the whole idea of permissions sometimes surprises people. Permissions on removable devices are particularly strange to deal with. (Also, obviously, mounting with nosuid,nodev and possibly noexec is a good idea if it's a user mount.) Also, it is probably still possible to feed the kernel a sufficiently corrupted ext[234] filesystem and have it execute arbitrary code, or crash (these bugs are still being squashed). But you're not allowing hostile local users to mount these filesystems so that's not a concern. > Ext3 also has the advantage over NTFS that it's stable and documented, > which means that you can if you want use it on either Windows or Mac OS. Take care. It's not as 'stable' as all that. A filesystem with a given combination of feature flags turned on should always be readable, and depending on the flags, writable, on a given system. But changes to the FS *do* happen and flip new flags on (e.g. htree indexing). From benjamin at py-soft.co.uk Sun Mar 7 13:42:18 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Sun, 7 Mar 2010 13:42:18 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <87pr3gs1em.fsf@spindle.srvr.nix> References: <4B929882.9030608@sinodun.org.uk> <4B935709.8070809@sinodun.org.uk> <87pr3gs1em.fsf@spindle.srvr.nix> Message-ID: <732076a81003070542j17bf852br69ed0b7533b1068e@mail.gmail.com> On 7 March 2010 13:32, Nix wrote: > Also, it is probably still possible to feed the kernel a sufficiently > corrupted ext[234] filesystem and have it execute arbitrary code, or > crash (these bugs are still being squashed). But you're not allowing > hostile local users to mount these filesystems so that's not a concern. Interesting... Have you got any sources for this to hand? Ben From rich at annexia.org Sun Mar 7 14:07:22 2010 From: rich at annexia.org (Richard Jones) Date: Sun, 7 Mar 2010 14:07:22 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <732076a81003070542j17bf852br69ed0b7533b1068e@mail.gmail.com> References: <4B929882.9030608@sinodun.org.uk> <4B935709.8070809@sinodun.org.uk> <87pr3gs1em.fsf@spindle.srvr.nix> <732076a81003070542j17bf852br69ed0b7533b1068e@mail.gmail.com> Message-ID: <20100307140722.GA9261@annexia.org> On Sun, Mar 07, 2010 at 01:42:18PM +0000, Benjamin Donnachie wrote: > On 7 March 2010 13:32, Nix wrote: > > Also, it is probably still possible to feed the kernel a sufficiently > > corrupted ext[234] filesystem and have it execute arbitrary code, or > > crash (these bugs are still being squashed). But you're not allowing > > hostile local users to mount these filesystems so that's not a concern. > > Interesting... Have you got any sources for this to hand? These do pop up from time to time. The latest ext3 one was, I think, this one from four years ago: http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-6053 Here's another one from 2006 affecting ISO 9660 handling (ie. CDs, so this is really quite serious): http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-5757 This is an HFS buffer overflow in the kernel discovered a few months ago: https://bugzilla.redhat.com/show_bug.cgi?id=540736 The general advice is still (unfortunately) don't mount *any* untrusted filesystems directly on your machines. Libguestfs gives you some greater protection here, because not only would the crafted filesystem have to exploit the kernel (which is virtualized in libguestfs), but it would also have to exploit either the libguestfs protocol (well understood, simple, robust and checked), or the qemu container. Also libguestfs encourages you to _not_ run everything as root. Rich. -- Richard Jones Red Hat From nix at esperi.org.uk Sun Mar 7 16:50:43 2010 From: nix at esperi.org.uk (Nix) Date: Sun, 07 Mar 2010 16:50:43 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <20100307140722.GA9261@annexia.org> (Richard Jones's message of "Sun, 7 Mar 2010 14:07:22 +0000") References: <4B929882.9030608@sinodun.org.uk> <4B935709.8070809@sinodun.org.uk> <87pr3gs1em.fsf@spindle.srvr.nix> <732076a81003070542j17bf852br69ed0b7533b1068e@mail.gmail.com> <20100307140722.GA9261@annexia.org> Message-ID: <87iq98oz30.fsf@spindle.srvr.nix> On 7 Mar 2010, Richard Jones stated: > On Sun, Mar 07, 2010 at 01:42:18PM +0000, Benjamin Donnachie wrote: >> On 7 March 2010 13:32, Nix wrote: >> > Also, it is probably still possible to feed the kernel a sufficiently >> > corrupted ext[234] filesystem and have it execute arbitrary code, or >> > crash (these bugs are still being squashed). But you're not allowing >> > hostile local users to mount these filesystems so that's not a concern. >> >> Interesting... Have you got any sources for this to hand? > > These do pop up from time to time. The latest ext3 one was, I think, > this one from four years ago: > > http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-6053 DoS attacks are relatively common. Spotted via a search of the fs/ext3 history for 'corrupt': commit de18f3b2d68c1f3481839be760a5ff93f6a9a5e5 Author: Bryan Donlan Date: Thu Apr 2 16:57:15 2009 -0700 ext3: return -EIO not -ESTALE on directory traversal through deleted inode commit a21102b55c4f8dfd3adb4a15a34cd62237b46039 Author: Theodore Ts'o Date: Fri Jan 16 11:13:47 2009 -0500 ext3: Add sanity check to make_indexed_dir commit b5ed3112b5f74c8ec1c7aa03a76c596635e85197 Author: Duane Griffin Date: Fri Dec 19 20:47:14 2008 +0000 ext3: ensure fast symlinks are NUL-terminated commit cdbf6dba28e8e6268c8420857696309470009fd9 Author: Eric Sandeen Date: Sat Oct 18 20:28:00 2008 -0700 ext3: avoid printk floods in the face of directory corruption commit 3ccc3167b0e5d46ab3bf03e22fbdb7616ce038cd Author: Duane Griffin Date: Fri Jul 25 01:46:26 2008 -0700 ext3: handle deleting corrupted indirect blocks [...] > Here's another one from 2006 affecting ISO 9660 handling (ie. CDs, so > this is really quite serious): > > http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-5757 Some iso9660 and ufs ones: commit 2deb1acc653cbd5384b107d050d2deba089db2bd Author: Jan Kara Date: Wed Apr 30 00:52:33 2008 -0700 isofs: fix access to unallocated memory when reading corrupted filesystem commit 817794e0df5fea495396c18878804044436832be Author: Kirill Kuvaldin Date: Tue Jul 31 00:38:58 2007 -0700 isofs: mounting to regular file may succeed commit b12903f1384cd176a3994a6bf6caf5a482169cc8 Author: Duane Griffin Date: Thu Jan 8 22:43:50 2009 +0000 ufs: ensure fast symlinks are NUL-terminated commit f33219b7a90c4779a0b59e11fb35ebc4542db328 Author: Duane Griffin Date: Thu Jan 8 22:43:49 2009 +0000 ufs: don't truncate longer ufs2 fast symlinks (note that one of these at least is a cross-FS implementation error). From jjllmmss at googlemail.com Sun Mar 7 19:57:35 2010 From: jjllmmss at googlemail.com (JLMS) Date: Sun, 7 Mar 2010 19:57:35 +0000 Subject: [Gllug] Automatic mail archiving In-Reply-To: <1267876925.26839.17.camel@millaralpt.co.tradefair> References: <1267876925.26839.17.camel@millaralpt.co.tradefair> Message-ID: On Sat, Mar 6, 2010 at 12:02 PM, Andy Millar wrote: > On Sat, 2010-03-06 at 11:20 +0000, Martin A. Brooks wrote: >> If you're on a lot of mailing lists, you might find this handy: >> http://bit.ly/axamTe > > That's quite a nice idea. > > I do exactly the same thing, manually, when I've read/dealt with an > email, that way my Inbox really is for things that still need dealing > with. > > I *hate* the Gmail idea of everything being in your Inbox (that said, > I've never actually used Gmail) and people who use their Deleted Items > folder as an archive for their old mail (and they really don't see why > this is a bad idea). > > That said, I actually organise mine by month instead of year. > > Andy How can you hate something you have not used? - Not everything has to be in your Inbox if you don't want to. Inbox is just a label of many, and if you so wish all your email can be filtered on arrival with different lables so your "Inbox" is empty. - You actually don't have to delete email, you archive it and label it (or delete the ones you really know you will not need). The real concern with Googlemail is privacy, if they want to they could make your life a complete misery... From gllug at getaroundtoit.co.uk Mon Mar 8 02:14:51 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Mon, 08 Mar 2010 15:14:51 +1300 Subject: [Gllug] [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <20100306112045.GC13486@annexia.org> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <4B902327.70707@GetAroundToIt.co.uk> <20100306112045.GC13486@annexia.org> Message-ID: <4B945D9B.9010704@getaroundtoit.co.uk> On 07/03/10 00:20, Richard Jones wrote: > On Fri, Mar 05, 2010 at 10:16:23AM +1300, David L Neil wrote: >> Compare this with the UK and US governments giving wads of cash to the >> merchant bankers, who instead of passing it on (even to their >> oft-sacrificed retail brethren), simply held on to it and used it as >> fuel for the bonuses they are paying themselves today. > > The US also distributed cash to the population, in the form of "rebate > checks" (cheques) in the $300-$600 range to most taxpayers. It's not > really clear how much effect that had. Wouldn't sensible people have > mainly held on to this money because of their uncertainly over future > prospects? They would have saved it in the bank, and the banks would > still have had to increase their minimum capital ratios, so would not > have been lending it to anyone. > > Rich. Rich (et al), You are correct about the lack of clarity. It takes time to see these things, and given the size of the economic slide and the duration of the downturn it becomes even more difficult to see something as more than a 'blip', particularly if it was too little. (or in the case of the heralded (US) investment in a "broadband infrastructure", way too late (it still hasn't really started, and certainly not involved expenditure at the $scale touted). The bottom line, as one would say, is that the people who needed it, either saved it and eked-out its benefit (hard to see statistically, ie from aggregated data) or perhaps foolishly blew it all in one go (a visible 'blip' if enough people did that). My earlier observation about Aussie youngsters enjoying the Queenstown (New Zealand, South Island) ski season was quickly observed anecdotally through sheer numbers, and then as the formal stats came in the year-on-year comparison of both visitor numbers and apparent spending per person revealed the benefit (in this case, to New Zealand rather than Australia). The logic of holding-onto the money is an assumption. If someone has 'enough' then yes I'd agree. However a second person thrown out of work and with few resources would probably employ the funds (hopefully on necessities) fairly quickly - and by the same token those with less personal responsibility (or experience of economic lean-times?) would probably contribute to an immediate 'blip' (apropos the Australian example, above). The 'immediate' is what the politician/economists want (see, understand, ... [choose your own word here]). The "sensible" course is not! What? The thing about hunkering-down and saving today's 'ciggie and beer' money in case you need it for tomorrow's 'bread', is that this takes money "out of circulation" - not quite true if you bank it - unless the bank also holds onto funds!!! (see also, below) What was (?is) needed, is to have money being spent on 'stuff': if you buy something from me, I will tend to take that cash and spend it with someone else, and he will ... This is the "money-multiplier effect" illustrated: the same 'wad of notes' passes through three (or likely more) hands and transactions (the latter important when one takes tax revenues and ultimately, government spending, into account). If you (sensibly as an individual, but maybe not in the community view) hold onto that cash, I end up 'short', as does the person I would have paid it to... My immediate reaction is to say 'business is bad' and thus I 'hunker-down' too - and this view 'ripples' to our mythical third-man, etc, etc. Because we have (all) stopped spending, there's 'no' sales and thus no point in manufacturing (or writing software, except for fun/employment of the unemployed mind - ok, given the audience: "scratching an itch"). Thus we have recession and/or depression [no, for those of you who know, I'm not going there!] Thus if by giving you/us all some $ or £, the government persuades you to keep spending, money will continue to flow around the economy and your perception of 'impending doom' will be forestalled. Of course the (fervent) hope is that this delay in perception will be just long enough to ensure that the initial 'shock' is minimised and the natural optimism of the community (?wisdom of crowds) will carry the day. (see also phone deals which give you free minutes during the first month in the hope that you will develop and maintain expensive habits thereafter!) The other assumption is that the money 'gift' is made quickly enough so that pessimism and negativism don't set-in first. Politicians move quickly??? Having avoided going back there for some years, I can't speak to the American situation beyond what I read (and not just 'the newspapers'!) but the choice of approach utilised down-under at least contributed to Australia's extremely confident position when compared to any/all of the western countries. Unfortunately the UK government put their money into the hands of 'bankers' - and when I say this, I mean the merchant and investment bankers, the very people who caused the crisis: (a) by their 'complex swaps'* and (b) by pushing their seen-to-be-junior retail arms, ie the 'high street banks', into offering ever more risky mortgage amounts and contracts (as mentioned elsewhere in this thread), which in turn were revealed as houses-of-cards when liquidity dried-up and (house) prices fell (negative-equity). * which were a means of describing 'paper' as "assets", redefining the same paper elsewhere as a risk (always were "liabilities" under any view of accounting), and by slicing-and-dicing, re-selling the same 'items' over and over and in slightly different ways, for ever-increasing sums (pocketing a fee for their 'expertise' at every transaction) - despite there being no value-added if indeed there was ever any real value (a liability = value???) from the start. Those bankers did exactly as you suggested a (conservative and sensible) individual would do. They figured that if they held onto a war chest they would be able to buy-out and otherwise bludgeon their (financial) competitors, or at the very least, that they would have some assets on the books with which to counter the enormous paper-losses they had created earlier (as above). Most didn't even share their largess with their retail arms... Thus, the great British public only saw 'relief' in the form of lower interest rates (on their credit debts - but effectively losses and disincentives to save, also mentioned elsewhere) - and ONLY saw that IF they held a mortgage!!! I lived in Shepherd's Bush which displays the complete socio-economic range. I can tell you that when a 'factory' shut down (lack of sales, everyone holding their cash...) for every four or five executives worrying about their mortgages (whilst 'enjoying' this relief), there were hundreds or even thousands of folk on the White City council estate who not only lost their income but received no help from these 'economic measures' at all (rent assistance comes from the regular and every-day social security schemes - ie government expenditure which is funded by a decreasing work-force/spending population... So a "double-whammy" to use the economists' term to describe the government's dilemma). A lot of Brits fail(ed) to see an economic downturn (perhaps I moved in the 'wrong circles'?) - certainly a visit to ShepBush's Westfield shopping center showed no shortage of people who had money and who were prepared to spend (even if it also showed shops opening and then closing-down unfortunately rapidly). Maybe a cash handout or rebate would simply have ended-up as coffees, Burberry, and bling? However the folk who were affected, were affected badly, and in large numbers (as evidenced by any economic data you care to quote), despite being relatively 'invisible' to those still working/spending/partying, and despite the huge impact they have had on both the money-multiplier and public funds. The comment about capital ratios disappears with a bit more thought. If the amount of money passing through banks increases, that's no problem. If cash-money enters a bank and is then held/amalgamated to be utilised in a long-term transaction, eg a mortgage, then ratio regulations come into play (necessarily). Most ratios were reduced or relaxed, although they were so low in the UK and US that they completely failed to cope with the very situation for which they were designed (to protect the bank, the mortgagees, the savers, and the national economies) AND the bankers knew this! Thus the example mythical transactions from you, to me, to 'Fred', were simply deposits into cheque accounts followed by withdrawals - no bank ratio (beyond the overnight money market...) in sight. The 'ratio' that is a problem in our 'chain' is not the bank's but the government's. Yes our much beloved VAT/GST! The decision to reduce sales tax was for exactly the reason (you) stated. However, again it benefited those who needed less help by more (in $/£ terms). It probably didn't encourage those 'hunkered down' to spend more, but it did reduce the 'drag' on the amount of spending that was subsequently re-spent (multiplied by more, by de-multiplying by less). Whereas in the UK the government has been forced by its own income needs to re-instate the rate of sales tax, here in New Zealand they did not but discussion is underway to increase sales taxes beyond any historic level. The Opposition has opened-fire in predictable fashion (probably THE point in which they are closets to populist thinking!?). The government's plan is to commensurately (their claim, but without detail) increase social security and pensions (etc) so that the lower socio-economic group(s) will be no-worse off. They claim to seek an "adjustment" which attempts to shift more of the tax burden onto "the rich" without appearing to/actually taxing the income which makes them 'rich'. Yeah right! Another whole discussion or political smoke-and-mirrors re-interpreting of monetary policy... but mentioned to show that this economy is rising again and it is thought, can afford the 'hit' of such a measure - both in economic thinking and individual-perceptions... Regards, =dn From james.dutton at gmail.com Mon Mar 8 11:36:45 2010 From: james.dutton at gmail.com (James Courtier-Dutton) Date: Mon, 8 Mar 2010 11:36:45 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: References: Message-ID: On 3 March 2010 16:26, Stephen Nelson-Smith wrote: > Hi, > > I have a number of Macbooks which contain commercially sensitive > information, and which are sometimes taken off site. > > I've been asked to provide disk-based encryption to protect the data > in the event of loss or theft. > > Have any of you done this before?  I'm currently looking at: > > * http://www.pgp.com/products/wholediskencryption/ > * FileVault > * Knox > > Anything else I should look at?  Any success or war stories to share? > The company I work for have tried: becrypt disk protect (http://www.becrypt.com) pgp whole disk encryption. Of those two, pgp came out preferred as it can me managed centrally better than becrypt. There are central enterprise tools for pgp that lets one do installs on one's entire user base from a central location. For example, they rolled out pgp to 10000 users in my company without the users having to do anything themselves. But, if you do not need too much central management, another one to look at would be Truecrypt and luks. Whole disk encryption is very much preferred over individual folder or virtual disk encryption. Please note that whole disk encryption only provides protection if the PC is stolen while powered off. So, very useful for the "left in a taxi" or "left on a train" scenarios!!! Kind Regards James From sanelson at gmail.com Mon Mar 8 13:39:40 2010 From: sanelson at gmail.com (Stephen Nelson-Smith) Date: Mon, 8 Mar 2010 13:39:40 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: References: Message-ID: Hello, >> * FileVault Seems FileVault creates one massive file in TimeMachine - not so handy for restores. > Whole disk encryption is very much preferred over individual folder or > virtual disk encryption. That appears to be the guidance of Group Security here. > Please note that whole disk encryption only provides protection if the > PC is stolen while powered off. > So, very useful for the "left in a taxi" or "left on a train" scenarios!!! Hrm - whereas file security this applies only if the file is open? S. From jjllmmss at googlemail.com Mon Mar 8 14:05:28 2010 From: jjllmmss at googlemail.com (JLMS) Date: Mon, 8 Mar 2010 14:05:28 +0000 Subject: [Gllug] Topics you would like on a training class. Message-ID: Dear all, I am tinkering with the idea of starting a company to provide training in the Linux-FOSS field. So I have a question for you all, long time veterans and newbies (are there any here?) alike. Which topics would you want to see offered by such a company? I am not thinking about full certification of any flavour of Linux or specific products for now (there are tons of people doing that now), but rather specific topics that one wishes would have been explained by somebody instead of having to jump on your own completely ignorant of how to do things :-) For example, some of my pet peeves: - Setting up a kerberos server and environment. - Setting and using OpenLDAP. - Shell programming (basic, advanced, guru topics :-) ). - Popular (perl, python) and no so popular scripting languages. - Performance analysis. - Security. What else do you think may be of interest? Regards From walter.stanish at saffrondigital.com Mon Mar 8 14:25:41 2010 From: walter.stanish at saffrondigital.com (Walter Stanish) Date: Mon, 8 Mar 2010 14:25:41 -0000 Subject: [Gllug] Topics you would like on a training class. In-Reply-To: References: Message-ID: > Which topics would you want to see offered by such a company? I think larger places want either programming talent which is often (arguably quite mistakenly) linked to immediate requirements (ie: a particular language or framework), or likewise sysadmin talent with experience in a certain platform with immediate demand. As you mentioned existing FOSS certifications seems to take care of some of the above. I would consider (even prefer) a candidate with more general training however the issue with training is that people do not necessarily have the perception that a given course has any content in particular if they have not heard of it or it is not platform-linked. Therefore I would recommend including extensive information on ground covered in any certificates issued, so that potential or existing employers reviewing output can get a really clear idea of which areas have been covered, and to what extent. > For example, some of my pet peeves: > - Setting up a kerberos server and environment. > - Setting and using OpenLDAP. IMHO less likely to win clients. > - Shell programming (basic, advanced, guru topics :-) ). IMHO this is often of lesser (even dubious) utility these days... > - Popular (perl, python) and no so popular scripting languages. IMHO those who would claim recognition of a programming language based on an intro-course rather than a past project or commercial experience are an instant worry - the curiosity and drive to use information freely available to 'skill up' or keep up to date over time represent a real type of personality attribute that many(*) see as a fundamental requirement for good programmers. So errr ... I'd recommend against this as a focus from a pragmatic perspective as a potential employer of course-attendees, though it may very well work out to be a profitable enterprise. (Profit-making and sustainability are often two very separate things!) (* For example some chapters of 'Coders at work' by Peter Siebel) > - Performance analysis. Very interesting area and IMHO one that you could consult in exclusively if you wanted to and have the ability to get the clients in the door. > - Security. Likewise, but the market seems to be maturing for this so you'd better have some good credentials to make a dent on your own. > What else do you think may be of interest? Mostly business-side in areas with recent change... - Groupware - Unified Communications (UM) / Unified Messaging (UM) with Asterisk + XMPP, fax servers, automated email processing, etc. - Automated open source accounting solutions Hope the above is of some use though the normal "take it or leave it", "my views only", "I'm not claiming omniscience", etc. do naturally apply ;) - Walter From bap at shrdlu.com Mon Mar 8 15:13:12 2010 From: bap at shrdlu.com (Bernard Peek) Date: Mon, 08 Mar 2010 15:13:12 +0000 Subject: [Gllug] Topics you would like on a training class. In-Reply-To: References: Message-ID: <4B951408.7070109@shrdlu.com> On 08/03/10 14:05, JLMS wrote: > Dear all, > > I am tinkering with the idea of starting a company to provide training > in the Linux-FOSS field. > > So I have a question for you all, long time veterans and newbies (are > there any here?) alike. > > Which topics would you want to see offered by such a company? > > I am not thinking about full certification of any flavour of Linux or > specific products for now (there are tons of people doing that now), > but rather specific topics that one wishes would have been explained > by somebody instead of having to jump on your own completely ignorant > of how to do things :-) > My situation is that I'm an IT Manager who usually works with Windows systems and users, but I expect to be encountering Linux more in the future. There are some problems that I can see arising and there are some training courses that I can see being very useful. Whether you would want to deliver them is a different question. First and foremost I would want conversion training for users of Microsoft Office to Open Office. It's not very glamorous but it's by far the highest priority and for a training company I would expect it to become the cash-cow. I would want beginner, intermediate and power-user courses possibly including ECDL certification. Power-user training would include macro programming. I would want to switch users to Open Office before I switched them to Linux, so the training should cover both platforms. For the power-users I would want training in Bash. Again, beginner intermediate and advanced. Advanced would include basic Perl. I wouldn't expect to send very many people on to any course that involved use of the command-line. The second most important requirement will be for training in line-of-business applications. For the near future I expect Linux to be mainly restricted to sites where large numbers of people use specific applications for most of their work. Call-centres, that sort of site. I would want a training company that could come in, take a look at software before it is completed and build a training course on it, then deliver it to lots of people. -- Bernard Peek bap at shrdlu.com From social-bounces at gllug.org.uk Tue Mar 9 16:43:04 2010 From: social-bounces at gllug.org.uk (social-bounces at gllug.org.uk) Date: Tue, 09 Mar 2010 16:43:04 +0000 Subject: [Gllug] =?iso-8859-1?q?Forward_of_moderated_message?= Message-ID: An embedded message was scrubbed... From: =?ISO-8859-1?Q?L=E9on_Brocard?= Subject: [ANNOUNCE] London Perl Mongers Technical Meeting 18th March 2010 Date: Mon, 8 Mar 2010 20:04:14 +0000 Size: 4635 Url: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100309/af9802e9/attachment.eml From general_email at technicalbloke.com Wed Mar 10 05:06:00 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Wed, 10 Mar 2010 05:06:00 +0000 Subject: [Gllug] Port forwarding over ssh... Message-ID: <4B9728B8.70801@technicalbloke.com> Hi all, I'm currently testing the FreeNAS box I mentioned I was about to build a couple of weeks ago. It's going quite well but I've got a little stuck on tunneling over SSH. I have ssh setup with certs and have disabled interactive (challenge response) logons. All seems well with that and I can login with no problems but now I'd like to be able to access the NAS's web GUI via an ssh tunnel (tunneling being a new thing to me). The tutorial I read* said to simply SSH in using the -D option to specify a local port that will then act as a local socks proxy so I typed this... ssh -p 12345 -D 9999 username at 192.168.1.123 Sure enough it logged me in and I get a bash prompt. Sadly though when I set firefox's proxy settings to localhost:9999 I can't get anything back from the URL http://localhost/ where, AFAICT, I should get back the NAS's admin login page. I don't know whether this is because the SSH/SOCKS tunnel isn't actually up and running, whether running the above command should actually give me a prompt at all, whether the URL I am entering into firefox isn't right or if the firefox proxy settings are actually correct so I'm not sure how to proceed. I'd appreciate some suggestions as to how to check/rule out some of the above, or a simple correction if I've made a schoolboy error. Many thanks, Roger. *this one is for drupal but I figure it should apply generally... http://www.johnandcailin.com/blog/john/beef-your-drupal-security-apache-modrewrite-and-ssh From farnsaw at stonedoor.com Wed Mar 10 05:15:13 2010 From: farnsaw at stonedoor.com (Andrew Farnsworth) Date: Wed, 10 Mar 2010 00:15:13 -0500 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <4B9728B8.70801@technicalbloke.com> References: <4B9728B8.70801@technicalbloke.com> Message-ID: instead of http://localhost/, do http://192.168.1.123/ Andy On Wed, Mar 10, 2010 at 12:06 AM, general_email at technicalbloke.com < general_email at technicalbloke.com> wrote: > Hi all, > > I'm currently testing the FreeNAS box I mentioned I was about to build a > couple of weeks ago. It's going quite well but I've got a little stuck > on tunneling over SSH. I have ssh setup with certs and have disabled > interactive (challenge response) logons. All seems well with that and I > can login with no problems but now I'd like to be able to access the > NAS's web GUI via an ssh tunnel (tunneling being a new thing to me). > > The tutorial I read* said to simply SSH in using the -D option to > specify a local port that will then act as a local socks proxy so I > typed this... > > ssh -p 12345 -D 9999 username at 192.168.1.123 > > Sure enough it logged me in and I get a bash prompt. Sadly though when I > set firefox's proxy settings to localhost:9999 I can't get anything back > from the URL http://localhost/ where, AFAICT, I should get back the > NAS's admin login page. > > I don't know whether this is because the SSH/SOCKS tunnel isn't actually > up and running, whether running the above command should actually give > me a prompt at all, whether the URL I am entering into firefox isn't > right or if the firefox proxy settings are actually correct so I'm not > sure how to proceed. I'd appreciate some suggestions as to how to > check/rule out some of the above, or a simple correction if I've made a > schoolboy error. > > Many thanks, > > Roger. > > > > *this one is for drupal but I figure it should apply generally... > > http://www.johnandcailin.com/blog/john/beef-your-drupal-security-apache-modrewrite-and-ssh > -- > Gllug mailing list - Gllug at gllug.org.uk > http://lists.gllug.org.uk/mailman/listinfo/gllug > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100310/9d1264ea/attachment.htm From general_email at technicalbloke.com Wed Mar 10 16:29:12 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Wed, 10 Mar 2010 16:29:12 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: References: <4B9728B8.70801@technicalbloke.com> Message-ID: <4B97C8D8.1070802@technicalbloke.com> > On Wed, Mar 10, 2010 at 12:06 AM, general_email at technicalbloke.com < > general_email at technicalbloke.com> wrote: > > >> Hi all, >> >> I'm currently testing the FreeNAS box I mentioned I was about to build a >> couple of weeks ago. It's going quite well but I've got a little stuck >> on tunneling over SSH. I have ssh setup with certs and have disabled >> interactive (challenge response) logons. All seems well with that and I >> can login with no problems but now I'd like to be able to access the >> NAS's web GUI via an ssh tunnel (tunneling being a new thing to me). >> >> The tutorial I read* said to simply SSH in using the -D option to >> specify a local port that will then act as a local socks proxy so I >> typed this... >> >> ssh -p 12345 -D 9999 username at 192.168.1.123 >> >> Sure enough it logged me in and I get a bash prompt. Sadly though when I >> set firefox's proxy settings to localhost:9999 I can't get anything back >> from the URL http://localhost/ where, AFAICT, I should get back the >> NAS's admin login page. >> >> I don't know whether this is because the SSH/SOCKS tunnel isn't actually >> up and running, whether running the above command should actually give >> me a prompt at all, whether the URL I am entering into firefox isn't >> right or if the firefox proxy settings are actually correct so I'm not >> sure how to proceed. I'd appreciate some suggestions as to how to >> check/rule out some of the above, or a simple correction if I've made a >> schoolboy error. >> >> Many thanks, >> >> Roger. >> >> >> >> *this one is for drupal but I figure it should apply generally... >> >> http://www.johnandcailin.com/blog/john/beef-your-drupal-security-apache-modrewrite-and-ssh >> -- >> Gllug mailing list - Gllug at gllug.org.uk >> http://lists.gllug.org.uk/mailman/listinfo/gllug >> >> > > Andrew Farnsworth wrote: > instead of http://localhost/, do http://192.168.1.123/ > > Andy > Yeah I tried that, no joy, thanks anyway though. Interestingly I do get an empty white page instead of an error message :/ Is there a simple way to test if the tunnel is up? Manually send a GET request via telnet maybe?? Roger. From ege.turgay at maglabs.net Wed Mar 10 17:09:21 2010 From: ege.turgay at maglabs.net (Ege Turgay) Date: Wed, 10 Mar 2010 17:09:21 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <4B9728B8.70801@technicalbloke.com> References: <4B9728B8.70801@technicalbloke.com> Message-ID: <1268240961.26403.87.camel@ege.maglabs.net> Hi, this works for me: ssh -f username at remoteip -L 5901:127.0.0.1:5901 -N vncviewer 127.0.0.1 tunnels me to the remote host. should work for http too. so in your case replace the port number with 80. On Wed, 2010-03-10 at 05:06 +0000, general_email at technicalbloke.com wrote: > > ssh -p 12345 -D 9999 username at 192.168.1.123 > > Sure enough it logged me in and I get a bash prompt. Sadly though when I > set firefox's proxy settings to localhost:9999 I can't get anything back > from the URL http://localhost/ where, AFAICT, I should get back the > NAS's admin login page. Best Regards, Ege Turgay -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100310/342aed3c/attachment.htm From ege.turgay at maglabs.net Wed Mar 10 17:19:45 2010 From: ege.turgay at maglabs.net (Ege Turgay) Date: Wed, 10 Mar 2010 17:19:45 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <4B97C8D8.1070802@technicalbloke.com> References: <4B9728B8.70801@technicalbloke.com> <4B97C8D8.1070802@technicalbloke.com> Message-ID: <1268241585.32147.1.camel@ege.maglabs.net> Hi, $ curl -v -0 www.google.com would get you a GET request On Wed, 2010-03-10 at 16:29 +0000, general_email at technicalbloke.com wrote: > Is there a simple > way to test if the tunnel is up? Manually send a GET request via telnet > maybe?? > > Roger. > > Best Regards, Ege Turgay -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100310/bc46e8cd/attachment.htm From general_email at technicalbloke.com Wed Mar 10 18:06:50 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Wed, 10 Mar 2010 18:06:50 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <1268241585.32147.1.camel@ege.maglabs.net> References: <4B9728B8.70801@technicalbloke.com> <4B97C8D8.1070802@technicalbloke.com> <1268241585.32147.1.camel@ege.maglabs.net> Message-ID: <4B97DFBA.8000209@technicalbloke.com> Ege Turgay wrote: > Hi, > > $ curl -v -0 www.google.com > > would get you a GET request > > > > On Wed, 2010-03-10 at 16:29 +0000, general_email at technicalbloke.com > wrote: > > >> Is there a simple >> way to test if the tunnel is up? Manually send a GET request via telnet >> maybe?? >> >> Roger. >> >> >> > > > > Best Regards, > Ege Turgay > > Hi Eve, Thanks for your replies. Actually I've just tried with wget... xxx at xxx:~/tunneling_experiments$ wget http://192.168.1.73/index.php--2010-03-10 18:02:31-- http://192.168.1.73/index.php Connecting to 192.168.1.73:80... connected. HTTP request sent, awaiting response... 302 Found Location: login.php [following] --2010-03-10 18:02:31-- http://192.168.1.73/login.php Connecting to 192.168.1.73:80... connected. HTTP request sent, awaiting response... 200 OK Length: unspecified [text/html] Saving to: `login.php.2' [ <=> ] 1,372 --.-K/s in 0s 2010-03-10 18:02:31 (88.8 MB/s) - `login.php.2' saved [1372] xxx at xxx:~/tunneling_experiments$ http_proxy=127.0.0.1:9999 wget http://192.168.1.73/index.php --2010-03-10 18:02:40-- http://192.168.1.73/index.php Connecting to 127.0.0.1:9999... connected. Proxy request sent, awaiting response... No data received. Retrying. As you can see the local port is open but I get No data received :( I'll try your other suggestion next :) Roger. From general_email at technicalbloke.com Wed Mar 10 18:15:39 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Wed, 10 Mar 2010 18:15:39 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <1268240961.26403.87.camel@ege.maglabs.net> References: <4B9728B8.70801@technicalbloke.com> <1268240961.26403.87.camel@ege.maglabs.net> Message-ID: <4B97E1CB.8070603@technicalbloke.com> Ege Turgay wrote: > Hi, > > this works for me: > > ssh -f username at remoteip -L 5901:127.0.0.1:5901 -N > > vncviewer 127.0.0.1 tunnels me to the remote host. > > should work for http too. > > so in your case replace the port number with 80. > > On Wed, 2010-03-10 at 05:06 +0000, general_email at technicalbloke.com > wrote: > > > >> ssh -p 12345 -D 9999 username at 192.168.1.123 >> >> Sure enough it logged me in and I get a bash prompt. Sadly though when I >> set firefox's proxy settings to localhost:9999 I can't get anything back >> from the URL http://localhost/ where, AFAICT, I should get back the >> NAS's admin login page. >> > > > > > Best Regards, > Ege Turgay > > Ege, you're a star! It turns out I needed -L not -D, thanks a lot for that! And v.sorry for spelling your name wrong in my previous post too! >///< Roger. From joel at fysh.org Wed Mar 10 18:19:49 2010 From: joel at fysh.org (Joel Bernstein) Date: Wed, 10 Mar 2010 18:19:49 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <4B97E1CB.8070603@technicalbloke.com> References: <4B9728B8.70801@technicalbloke.com> <1268240961.26403.87.camel@ege.maglabs.net> <4B97E1CB.8070603@technicalbloke.com> Message-ID: <893ec6981003101019l74ef461byb60f52caaa8fd8b4@mail.gmail.com> On 10 March 2010 18:15, general_email at technicalbloke.com wrote: > Ege, you're a star! It turns out I needed -L not -D, thanks a lot for > that! And v.sorry for spelling your name wrong in my previous post too! I can't imagine why the DynamicForward option didn't work. Can you paste the output of "ssh -vvv -D 9999 you at host"? If the remote sshd permits SOCKS proxying then this should Just Work. Certainly a lot less clunky than pretending your remote machine is actually your local machine with ssh -Llocalport:remotehost:remoteport. It'd be nice to get to the bottom of this anyway! /joel From chrisbell at 3966.ukfsn.org Wed Mar 10 21:26:03 2010 From: chrisbell at 3966.ukfsn.org (Chris Bell) Date: Wed, 10 Mar 2010 21:26:03 +0000 (GMT) Subject: [Gllug] Adding another OpenVPN connection via IPCop Message-ID: Hello, I have an IPCop machine protecting a local network. I have two OpenVPN connections working via Zerina on the IPCop box, one from another fairly local network, another from a laptop travelling in the UK. I now need to provide a connection from a computer in India, but it is not working. I have tried with two remote computers, both running Microsoft Windows XP Pro, and both fail to connect. I have seen a note that the clock setting is important. The computers in India are set for a time zone six and a half hours away from the UK, is this OK? I am running out of ideas on the cause of the problem. -- Chris Bell www.chrisbell.org.uk (was www.overview.demon.co.uk) Microsoft sells you Windows ... Linux gives you the whole house. From mailing-lists at kerneljack.com Wed Mar 10 20:46:31 2010 From: mailing-lists at kerneljack.com (Khusro Jaleel) Date: Wed, 10 Mar 2010 20:46:31 +0000 Subject: [Gllug] Topics you would like on a training class. In-Reply-To: References: Message-ID: <6DB3CF96-421C-4DE5-9079-4D0B5A38A9AB@kerneljack.com> On 8 Mar 2010, at 14:05, JLMS wrote: > Which topics would you want to see offered by such a company? > > - Setting up a kerberos server and environment. > > - Setting and using OpenLDAP. > > - Shell programming (basic, advanced, guru topics :-) ). > > - Popular (perl, python) and no so popular scripting languages. > > - Performance analysis. > > - Security. > I would go for all of the above, plus the following. I think that businesses will consider all of them a plus, except perhaps the last one (SELinux): - Setting up a high availability cluster (with or without DRBD included) - Creating your own RPMS + a repository to host them internally for your own business - Setting up and maintaining Puppet (centralised configuration management for your business) - Setting up a virtualised infrastructure (Xen/KVM) and how to properly secure it, maintain it, and back it up - Setting up Nagios (monitoring your infrastructure) - SELinux perhaps? It confuses a lot of people so they just turn it off From phil-gllug at tinsleyviaduct.com Thu Mar 11 00:21:15 2010 From: phil-gllug at tinsleyviaduct.com (Phil Reynolds) Date: Thu, 11 Mar 2010 00:21:15 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> Message-ID: <20100311002115.17901wg7w5smgigr@topdeck.tinsleyviaduct.com> Quoting "Ciarán Mooney" : > Hi, > > Friendly reminder of this months meeting, details about the talk can > be found here: > > http://lists.gllug.org.uk/pipermail/gllug/2010-February/082543.html > > Seán Tohill has asked that I keep track of the number of attendees. > Currently 15 people have said that they will be attending. Please > email the list if you plan to attend. Unfortunately, I cannot now attend - I have a crisis to attend to in the Midlands. -- Phil Reynolds mail: phil-gllug at tinsleyviaduct.com Web: http://www.tinsleyviaduct.com/phil/ Waltham 66, Emley Moor 69, Droitwich 79, Windows 95 ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From general_email at technicalbloke.com Thu Mar 11 05:27:29 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Thu, 11 Mar 2010 05:27:29 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <893ec6981003101019l74ef461byb60f52caaa8fd8b4@mail.gmail.com> References: <4B9728B8.70801@technicalbloke.com> <1268240961.26403.87.camel@ege.maglabs.net> <4B97E1CB.8070603@technicalbloke.com> <893ec6981003101019l74ef461byb60f52caaa8fd8b4@mail.gmail.com> Message-ID: <4B987F41.4020909@technicalbloke.com> Joel Bernstein wrote: > On 10 March 2010 18:15, general_email at technicalbloke.com > wrote: > >> Ege, you're a star! It turns out I needed -L not -D, thanks a lot for >> that! And v.sorry for spelling your name wrong in my previous post too! >> > > I can't imagine why the DynamicForward option didn't work. Can you > paste the output of "ssh -vvv -D 9999 you at host"? If the remote sshd > permits SOCKS proxying then this should Just Work. Certainly a lot > less clunky than pretending your remote machine is actually your local > machine with ssh -Llocalport:remotehost:remoteport. > > It'd be nice to get to the bottom of this anyway! > > /joel > Cheers Joel, I can paste it but it would be a couple of sides of A4 and -L's working out well enough for what I need so it's probably not worth it. Unless it's going to cause (or it indicates) a security problem I don't really have time to dwell on it, gotta install the bugger at the weekend! Thanks for offering to take a look though :) Regards, Roger. From phil at tinsleyviaduct.com Wed Mar 10 20:18:31 2010 From: phil at tinsleyviaduct.com (Phil Reynolds) Date: Wed, 10 Mar 2010 20:18:31 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> Message-ID: <20100310201831.12201kf9tuni7grb@topdeck.tinsleyviaduct.com> Quoting "Ciarán Mooney" : > Hi, > > Friendly reminder of this months meeting, details about the talk can > be found here: > > http://lists.gllug.org.uk/pipermail/gllug/2010-February/082543.html > > Seán Tohill has asked that I keep track of the number of attendees. > Currently 15 people have said that they will be attending. Please > email the list if you plan to attend. Unfortunately I can no longer attend - a crisis in the Midlands needs my attention. -- Phil Reynolds mail: phil at tinsleyviaduct.com Web: http://www.tinsleyviaduct.com/phil/ Waltham 66, Emley Moor 69, Droitwich 79, Windows 95 ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program. From rich at annexia.org Thu Mar 11 10:58:30 2010 From: rich at annexia.org (Richard Jones) Date: Thu, 11 Mar 2010 10:58:30 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin Message-ID: <20100311105830.GA26352@annexia.org> My internet connection wasn't working this morning, and the problem seemed to be with the line itself. The router reports several mysterious numbers -- I wonder if anyone can help to decode them? When it wasn't working: ADSL Link Downstream Upstream Connection Speed 3250 kbps 915 kbps Line Attenuation 49 db 13.5 db Noise Margin 9 db 5 db Now that it is working: ADSL Link Downstream Upstream Connection Speed 3277 kbps 919 kbps Line Attenuation 48 db 13.5 db Noise Margin 9 db 5 db Reading around the issue, it seems like the tiny difference in line attenuation (49 db vs 48 db) may be significant for ADSL 2+? In general, are these numbers good, bad or indifferent? Rich. -- Richard Jones Red Hat From jason at ukfsn.org Thu Mar 11 11:02:14 2010 From: jason at ukfsn.org (Jason Clifford) Date: Thu, 11 Mar 2010 11:02:14 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: <20100311105830.GA26352@annexia.org> References: <20100311105830.GA26352@annexia.org> Message-ID: <1268305334.2482.72.camel@L1> On Thu, 2010-03-11 at 10:58 +0000, Richard Jones wrote: > Reading around the issue, it seems like the tiny difference in line > attenuation (49 db vs 48 db) may be significant for ADSL 2+? They are not significant. That you had line sync as shown when it wasn't working shows that the problem was nothing to do with attenuation or noise margin. Is your connection a C&W LLU one? If so the reason it was down was a major service outage in C&W's systems. From rich at annexia.org Thu Mar 11 11:04:38 2010 From: rich at annexia.org (Richard Jones) Date: Thu, 11 Mar 2010 11:04:38 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: <1268305334.2482.72.camel@L1> References: <20100311105830.GA26352@annexia.org> <1268305334.2482.72.camel@L1> Message-ID: <20100311110438.GB26352@annexia.org> On Thu, Mar 11, 2010 at 11:02:14AM +0000, Jason Clifford wrote: > On Thu, 2010-03-11 at 10:58 +0000, Richard Jones wrote: > > Reading around the issue, it seems like the tiny difference in line > > attenuation (49 db vs 48 db) may be significant for ADSL 2+? > > They are not significant. > > That you had line sync as shown when it wasn't working shows that the > problem was nothing to do with attenuation or noise margin. > > Is your connection a C&W LLU one? If so the reason it was down was a > major service outage in C&W's systems. murphx.com isn't it? I am kind of interested in the numbers, what they mean, and how to improve them anyhow :-) Rich. -- Richard Jones Red Hat From tethys at gmail.com Thu Mar 11 11:07:04 2010 From: tethys at gmail.com (- Tethys) Date: Thu, 11 Mar 2010 11:07:04 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: <1268305334.2482.72.camel@L1> References: <20100311105830.GA26352@annexia.org> <1268305334.2482.72.camel@L1> Message-ID: <264d99b01003110307x7994fdc5t9362cac5928fb94a@mail.gmail.com> On Thu, Mar 11, 2010 at 11:02 AM, Jason Clifford wrote: > Is your connection a C&W LLU one? If so the reason it was down was a > major service outage in C&W's systems. Interesting. My ADSL line was down this morning, too. My ISP is claiming a fault on BT's network. I wonder if these two incidents are connected or if it's just coincidence. Tet -- ?It seems intuitively obvious to me, which means that it might be wrong.? -- Chris Torek From tim at seacon.co.uk Thu Mar 11 11:17:27 2010 From: tim at seacon.co.uk (t.clarke) Date: Thu, 11 Mar 2010 11:17:27 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: <264d99b01003110307x7994fdc5t9362cac5928fb94a@mail.gmail.com> References: <264d99b01003110307x7994fdc5t9362cac5928fb94a@mail.gmail.com> <20100311105830.GA26352@annexia.org> <1268305334.2482.72.camel@L1> Message-ID: I don't think there is much you can do about the line attenuation or noise margin numbers! Line attenuation is a presumably a result mainly of the distance you are from the exchange (as the wires run - eg my home is much closer to the exchange than the 'wire distance' due to the wires taking a longer route to get across the local stream!). I believe the noise margin is set my the DSLAM - but am open to correction on this! My brother recently purchased a new master socket with built-in microfilter, specifically designed for 'longer distance' connections and saw an immediate significant improvement over his previous separate microfilter arrangement. So it is quite possible the quality of the microfilters will have an affect on these figures as well! Certainly in my experience microfilters can be a significant factor in determining connection quality. Tim From gllug at acp-online.co.uk Thu Mar 11 11:55:51 2010 From: gllug at acp-online.co.uk (InstincT) Date: Thu, 11 Mar 2010 11:55:51 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: <20100311110438.GB26352@annexia.org> References: <20100311105830.GA26352@annexia.org> <1268305334.2482.72.camel@L1> <20100311110438.GB26352@annexia.org> Message-ID: <4B98DA47.4030504@acp-online.co.uk> On 11/03/2010 11:04, Richard Jones wrote: > > I am kind of interested in the numbers, what they mean, and how > to improve them anyhow :-) > > Rich. > > I think of line attenuation as how 'loud' your modem has to shout to your local exchange to communicate. The louder it has to shout the lower speed you will get. The further away you are the louder it has to shout. Also the quality of the line effects how your 'shout' travels. Technically it is how much signal is lost along the wire. The lower this number the better. I think the 50+db range may be quite high as I normally see 20-35 but not really sure. Signal to Noise Ratio shows how much of your line is being taken up by your adsl signal [your shout] and how much is being taken up by noise [ie part of the line you cannot shout down]. The more signal you have in relation to noise the better. There are a few things that you can do to try and improve the numbers. To start with it's best to see what numbers you get if you plug the router directly into the BT master socket [take the front off and plug into the main socket - this disconnects all internal wiring] This will tell you the best you can hope for. From there you can see what effect your internal wiring is having. _Decent_ microfilters are worth the investment. I like the replcement master socket face plates which handles all filtering for all internal extensions and allows your router to get as close as possible to where your wire comes in: http://www.adslnation.com/products/xte2005.php Cheers Marcus From lesleyb at pgcroft.net Thu Mar 11 13:03:14 2010 From: lesleyb at pgcroft.net (Lesley B) Date: Thu, 11 Mar 2010 13:03:14 +0000 Subject: [Gllug] Port forwarding over ssh... In-Reply-To: <4B987F41.4020909@technicalbloke.com> References: <4B9728B8.70801@technicalbloke.com> <1268240961.26403.87.camel@ege.maglabs.net> <4B97E1CB.8070603@technicalbloke.com> <893ec6981003101019l74ef461byb60f52caaa8fd8b4@mail.gmail.com> <4B987F41.4020909@technicalbloke.com> Message-ID: <20100311130314.GB4203@pgcroft.net> On Thu, Mar 11, 2010 at 05:27:29AM +0000, general_email at technicalbloke.com wrote: > Joel Bernstein wrote: > > On 10 March 2010 18:15, general_email at technicalbloke.com > > wrote: > > > >> Ege, you're a star! It turns out I needed -L not -D, thanks a lot for > >> that! And v.sorry for spelling your name wrong in my previous post too! > >> > > > > I can't imagine why the DynamicForward option didn't work. Can you > > paste the output of "ssh -vvv -D 9999 you at host"? If the remote sshd > > permits SOCKS proxying then this should Just Work. Certainly a lot > > less clunky than pretending your remote machine is actually your local > > machine with ssh -Llocalport:remotehost:remoteport. > > > > It'd be nice to get to the bottom of this anyway! > > > > /joel > > > > > Cheers Joel, > > I can paste it but it would be a couple of sides of A4 and -L's working > out well enough for what I need so it's probably not worth it. Unless > it's going to cause (or it indicates) a security problem I don't really > have time to dwell on it, gotta install the bugger at the weekend! > Thanks for offering to take a look though :) > why not use pastebin? Regards L. From rich at annexia.org Thu Mar 11 21:27:49 2010 From: rich at annexia.org (Richard Jones) Date: Thu, 11 Mar 2010 21:27:49 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: References: <264d99b01003110307x7994fdc5t9362cac5928fb94a@mail.gmail.com> <20100311105830.GA26352@annexia.org> <1268305334.2482.72.camel@L1> Message-ID: <20100311212749.GA25714@annexia.org> On Thu, Mar 11, 2010 at 11:17:27AM +0000, t.clarke wrote: > My brother recently purchased a new master socket with built-in microfilter, > specifically designed for 'longer distance' connections and saw an immediate > significant improvement over his previous separate microfilter arrangement. > So it is quite possible the quality of the microfilters will have an affect > on these figures as well! Certainly in my experience microfilters can be a > significant factor in determining connection quality. Have you got a reference for that? I'm assuming it would be something like the "faceplate splitter" here: http://www.solwise.co.uk/adsl_splitters.htm This house has phone cabling going into just about every room, presumably from the days before cordless phones. It's a bit of a nightmare and I intend to pull most of it out ... Rich. -- Richard Jones Red Hat From tim at seacon.co.uk Thu Mar 11 21:37:07 2010 From: tim at seacon.co.uk (t.clarke) Date: Thu, 11 Mar 2010 21:37:07 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: <20100311212749.GA25714@annexia.org> References: <20100311212749.GA25714@annexia.org> <264d99b01003110307x7994fdc5t9362cac5928fb94a@mail.gmail.com> Message-ID: The 'master socket' my brother bought came from broadbandhelp.co.uk (or something like that) amd he thinks its the adslnation one. It could well be the same one that solwise sell - he checked the other day and the one he bought is now apparently out of stock til April! He also bought a 'high quality modem cable' for two quid - the cable that sits between the wall socket and the router itself; whether or not this contributes anything noticeable or not is a matter of conjecture! I will be getting one for our ADSL connection at the office to try as well - we only get a crap 2Mbps due to distance from the exchange, so any improvement will be worth having. Will report further on that in due course. Tim From cehunter at gb-x.org Fri Mar 12 05:13:56 2010 From: cehunter at gb-x.org (Christopher Hunter) Date: Fri, 12 Mar 2010 05:13:56 +0000 Subject: [Gllug] Advice on ADSL line attenuation and noise margin In-Reply-To: References: <20100311212749.GA25714@annexia.org> <264d99b01003110307x7994fdc5t9362cac5928fb94a@mail.gmail.com> Message-ID: <1268370836.32115.52.camel@localhost> On Thu, 2010-03-11 at 21:37 +0000, t.clarke wrote: > The 'master socket' my brother bought came from broadbandhelp.co.uk (or > something like that) amd he thinks its the adslnation one. > It could well be the same one that solwise sell - he checked the other day > and the one he bought is now apparently out of stock til April! > He also bought a 'high quality modem cable' for two quid - the cable that > sits between the wall socket and the router itself; whether or not this > contributes anything noticeable or not is a matter of conjecture! > > I will be getting one for our ADSL connection at the office to try as well - > we only get a crap 2Mbps due to distance from the exchange, so any improvement > will be worth having. Will report further on that in due course. > > Tim I've found that some pieces of gear connected to a phone line affect the level of the ADSL signal more than others. Everything connected to a phone line is supposed to be "approved" and each item is meant to appear (electrically) as "telephone instrument". Years ago, when I designed answering machines, fax machines, cordless phones and the like, I had to take great pains to get the complex impedance presented to line to "look" just like a 746 telephone! These days, the manufacturers seem less interested in that kind of accuracy, and the approval bodies are much less stringent. The "Sky Box" is a case in point, and attenuates ADSL quite markedly. I looked at this be the simple expedient of connecting my 'scope across the phone line, and going 'round the house connecting and disconnecting all the various bits of telephonic gear to see what effect each had. If you have spare microfilters, it's often worth using two in series with some gear. It sounds peculiar, but the attenuation due to the item is halved, and I found that (for example) connecting two microfilters in series before the Sky receiver meant almost no change in received ADSL level, less reflection (so a better quality ADSL signal) and thereby a significantly faster connection. I also have a Panasonic Digital Cordless Phone (one of my designs from the late 80s) which has no effect on the ADSL level when connected through a microfilter, but a more modern model was almost as bad as the Sky receiver! The ideal is, I suppose, to put the filter at the incoming "master" socket, and have all the telephone equipment fed entirely separately from the feed to the modem. Again, it might prove worthwhile to use two filters - I'm going to try this at the weekend. I have a "Master" socket-type filter which I can easily fit, then run some Cat 5 cable with the Western Electric plugs crimped on to it for the modem, and use the existing house phone wiring for the instruments. I also have a spare microfilter or two, so I'll try adding one after the master socket, and see what happens. I'll let you all know. Chris From danthegeekman at googlemail.com Fri Mar 12 10:59:43 2010 From: danthegeekman at googlemail.com (Dan) Date: Fri, 12 Mar 2010 10:59:43 +0000 Subject: [Gllug] [OT] Disk Encryption In-Reply-To: <20100306121648.GF13486@annexia.org> References: <732076a81003040501t2e78ebb6of31a7b99d7aba34b@mail.gmail.com> <20100304180517.GA4136@basis.uklinux.net> <732076a81003041007n768887f4o79260357a414c8c4@mail.gmail.com> <20100304182756.GA4253@basis.uklinux.net> <20100304183609.GV28166@cornerstonelinux.co.uk> <20100306121648.GF13486@annexia.org> Message-ID: <4e73d5ed1003120259uc447bf8w59e85fd3a6f21710@mail.gmail.com> On Sat, Mar 6, 2010 at 12:16 PM, Richard Jones wrote: > Finally, if you want to hide personal / sensitive / illegal > information from authorities, then you would want a two-level system, > where you use another layer of encryption that cannot be detected or > you can plausibly deny knowledge of. TrueCrypt can allegedly do this > (it's called a "hidden volume" -- I've never tried). LUKS can't do > this, except by just obfuscating the file so it will still be quite > obvious to anyone looking that there is some encrypted data they > cannot read. There's a BZ open for a long time for supporting hidden > volumes in Fedora: https://bugzilla.redhat.com/show_bug.cgi?id=454855 > > Rich. > Yup, hidden volume's is the way to go for anything dodgy. But again traces can be left in the OS. This will show someone who really know what there doing that something odd is going on. The best thing is a hidden OS ( http://www.truecrypt.org/docs/hidden-operating-system). Truecrypt now supports this for windows desktops. You will loose a significant amount of harddisk space. 50% of disk space in fact. That's where the 'decoy' OS will live. But once you've got the system up and running you can either run your other OS's in VM or install something like WUBI to give you a Linux install. Again no help for Mac's I'm afraid. Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100312/18f6ba7d/attachment.htm From danthegeekman at googlemail.com Fri Mar 12 11:17:59 2010 From: danthegeekman at googlemail.com (Dan) Date: Fri, 12 Mar 2010 11:17:59 +0000 Subject: [Gllug] [OT] Re: [Long] Down-under response. Was: File On 4 on waste in Govt computing projects In-Reply-To: <20100306103756.GB13486@annexia.org> References: <20100302215244.GA12176@annexia.org> <4B8DA380.7010808@GetAroundToIt.co.uk> <1267603813.10915.73.camel@localhost> <87635df3b9.fsf@spindle.srvr.nix> <1267662462.1783.114.camel@localhost> <87wrxsedkx.fsf@spindle.srvr.nix> <20100306103756.GB13486@annexia.org> Message-ID: <4e73d5ed1003120317l1c6090f6w8005b434f6b1049b@mail.gmail.com> On Sat, Mar 6, 2010 at 10:37 AM, Richard Jones wrote: > Yes. It's one reason why my parent's generation all live in much > bigger houses than I will ever do. Mortgages they took out in the > early 70s inflated away. Look at the bulge in this list of figures: > > http://safalra.com/other/historical-uk-inflation-price-conversion/ > > Rich. > Great link Rich, I shall make a note of that one for all time. The 'great slump' is an interesting period recorded well on your link. http://en.wikipedia.org/wiki/Great_Depression_in_the_United_Kingdom Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100312/9fe93ba1/attachment.htm From gllug at si-designs.co.uk Fri Mar 12 12:02:41 2010 From: gllug at si-designs.co.uk (Simon Perry) Date: Fri, 12 Mar 2010 12:02:41 +0000 Subject: [Gllug] Adding another OpenVPN connection via IPCop In-Reply-To: References: Message-ID: <4B9A2D61.9000805@si-designs.co.uk> Chris Bell wrote: > Hello, > I have an IPCop machine protecting a local network. I have two OpenVPN > connections working via Zerina on the IPCop box, one from another fairly > local network, another from a laptop travelling in the UK. I now need to > provide a connection from a computer in India, but it is not working. I have > tried with two remote computers, both running Microsoft Windows XP Pro, and > both fail to connect. I have seen a note that the clock setting is > important. The computers in India are set for a time zone six and a half > hours away from the UK, is this OK? I am running out of ideas on the cause > of the problem. > > What do the OpenVpn logs on the ipcop box say? Do they even register the connection attempt? If not it could be firewall or ISP policy that is preventing connections on the required ports for the Indian computers. Hth Simon From k.kudlacek at gmail.com Fri Mar 12 12:04:33 2010 From: k.kudlacek at gmail.com (Karel Kudlacek) Date: Fri, 12 Mar 2010 13:04:33 +0100 Subject: [Gllug] automounting volumes in gnome Message-ID: <1268395473.2321.0@debian> Hi, could anybody please advise how I can change the default automount directory in gnome from /media to /mnt Is there some simple way to do it, without the need to install additional packages (ie gnome-volume-manager) thanks, karel From pawelfita at yahoo.co.uk Fri Mar 12 12:11:30 2010 From: pawelfita at yahoo.co.uk (Pawel Fita) Date: Fri, 12 Mar 2010 04:11:30 -0800 (PST) Subject: [Gllug] automounting volumes in gnome In-Reply-To: <1268395473.2321.0@debian> Message-ID: <916632.31467.qm@web25804.mail.ukl.yahoo.com> Hi Karel, Check that page, I believe that's what you are after: http://felipec.wordpress.com/2006/12/28/automounting-a-storage-device-with-gnome/ Regards, Pawel Fita --- On Fri, 12/3/10, Karel Kudlacek wrote: From: Karel Kudlacek Subject: [Gllug] automounting volumes in gnome To: gllug at gllug.org.uk Date: Friday, 12 March, 2010, 13:04 Hi, could anybody please advise how I can change the default automount directory in gnome from /media to /mnt Is there some simple way to do it, without the need to install additional packages (ie gnome-volume-manager) thanks, karel -- Gllug mailing list  -  Gllug at gllug.org.uk http://lists.gllug.org.uk/mailman/listinfo/gllug -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100312/c42f9744/attachment-0001.htm From sanelson at gmail.com Fri Mar 12 14:58:20 2010 From: sanelson at gmail.com (Stephen Nelson-Smith) Date: Fri, 12 Mar 2010 14:58:20 +0000 Subject: [Gllug] [JOB] Site Reliability Engineer - Capita National Strategies Message-ID: Hello, National Strategies is looking for a talented person with Linux Administration and Python programming skills to join our team to support us in our drive to automate the whole infrastructure, set the standard for deployment and continuous integration, build a world-class monitoring and orchestration environment, and use creativity and intelligence to bring about continuous improvement in site speed and resilience. The project is moving from a managed hosting environment to a colocation environment, which brings with it the opportunity to bring the entire current site, and all of the new site under puppet control. The role is based around a 5 day week, in the project team building in the centre of Reading, 7 minutes walk from the railway station (which is 35 mins from Paddington). You will be reporting into the Infrastructure Lead, working in a small, self-managing team. Daily work is carried out from a ticket-driven Kanban system. Activities are typically broken into project and support work, support being for the current website and the new website, and the surrounding project team. About National Strategies National Strategies Online (NSO) is a hugely successful Drupal-based resource for the whole education system - offering a wide range of content to improve standards in the UK school system. The site has in excess of a million visitors a month, and has over 20,000 Drupal nodes. In 2010-2011 the project team will be taking the lessons learned from the current site, and delivering a new site containing some of the best content from the current site, new content, and syndicated content from the leading educational websites in the UK. The technology stack is Redhat Linux, Apache, Drupal, Memcache, Varnish and Solr. A number of custom tools have been developed in Python to provide indexing, queue-based sharing capabilities and general system programming. The whole project team is strongly influenced by Agile methodologies, with Scrum and Kanban implemented throughout. XP-derived practices such as pair programming and test-driven development are being introduced. Desired competencies: * Python programming * Configuration management (preferably Puppet) * Experience with virtualisation technologies (VMware, Xen, KVM, Virtualbox) * Familiarity with and interest in Agile ways of working * Huge capacity and hunger for learning * Solid understanding of the Linux operating system (exposure to both Red Hat and Debian-based systems would be ideal) In return, you will get: * A reliable 6 month contract, with a good chance of a further 6 month extension * Experience on a large government project * The chance to work with a large number of highly skilled programmers, technical architects and sysadmins * A license to innovate and make things awesome Rate negotiations will start at around £300 a day. If you're interested, please contact Stephen Nelson-Smith at stephen.nelson-smith at nationalstrategies.co.uk. Strictly no agencies, under any circumstances at all. S. -- Stephen Nelson-Smith Technical Director Atalanta Systems Ltd www.atalanta-systems.com From dev at archonet.com Fri Mar 12 15:24:21 2010 From: dev at archonet.com (Richard Huxton) Date: Fri, 12 Mar 2010 15:24:21 +0000 Subject: [Gllug] [JOB] Site Reliability Engineer - Capita National Strategies In-Reply-To: References: Message-ID: <4B9A5CA5.6060900@archonet.com> On 12/03/10 14:58, Stephen Nelson-Smith wrote: > In return, you will get: ... > * A license to innovate and make things awesome It would be nice to think this was an actual piece of paper with "By the authority of her most Imperial Majesty Queen Elizabeth II, I hereby grant the bearer..." -- Richard Huxton Archonet Ltd From tim at seacon.co.uk Fri Mar 12 15:29:48 2010 From: tim at seacon.co.uk (t.clarke) Date: Fri, 12 Mar 2010 15:29:48 +0000 Subject: [Gllug] [JOB] Site Reliability Engineer - Capita National In-Reply-To: <4B9A5CA5.6060900@archonet.com> References: <4B9A5CA5.6060900@archonet.com> Message-ID: I thought it was 'Her Britannic Majesty' these days, now we no longer have an empire !! Tim From k.kudlacek at gmail.com Fri Mar 12 16:03:47 2010 From: k.kudlacek at gmail.com (Karel Kudlacek) Date: Fri, 12 Mar 2010 17:03:47 +0100 Subject: [Gllug] automounting volumes in gnome In-Reply-To: <916632.31467.qm@web25804.mail.ukl.yahoo.com> (from pawelfita@yahoo.co.uk on Fri Mar 12 13:11:30 2010) Message-ID: <1268409827.2321.1@debian> Hi Pawel, thanks for the link. It seems to be an instruction how to mount a specific device to a specific location. But I would prefer to have some general rule. I would like all removable devices to be mounted in /mnt instead of /media. Would there be some simpler way to do it ? Karel > Hi Karel, > > Check that page, I believe that's what you are after: > http://felipec.wordpress.com/2006/12/28/automounting-a-storage- > device-with-gnome/ > > > Regards, > > Pawel Fita > > --- On Fri, 12/3/10, Karel Kudlacek wrote: > > From: Karel Kudlacek > Subject: [Gllug] automounting volumes in gnome > To: gllug at gllug.org.uk > Date: Friday, 12 March, 2010, 13:04 > > Hi, > could anybody please advise how I can change the default automount > directory in gnome from /media to /mnt > > Is there some simple way to do it, without the need to install > additional packages (ie gnome-volume-manager) > > thanks, > karel > > -- > Gllug mailing list  -  Gllug at gllug.org.uk > http://lists.gllug.org.uk/mailman/listinfo/gllug > > > > ------quoted attachment------ > -- > Gllug mailing list - Gllug at gllug.org.uk > http://lists.gllug.org.uk/mailman/listinfo/gllug > From benjamin at py-soft.co.uk Fri Mar 12 16:17:21 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Fri, 12 Mar 2010 16:17:21 +0000 Subject: [Gllug] automounting volumes in gnome In-Reply-To: <1268409827.2321.1@debian> References: <1268409827.2321.1@debian> Message-ID: <3251583748312749582@unknownmsgid> On 12 Mar 2010, at 16:03, Karel Kudlacek wrote: > I would like all removable devices to be mounted in /mnt instead > of /media. Would there be some simpler way to do it ? Why not just make /media a hard link to /mnt ? Ben From k.kudlacek at gmail.com Fri Mar 12 17:14:07 2010 From: k.kudlacek at gmail.com (Karel Kudlacek) Date: Fri, 12 Mar 2010 18:14:07 +0100 Subject: [Gllug] automounting volumes in gnome In-Reply-To: <3251583748312749582@unknownmsgid> (from benjamin@py-soft.co.uk on Fri Mar 12 17:17:21 2010) References: <1268409827.2321.1@debian> <3251583748312749582@unknownmsgid> Message-ID: <1268414047.2321.2@debian> :-) that's clever, I'll do that many thanks, Karel > On 12 Mar 2010, at 16:03, Karel Kudlacek > wrote: > > > I would like all removable devices to be mounted in /mnt instead > > of /media. Would there be some simpler way to do it ? > > Why not just make /media a hard link to /mnt ? > > Ben > -- > Gllug mailing list - Gllug at gllug.org.uk > http://lists.gllug.org.uk/mailman/listinfo/gllug > From john at cornerstonelinux.co.uk Fri Mar 12 17:31:43 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Fri, 12 Mar 2010 17:31:43 +0000 Subject: [Gllug] automounting volumes in gnome In-Reply-To: <3251583748312749582@unknownmsgid> References: <1268409827.2321.1@debian> <3251583748312749582@unknownmsgid> Message-ID: <20100312173143.GA4831@cornerstonelinux.co.uk> On Fri, Mar 12, 2010 at 04:17:21PM +0000, Benjamin Donnachie wrote: > On 12 Mar 2010, at 16:03, Karel Kudlacek wrote: > >> I would like all removable devices to be mounted in /mnt instead >> of /media. Would there be some simpler way to do it ? > > Why not just make /media a hard link to /mnt ? I guess you mean a symbolic link. You can't hard link directories. As /media is more controlled by the OS and more likely to be changed without your knowledge, I think a better way may be to make /mnt the symlink to /media. Personally, I like them separate. OS controlled mounts can go under /media and manual mounts go under /mnt. That way /mnt mounts don't get suddenly changed without my knowledge. -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100312/1cc4350d/attachment.pgp From benjamin at py-soft.co.uk Fri Mar 12 17:52:18 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Fri, 12 Mar 2010 17:52:18 +0000 Subject: [Gllug] automounting volumes in gnome In-Reply-To: <20100312173143.GA4831@cornerstonelinux.co.uk> References: <1268409827.2321.1@debian> <3251583748312749582@unknownmsgid> <20100312173143.GA4831@cornerstonelinux.co.uk> Message-ID: <732076a81003120952o21b7a0d9p74d5deb2fcdc7f59@mail.gmail.com> On 12 March 2010 17:31, John Edwards wrote: > I guess you mean a symbolic link. You can't hard link directoriesen Fair point - stuck in front of Windows servers all day and no opportunity to check. > Personally, I like them separate. Me too, but OP obviously feels differently. Take care, Ben From nix at esperi.org.uk Fri Mar 12 21:06:48 2010 From: nix at esperi.org.uk (Nix) Date: Fri, 12 Mar 2010 21:06:48 +0000 Subject: [Gllug] [JOB] Site Reliability Engineer - Capita National Strategies In-Reply-To: <4B9A5CA5.6060900@archonet.com> (Richard Huxton's message of "Fri, 12 Mar 2010 15:24:21 +0000") References: <4B9A5CA5.6060900@archonet.com> Message-ID: <873a05tfkn.fsf@spindle.srvr.nix> On 12 Mar 2010, Richard Huxton uttered the following: > On 12/03/10 14:58, Stephen Nelson-Smith wrote: >> In return, you will get: > ... >> * A license to innovate and make things awesome > > It would be nice to think this was an actual piece of paper with "By the > authority of her most Imperial Majesty Queen Elizabeth II, I hereby > grant the bearer..." Does that mean if I don't have one of these, I can't think up anything new and can only make things less impressive? That's disturbing. From bap at shrdlu.com Fri Mar 12 22:33:31 2010 From: bap at shrdlu.com (Bernard Peek) Date: Fri, 12 Mar 2010 22:33:31 +0000 Subject: [Gllug] [JOB] Site Reliability Engineer - Capita National Strategies In-Reply-To: <873a05tfkn.fsf@spindle.srvr.nix> References: <4B9A5CA5.6060900@archonet.com> <873a05tfkn.fsf@spindle.srvr.nix> Message-ID: <4B9AC13B.9030100@shrdlu.com> On 12/03/10 21:06, Nix wrote: > On 12 Mar 2010, Richard Huxton uttered the following: > > >> On 12/03/10 14:58, Stephen Nelson-Smith wrote: >> >>> In return, you will get: >>> >> ... >> >>> * A license to innovate and make things awesome >>> >> It would be nice to think this was an actual piece of paper with "By the >> authority of her most Imperial Majesty Queen Elizabeth II, I hereby >> grant the bearer..." >> > Does that mean if I don't have one of these, I can't think up anything > new and can only make things less impressive? > > That's disturbing. > Don't worry. For a small fee I can make the license retroactive. -- Bernard Peek bap at shrdlu.com From john at sinodun.org.uk Sun Mar 14 07:53:58 2010 From: john at sinodun.org.uk (John Winters) Date: Sun, 14 Mar 2010 07:53:58 +0000 Subject: [Gllug] [JOB] Site Reliability Engineer - Capita National Strategies In-Reply-To: <4B9A5CA5.6060900@archonet.com> References: <4B9A5CA5.6060900@archonet.com> Message-ID: <4B9C9616.8020202@sinodun.org.uk> Richard Huxton wrote: > On 12/03/10 14:58, Stephen Nelson-Smith wrote: >> In return, you will get: > ... >> * A license to innovate and make things awesome > > It would be nice to think this was an actual piece of paper with "By the > authority of her most Imperial Majesty Queen Elizabeth II, I hereby > grant the bearer..." But there again, I expect Brenda can spell "licence". John From general.mooney at googlemail.com Mon Mar 15 09:19:18 2010 From: general.mooney at googlemail.com (=?UTF-8?Q?Ciar=C3=A1n_Mooney?=) Date: Mon, 15 Mar 2010 09:19:18 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 18th March - Reminder In-Reply-To: <20100310201831.12201kf9tuni7grb@topdeck.tinsleyviaduct.com> References: <3e4e51a81003050108je9b995bw111f01ae421c80fa@mail.gmail.com> <20100310201831.12201kf9tuni7grb@topdeck.tinsleyviaduct.com> Message-ID: <3e4e51a81003150219t154fbfe4if0fab4fbe6790e11@mail.gmail.com> Hi, Just another friendly reminder that the talk by Richard Jones is this Thursday (18th) at the University of Westminster at 1930. Regards, Ciarán From richardlewis at fastmail.co.uk Mon Mar 15 12:34:33 2010 From: richardlewis at fastmail.co.uk (Richard Lewis) Date: Mon, 15 Mar 2010 12:34:33 +0000 Subject: [Gllug] GDM Xclients script Message-ID: <87hbohepba.wl%richard.lewis@gold.ac.uk> Hi GLLUG, I updated my Debian unstable for the first time in a few weeks last night. I decided I'd better switch off rather than just hibernating afterwards. Now I can't seem to log in using the "Run Xclients script" option from GDM properly. When I select that option, it starts full GNOME sessions. It seems that when I run /etc/X11/Xsession it starts GNOME, when I'd expect it to run my ~/.xsession script as it used to. I've implemented a temporary hack with a /usr/share/xsessions/foo.desktop which looks like this: [Desktop Entry] Encoding=UTF-8 Name=Xsession Comment=Run my .xsession script Exec=/home/richard/.xsession Terminal=False Type=Application [Window Manager] SessionManaged=true However, this isn't really desirable. And also I'd like to know why GDM isn't doing what it used to, rather than just apply pasters. Any thoughts? Anyone seen anything like this before? -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Richard Lewis ISMS, Computing Goldsmiths, University of London Tel: +44 (0)20 7078 5134 Skype: richardjlewis JID: ironchicken at jabber.earth.li http://www.richard-lewis.me.uk/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- +----------------------------------------------+ | Support open access to scholarship | | http://freeculture.org/ http://www.doaj.org/ | +----------------------------------------------+ From sanelson at gmail.com Mon Mar 15 13:14:24 2010 From: sanelson at gmail.com (Stephen Nelson-Smith) Date: Mon, 15 Mar 2010 13:14:24 +0000 Subject: [Gllug] [JOB] Site Reliability Engineer - Capita National Strategies In-Reply-To: <4B9C9616.8020202@sinodun.org.uk> References: <4B9A5CA5.6060900@archonet.com> <4B9C9616.8020202@sinodun.org.uk> Message-ID: On Sun, Mar 14, 2010 at 7:53 AM, John Winters wrote: > Richard Huxton wrote: >> On 12/03/10 14:58, Stephen Nelson-Smith wrote: >>> In return, you will get: >> ... >>> * A license to innovate and make things awesome >> >> It would be nice to think this was an actual piece of paper with "By the >> authority of her most Imperial Majesty Queen Elizabeth II, I hereby >> grant the bearer..." > > But there again, I expect Brenda can spell "licence". Mea culpa and good spot. Sorry! S. From itsbruce at workshy.org Mon Mar 15 13:27:57 2010 From: itsbruce at workshy.org (Bruce Richardson) Date: Mon, 15 Mar 2010 13:27:57 +0000 Subject: [Gllug] GDM Xclients script In-Reply-To: <87hbohepba.wl%richard.lewis@gold.ac.uk> References: <87hbohepba.wl%richard.lewis@gold.ac.uk> Message-ID: <20100315132757.GB9234@phaistos.bruce> On Mon, Mar 15, 2010 at 12:34:33PM +0000, Richard wrote: > > Any thoughts? Anyone seen anything like this before? Have you had a look at /etc/X11/Xsession to see what it does before it runs the user session script? I don't have an unstable installation to hand, but the typical behaviour is to run any scripts in /etc/X11/Xsession.d/ first. Could be something in there that is preempting your user script. -- Bruce What would Edward Woodward do? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100315/5b738028/attachment.pgp From itsbruce at workshy.org Mon Mar 15 13:49:26 2010 From: itsbruce at workshy.org (Bruce Richardson) Date: Mon, 15 Mar 2010 13:49:26 +0000 Subject: [Gllug] Partition a 2 TB drive for storing films In-Reply-To: <9f8092cc1003061307p2ec83508o74befbb39db60ca0@mail.gmail.com> References: <4B929882.9030608@sinodun.org.uk> <9f8092cc1003061307p2ec83508o74befbb39db60ca0@mail.gmail.com> Message-ID: <20100315134926.GC9234@phaistos.bruce> On Sat, Mar 06, 2010 at 09:07:10PM +0000, John wrote: > I would use XFS for a large filesystem. I would tend to back John on this but would add the caveat that it depends what you mean by "large". XFS is good for dealing with large files, much faster than ext3 for fscking large filesystems, less painful than ext3 (*much* less painful than jfs) for iterating over filesystems with a large number of nested directories. Oh, and it's good (and quick) at online resizing, which may not be relevant to you. On the negative side, it is slow at deleting files and even slower at deleting directories with many files in or nested directory trees. I have also known it to be unstable where it was installed on top of multiple VFS layers (e.g. XFS on LVM on software RAID); that said, I think the usual culprit for that is write barriers, which you can disable and which XFS on Linux now routinely disables if it thinks the underlying device won't respect them. I do like XFS and use it extensively, just to be clear. -- Bruce It is impolite to tell a man who is carrying you on his shoulders that his head smells. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100315/61d2580f/attachment.pgp From gllug at getaroundtoit.co.uk Tue Mar 16 03:37:16 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Tue, 16 Mar 2010 16:37:16 +1300 Subject: [Gllug] ACTA coming to New Zealand Message-ID: <4B9EFCEC.7030908@getaroundtoit.co.uk> For those who are interested in the under-the-covers shenanigans of the 'IP industry': Greetings from one of the 'greater' parts of Greater London! New Zealanders are not great fans of American imperialism and seem to have found themselves in the fore-front of this issue, not least because of the government's insistence on concepts such as 'innocent until proven guilty'... "NZ Doing Good in ACTA Negotiation March 1, 2010 ? 1:15 pm The Anti-Counterfeiting Trade Agreement (ACTA) is an agreement between countries around IP rights and enforcement. The negotiations have been happening in secret, with every country saying "well, we'd love to reveal what we're talking about but those other countries just won't let us". Fortunately there have been leaks, and the latest is a fascinating glimpse at how these things are put together and where the parties stand. It seems bizarre at first, but the draft is laid out like a spreadsheet: one article per row and with three columns, one each for the US/Japan version, the EU version, and comments. Inside each sentence square brackets mark the attributed proposed alternatives for language. From this we can tell some very interesting things about the New Zealand position:" ... http://nathan.torkington.com/blog/2010/03/01/nz-acta-negotiation/ (Nat will be well-known to members of the Perl community) and: "InternetNZ to take public message to ACTA negotiators Media Release - 2 March 2010 - InternetNZ (Internet New Zealand Inc) will assist the public in voicing its concerns about the controversial international Anti-Counterfeiting Trade Agreement (ACTA) through an open conference to be held next month in Wellington, New Zealand. "We're going to give the public the chance to have their say - in contrast to the secrecy of the negotiation process," says Jordan Carter, InternetNZ Policy Director. PublicACTA will be held on Saturday, 10 April 2010, two days ahead of Round 8 of the ACTA negotiations on 12-16 April in Wellington. The outputs of PublicACTA will be provided to the New Zealand government negotiators." ... http://www.internetnz.net.nz/media/media-releases-2010/internetnz-to-take-public-message-to-acta-negotiators Regards, =dn PS hope you'll support this week's meeting, especially to encourage the organisers and speakers - the first for some time! (do as I say, not as I do! Hereby apologies tendered) From general.mooney at googlemail.com Thu Mar 18 09:48:49 2010 From: general.mooney at googlemail.com (=?UTF-8?Q?Ciar=C3=A1n_Mooney?=) Date: Thu, 18 Mar 2010 09:48:49 +0000 Subject: [Gllug] [ANN] Meeting TONIGHT - libguestfs Message-ID: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> Hi, Final reminder of the meeting tonight at the University of Westminster by Richard Jones on libguestfs. Original announcement : http://lists.gllug.org.uk/pipermail/gllug/2010-February/082543.html Currently there are 19 confirmed attendees, if anyone else is planning to attend please email me. Although, I feel, this is non-compulsory, Sean has requested that I ask. Regards, Ciarán From rich at annexia.org Thu Mar 18 15:47:22 2010 From: rich at annexia.org (Richard Jones) Date: Thu, 18 Mar 2010 15:47:22 +0000 Subject: [Gllug] [ANN] Meeting TONIGHT - libguestfs In-Reply-To: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> References: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> Message-ID: <20100318154722.GA17389@annexia.org> On Thu, Mar 18, 2010 at 09:48:49AM +0000, Ciarán Mooney wrote: > Hi, > > Final reminder of the meeting tonight at the University of Westminster > by Richard Jones on libguestfs. > > Original announcement : > http://lists.gllug.org.uk/pipermail/gllug/2010-February/082543.html Yup, look forward to seeing everyone there. Hopefully neither my car nor my laptop will break down, otherwise it'll be a very short meeting ... Rich. -- Richard Jones Red Hat From benjamin at py-soft.co.uk Thu Mar 18 15:48:37 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Thu, 18 Mar 2010 15:48:37 +0000 Subject: [Gllug] [ANN] Meeting TONIGHT - libguestfs In-Reply-To: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> References: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> Message-ID: <732076a81003180848o2c9f5940vadd0472ad7a640c5@mail.gmail.com> On 18 March 2010 09:48, Ciarán Mooney wrote: > Currently there are 19 confirmed attendees, if anyone else is planning > to attend please email me. I'm afraid that I will not be able to make it. > Although, I feel, this is non-compulsory, Sean has requested that I ask. You won't get past security if your name isn't on the list! :) Ben From rich at annexia.org Thu Mar 18 22:29:33 2010 From: rich at annexia.org (Richard Jones) Date: Thu, 18 Mar 2010 22:29:33 +0000 Subject: [Gllug] [ANN] Meeting TONIGHT - libguestfs In-Reply-To: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> References: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> Message-ID: <20100318222933.GA1380@annexia.org> Hope everyone enjoyed it. I think we had a good turn out for a weekday evening. I will (eventually) put up the talk on my blog: http://rwmj.wordpress.com/ (I'll notify here when I get round to it). Rich. -- Richard Jones Red Hat From tm at tm.uklinux.net Fri Mar 19 07:45:56 2010 From: tm at tm.uklinux.net (TM) Date: Fri, 19 Mar 2010 07:45:56 +0000 Subject: [Gllug] High-latency RS-232 In-Reply-To: References: Message-ID: <4BA32BB4.6050509@tm.uklinux.net> Hi All, We are facing a problem at work talking to a piece of kit via RS-232 from a Linux box. This problem did not exist in Windows and now a few guys (from the Windows world) are beginning to question the move to Linux. This annoys me no end because there is a lot of 'Windows is almighty good and Linux is a pile of crap' from people who haven't even looked at the issue properly. We talk to this piece of kit via RS-232 over TCL. We use xon/xoff. We have traditionally used Windows and the comms works a treat. Now they have moved to Linux and the serial port has turned sluggish. It takes about 1000 characters to stop/pause. Which is way to much. I suggested fiddling with the niceness of the process. This helped but wasn't good enough to solve the problem. They run a 'stock' kernel from Fedora and/or RedHat. Moving to a different distribution (thinking real-time kernel here) is complicated because there is a lot of code developed for the current kernel/distribution. I think they could just happen to have a serial port with a generic, poorly optimised driver. I have suggested googling for a serial card well supported in Linux. We have also played with the RTS-CTS handshake. This introduces a bunch of other issues (solvable, no doubt, but requiring time and effort). So, we would like to sick to Xon/Xoff. Can anyone suggest ways of tackling the problem please? Many thanks TM From tim at seacon.co.uk Fri Mar 19 08:05:20 2010 From: tim at seacon.co.uk (t.clarke) Date: Fri, 19 Mar 2010 8:05:20 +0000 Subject: [Gllug] High-latency RS-232 In-Reply-To: <4BA32BB4.6050509@tm.uklinux.net> References: <4BA32BB4.6050509@tm.uklinux.net> Message-ID: A bit more info on the RS232 setup would be helpful: an 'stty -a' print of the serial port would be good I assume you are talking about delays in Linux stopping SENDING characters to the device? How many characters are sent in each discrete 'write' call from the process? I don't know the internals of the linux kernel, but I always believed that serial flow control is implemented within the kernel layer and not the process - so I wouldn't expect 'nice' to make any real difference! Tim From danthegeekman at googlemail.com Fri Mar 19 09:54:19 2010 From: danthegeekman at googlemail.com (Dan) Date: Fri, 19 Mar 2010 09:54:19 +0000 (GMT Standard Time) Subject: [Gllug] [ANN] Meeting TONIGHT - libguestfs In-Reply-To: <20100318222933.GA1380@annexia.org> Message-ID: On Thu, Mar 18, 2010 at 10:29 PM, Richard Jones <rich at annexia.org> wrote: Hope everyone enjoyed it.  I think we had a good turn out for a weekday evening. Excellent talk Rich, really enjoyed it. I will (eventually) put up the talk on my blog: http://rwmj.wordpress.com/ (I'll notify here when I get round to it). I look forward to that. Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100319/3367c30d/attachment.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 302 bytes Desc: OpenPGP digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100319/3367c30d/attachment.pgp From james.dutton at gmail.com Fri Mar 19 10:07:20 2010 From: james.dutton at gmail.com (James Courtier-Dutton) Date: Fri, 19 Mar 2010 10:07:20 +0000 Subject: [Gllug] High-latency RS-232 In-Reply-To: <4BA32BB4.6050509@tm.uklinux.net> References: <4BA32BB4.6050509@tm.uklinux.net> Message-ID: On 19 March 2010 07:45, TM wrote: > > We talk to this piece of kit via RS-232 over TCL. We use xon/xoff. We > have traditionally used Windows and the comms works a treat. Now they > have moved to Linux and the serial port has turned sluggish. It takes > about 1000 characters to stop/pause. Which is way to much. > > I suggested fiddling with the niceness of the process. This helped but > wasn't good enough to solve the problem. > > They run a 'stock' kernel from Fedora and/or RedHat. Moving to a > different distribution (thinking real-time kernel here) is complicated > because there is a lot of code developed for the current > kernel/distribution. > > I think they could just happen to have a serial port with a generic, > poorly optimised driver. I have suggested googling for a serial card > well supported in Linux. > > We have also played with the RTS-CTS handshake. This introduces a bunch > of other issues (solvable, no doubt, but requiring time and effort). So, > we would like to sick to Xon/Xoff. > > Can anyone suggest ways of tackling the problem please? > I don't suppose you have a sample TCL program you can share. There are two ways to do XON/XOFF in Linux. One is at the application layer, the other is in the tty layer. The tty layer will react faster. You do not say which direction the problem is in. I.e. the Linux sending, or the other box sending chars. Another problem you might have is Linux by default uses the nagle algorithm. What this means is that the serial buffers will buffer up input chars until either the buffer is full, or there is a gap is transmission. So, what you might be seeing is characters that have already been received by the Linux box, but were just buffered. The XOFF causes a gap in transmission, and thus you see the buffer empty and pushed to the application. Now, XON/XOFF is quite rare now, so there might be a bug in the Linux driver. If it is using the tty layer correctly, I think that you should never see the XON/XOFF chars in the stream at the application layer. The tty layer should take care of them. If the problem is in the send direction. I.e. Linux is sending chars, and waiting to receive an XOFF. This will be slow to react. It always has been in Linux. There is no need to make it any quicker when Linux is talking to Linux. Linux is actually much faster at reacting in the other direction. I.e. It can send an XOFF very quickly when needed. It is just rather sluggish detecting the incoming XOFF. A way to speed this detection up is to try to ensure that your application reads the serial port as quickly as possible. Thus the input buffers are kept fairly empty and thus the XOFF is processed more quickly. One then needs to also only fill the tx buffers just in time, because the way Linux handles XOFF is that when it receives the XOFF, it continues to send everything in the ring buffer and then stops sending more. The ring buffer is about 4K in size, so only sending 1000 chars is actually quick a good result!!! It is technically possible to detect the XOFF earlier, but Linux choose not to do it that way. Detecting XOFF early makes everything more complex and uses more CPU resources. Extract from a Linux programming guide: "Hardware handshake is optional, but very useful. It allows either of the two stations to signal whether it is ready to receive more data, or if the other station should pause until the receiver is done processing the incoming data. The lines used for this are called "Clear to Send" (CTS) and "Ready to Send" (RTS), respectively, which explains the colloquial name for hardware handshake: "RTS/CTS." The other type of handshake you might be familiar with is called "XON/XOFF" handshaking. XON/XOFF uses two nominated characters, conventionally Ctrl-S and Ctrl-Q, to signal to the remote end that it should stop and start transmitting data, respectively. While this method is simple to implement and okay for use by dumb terminals, it causes great confusion when you are dealing with binary data, as you may want to transmit those characters as part of your data stream, and not have them interpreted as flow control characters. It is also somewhat slower to take effect than hardware handshake. Hardware handshake is clean, fast, and recommended in preference to XON/XOFF when you have a choice." So, you have two options. 1) Implement fast handling rx of XOFF in the kernel yourself. 2) Modify your program to ensure the tx and rx buffers are kept as empty as possible. 3) Use hardware flow control. Generally, XON/XOFF is really only meant to be used for interactive terminals where their are often gaps in tx and rx of data, so the XON/XOFF is processed in good time. All other datacomms should really use hardware flow control. I do not know exactly what your application is, but I would be surprised if PCs nowdays can not handle full speed serial comms without ever having to send an XOFF or even resort to flow control. From rich at annexia.org Fri Mar 19 10:28:20 2010 From: rich at annexia.org (Richard Jones) Date: Fri, 19 Mar 2010 10:28:20 +0000 Subject: [Gllug] Slides from the talk (was: Re: [ANN] Meeting TONIGHT - libguestfs) In-Reply-To: <20100318222933.GA1380@annexia.org> References: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> <20100318222933.GA1380@annexia.org> Message-ID: <20100319102819.GA25105@annexia.org> http://rwmj.wordpress.com/2010/03/19/gllug-talk-on-libguestfs-18th-march-2010/#content -- Richard Jones Red Hat From tethys at gmail.com Fri Mar 19 11:31:13 2010 From: tethys at gmail.com (- Tethys) Date: Fri, 19 Mar 2010 11:31:13 +0000 Subject: [Gllug] Slides from the talk (was: Re: [ANN] Meeting TONIGHT - libguestfs) In-Reply-To: <20100319102819.GA25105@annexia.org> References: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> <20100318222933.GA1380@annexia.org> <20100319102819.GA25105@annexia.org> Message-ID: <264d99b01003190431n122e034k5c9afedcce4da27e@mail.gmail.com> On Fri, Mar 19, 2010 at 10:28 AM, Richard Jones wrote: > http://rwmj.wordpress.com/2010/03/19/gllug-talk-on-libguestfs-18th-march-2010/#content Interesting talk, thanks. That said, the thing that grabbed me most from the evening was Tech Talk PSE. I'll *definitely* be using that in my future presentations. Tet -- ?It seems intuitively obvious to me, which means that it might be wrong.? -- Chris Torek From richardlewis at fastmail.co.uk Fri Mar 19 12:41:05 2010 From: richardlewis at fastmail.co.uk (Richard Lewis) Date: Fri, 19 Mar 2010 12:41:05 +0000 Subject: [Gllug] GDM Xclients script In-Reply-To: <20100315132757.GB9234@phaistos.bruce> References: <87hbohepba.wl%richard.lewis@gold.ac.uk> <20100315132757.GB9234@phaistos.bruce> Message-ID: <871vfg32n2.wl%richard.lewis@gold.ac.uk> At Mon, 15 Mar 2010 13:27:57 +0000, Bruce Richardson wrote: > > [1 ] > [1.1 ] > On Mon, Mar 15, 2010 at 12:34:33PM +0000, Richard wrote: > > > > Any thoughts? Anyone seen anything like this before? > > Have you had a look at /etc/X11/Xsession to see what it does before it > runs the user session script? I don't have an unstable installation to > hand, but the typical behaviour is to run any scripts in > /etc/X11/Xsession.d/ first. Could be something in there that is > preempting your user script. > Thanks for your suggestion. I had a look through and couldn't see anything obviously wrong. I did another upgrade last night and today when I logged in using "Run Xclients script" it started my normal session. So something is now working. Not sure what, though. -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Richard Lewis ISMS, Computing Goldsmiths, University of London Tel: +44 (0)20 7078 5134 Skype: richardjlewis JID: ironchicken at jabber.earth.li http://www.richard-lewis.me.uk/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- +----------------------------------------------+ | Support open access to scholarship | | http://freeculture.org/ http://www.doaj.org/ | +----------------------------------------------+ From itsbruce at workshy.org Fri Mar 19 13:54:21 2010 From: itsbruce at workshy.org (Bruce Richardson) Date: Fri, 19 Mar 2010 13:54:21 +0000 Subject: [Gllug] GDM Xclients script In-Reply-To: <871vfg32n2.wl%richard.lewis@gold.ac.uk> References: <87hbohepba.wl%richard.lewis@gold.ac.uk> <20100315132757.GB9234@phaistos.bruce> <871vfg32n2.wl%richard.lewis@gold.ac.uk> Message-ID: <20100319135421.GA3998@phaistos.bruce> On Fri, Mar 19, 2010 at 12:41:05PM +0000, Richard wrote: > > I did another upgrade last night and today when I logged in using "Run > Xclients script" it started my normal session. So something is now > working. Not sure what, though. If you are running unstable, it could be something as simple as a bug that was fixed between the two upgrades you did. Another possibility would be that a package was only incompletely installed first time round and the second upgrade fixed that. There are a number of circumstances which can cause an upgrade to fail in a way that can be fixed by immediately running it again or running "apt-get -f install"; those circumstances are more common with the unstable distribution. -- Bruce I see a mouse. Where? There, on the stair. And its clumsy wooden footwear makes it easy to trap and kill. -- Harry Hill -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100319/bb649c94/attachment.pgp From rich at annexia.org Fri Mar 19 17:38:01 2010 From: rich at annexia.org (Richard Jones) Date: Fri, 19 Mar 2010 17:38:01 +0000 Subject: [Gllug] Slides from the talk (was: Re: [ANN] Meeting TONIGHT - libguestfs) In-Reply-To: <264d99b01003190431n122e034k5c9afedcce4da27e@mail.gmail.com> References: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> <20100318222933.GA1380@annexia.org> <20100319102819.GA25105@annexia.org> <264d99b01003190431n122e034k5c9afedcce4da27e@mail.gmail.com> Message-ID: <20100319173800.GB25105@annexia.org> On Fri, Mar 19, 2010 at 11:31:13AM +0000, - Tethys wrote: > On Fri, Mar 19, 2010 at 10:28 AM, Richard Jones wrote: > > > http://rwmj.wordpress.com/2010/03/19/gllug-talk-on-libguestfs-18th-march-2010/#content > > Interesting talk, thanks. That said, the thing that grabbed me most > from the evening was Tech Talk PSE. I'll *definitely* be using that in > my future presentations. Patches welcome. There's a 'TODO' file if you're looking for ideas ... Rich. -- Richard Jones Red Hat From neil at safeharbourit.co.uk Fri Mar 19 20:55:04 2010 From: neil at safeharbourit.co.uk (Neil Tancock) Date: Fri, 19 Mar 2010 20:55:04 -0000 Subject: [Gllug] High-latency RS-232 Message-ID: Hiya, I've got a couple of Lantronix boxes if they'd be any help? You could use them to test the serial port or create a new port to test with the equipment? I agree about the window gripers - dogma isn't always a good thing! Neil -----Original Message----- From: gllug-bounces at gllug.org.uk [mailto:gllug-bounces at gllug.org.uk] On Behalf Of TM Sent: 19 March 2010 07:46 To: gllug at gllug.org.uk Subject: [Gllug] High-latency RS-232 Hi All, We are facing a problem at work talking to a piece of kit via RS-232 from a Linux box. This problem did not exist in Windows and now a few guys (from the Windows world) are beginning to question the move to Linux. This annoys me no end because there is a lot of 'Windows is almighty good and Linux is a pile of crap' from people who haven't even looked at the issue properly. We talk to this piece of kit via RS-232 over TCL. We use xon/xoff. We have traditionally used Windows and the comms works a treat. Now they have moved to Linux and the serial port has turned sluggish. It takes about 1000 characters to stop/pause. Which is way to much. I suggested fiddling with the niceness of the process. This helped but wasn't good enough to solve the problem. They run a 'stock' kernel from Fedora and/or RedHat. Moving to a different distribution (thinking real-time kernel here) is complicated because there is a lot of code developed for the current kernel/distribution. I think they could just happen to have a serial port with a generic, poorly optimised driver. I have suggested googling for a serial card well supported in Linux. We have also played with the RTS-CTS handshake. This introduces a bunch of other issues (solvable, no doubt, but requiring time and effort). So, we would like to sick to Xon/Xoff. Can anyone suggest ways of tackling the problem please? Many thanks TM -- Gllug mailing list - Gllug at gllug.org.uk http://lists.gllug.org.uk/mailman/listinfo/gllug From adrian at newgolddream.dyndns.info Sun Mar 21 22:54:19 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Sun, 21 Mar 2010 22:54:19 +0000 Subject: [Gllug] [OT] Hardware problem Message-ID: <1269212059.27507.6.camel@bossclass> The good news (that makes this ever-so-slightly less than a spam email) is that now I have finally acted on my oft repeated promise to the kids to get/build them a new/faster computer they have both agreed that they no longer want Windows on it and will be happier with Linux. The bad news is that, having bought a high spec Mobo (Asus P7H55-M Pro) for them, it seems unwilling to recognise their old 200G IDE (PATA) disk, though it will recognise the DVD drive on the same cable. Actually, not recognise is not the right word - not see at all is better. I cannot google anything to suggest that I have hit a known problem so am coming to believe the drive has been killed in some way, though don't know how. I have to test that obviously, but before I unscrew everything again, anybody ever had a similar problem and know of some heuristics I might employ to fix this in a software/BIOS sort of way. Long shot, I know... From addw at phcomp.co.uk Sun Mar 21 23:02:26 2010 From: addw at phcomp.co.uk (Alain Williams) Date: Sun, 21 Mar 2010 23:02:26 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269212059.27507.6.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> Message-ID: <20100321230226.GK420@phcomp.co.uk> On Sun, Mar 21, 2010 at 10:54:19PM +0000, Adrian McMenamin wrote: > The good news (that makes this ever-so-slightly less than a spam email) > is that now I have finally acted on my oft repeated promise to the kids > to get/build them a new/faster computer they have both agreed that they > no longer want Windows on it and will be happier with Linux. > > The bad news is that, having bought a high spec Mobo (Asus P7H55-M Pro) > for them, it seems unwilling to recognise their old 200G IDE (PATA) > disk, though it will recognise the DVD drive on the same cable. > > Actually, not recognise is not the right word - not see at all is > better. > > I cannot google anything to suggest that I have hit a known problem so > am coming to believe the drive has been killed in some way, though don't > know how. > > I have to test that obviously, but before I unscrew everything again, > anybody ever had a similar problem and know of some heuristics I might > employ to fix this in a software/BIOS sort of way. It might be simplest to just get a SATA disk, a quick peep at my local supplier and I see: * Hitachi Deskstar 320GB 3.5 7200rpm 16mb Sata -- #35.24 inc vat * Hitachi Deskstar 500GB SATA-II 16MB Cache -- #39.89 inc vat This means that you have the old machine as a backup/... -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php Past chairman of UKUUG: http://www.ukuug.org/ #include From gllug at getaroundtoit.co.uk Sun Mar 21 23:15:41 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Mon, 22 Mar 2010 12:15:41 +1300 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269212059.27507.6.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> Message-ID: <4BA6A89D.30601@getaroundtoit.co.uk> On 22/03/10 11:54, Adrian McMenamin wrote: ... > Actually, not recognise is not the right word - not see at all is > better. When you say "not see" are you talking about the Linux level or the BIOS level? Check the BIOS is including/prioritising parallel drives... Regards, =dn From adrian at newgolddream.dyndns.info Sun Mar 21 23:19:33 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Sun, 21 Mar 2010 23:19:33 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <4BA6A89D.30601@getaroundtoit.co.uk> References: <1269212059.27507.6.camel@bossclass> <4BA6A89D.30601@getaroundtoit.co.uk> Message-ID: <1269213573.27507.9.camel@bossclass> On Mon, 2010-03-22 at 12:15 +1300, David L Neil wrote: > On 22/03/10 11:54, Adrian McMenamin wrote: > ... > > > Actually, not recognise is not the right word - not see at all is > > better. > > When you say "not see" are you talking about the Linux level or the BIOS > level? > > Check the BIOS is including/prioritising parallel drives... It only reports the DVD drive as present. When I boot the thing with Knoppix it reports that the DVD is there as /dev/hdb - there is no /dev/hda (is this a clue does anyone think?). There are no bios options for PATA/IDE drives at all, other than to ensure the onboard interface is on (which obviously it is) From john at cornerstonelinux.co.uk Sun Mar 21 23:20:09 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Sun, 21 Mar 2010 23:20:09 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <4BA6A89D.30601@getaroundtoit.co.uk> References: <1269212059.27507.6.camel@bossclass> <4BA6A89D.30601@getaroundtoit.co.uk> Message-ID: <20100321232008.GE7740@cornerstonelinux.co.uk> On Mon, Mar 22, 2010 at 12:15:41PM +1300, David L Neil wrote: > On 22/03/10 11:54, Adrian McMenamin wrote: > ... > >> Actually, not recognise is not the right word - not see at all is >> better. > > When you say "not see" are you talking about the Linux level or the BIOS > level? > > Check the BIOS is including/prioritising parallel drives... It may also be worthwhile checking that the PATA controller is fully enabled. In some BIOS there may also be a "legacy" mode. -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100321/ac19deb3/attachment.pgp From john at cornerstonelinux.co.uk Sun Mar 21 23:23:53 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Sun, 21 Mar 2010 23:23:53 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <20100321232008.GE7740@cornerstonelinux.co.uk> References: <1269212059.27507.6.camel@bossclass> <4BA6A89D.30601@getaroundtoit.co.uk> <20100321232008.GE7740@cornerstonelinux.co.uk> Message-ID: <20100321232353.GF7740@cornerstonelinux.co.uk> On Sun, Mar 21, 2010 at 11:20:09PM +0000, John Edwards wrote: > On Mon, Mar 22, 2010 at 12:15:41PM +1300, David L Neil wrote: > > On 22/03/10 11:54, Adrian McMenamin wrote: > > ... > > > >> Actually, not recognise is not the right word - not see at all is > >> better. > > > > When you say "not see" are you talking about the Linux level or the BIOS > > level? > > > > Check the BIOS is including/prioritising parallel drives... > > It may also be worthwhile checking that the PATA controller is > fully enabled. In some BIOS there may also be a "legacy" mode. OK. Just read your recent email, so that has already been checked. Next thing is to check the data and power cables, and try replacements. Can you hear/feel the drive spinning up when the machine is powered on? Also try another IDE disk. If several IDE disks can not be read with different data and power cables, then the motherboard is probably faulty. -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100321/b92a3f54/attachment.pgp From gllug at getaroundtoit.co.uk Mon Mar 22 02:26:17 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Mon, 22 Mar 2010 15:26:17 +1300 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269213573.27507.9.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> <4BA6A89D.30601@getaroundtoit.co.uk> <1269213573.27507.9.camel@bossclass> Message-ID: <4BA6D549.9080701@getaroundtoit.co.uk> On 22/03/10 12:19, Adrian McMenamin wrote: > On Mon, 2010-03-22 at 12:15 +1300, David L Neil wrote: >> On 22/03/10 11:54, Adrian McMenamin wrote: >>> Actually, not recognise is not the right word - not see at all is >>> better. >> When you say "not see" are you talking about the Linux level or the BIOS >> level? >> Check the BIOS is including/prioritising parallel drives... > It only reports the DVD drive as present. When I boot the thing with > Knoppix it reports that the DVD is there as /dev/hdb - there is > no /dev/hda (is this a clue does anyone think?). > There are no bios options for PATA/IDE drives at all, other than to > ensure the onboard interface is on (which obviously it is) Is this a Knoppix live disk? In which case is it reserving /dev/hda for its own purposes? If there is no mention in the BIOS displays, (ensure all options for such displays are selected) what connections does it report during the BIOS boot phase? What happens if you select to boot from the (PATA) HDD as first priority? (at the moment you presumably have CD/DVD prioritised) Regards, =dn From walter.stanish at saffrondigital.com Mon Mar 22 02:45:25 2010 From: walter.stanish at saffrondigital.com (Walter Stanish) Date: Mon, 22 Mar 2010 02:45:25 -0000 Subject: [Gllug] [OT] Hardware problem References: <1269212059.27507.6.camel@bossclass><4BA6A89D.30601@getaroundtoit.co.uk> <1269213573.27507.9.camel@bossclass> Message-ID: > It only reports the DVD drive as present. Some possibilities as follows. If you are recycling a CDROM and sharing a PATA cable, then many of the PATA CDROM drives used to have their jumpers set to force master or slave. In this case, yours may be set to force slave. The default (no jumpers) tends to be 'cable select', which means that if you plug a single (multi-device) PATA cable in to both a disk and a CDROM it may refuse to see one of them entirely. Removing the jumper could fix it instantly. If not, then removing the jumper and swapping the order of the CD/HDD on the cable may work. If you are using dedicated cables, then I would suggest unplugging all IDE/SATA devices except the problem disk. If BIOS is unable to see it using a power cable that another drive works with, and a PATA port that another drive works with, and removing jumpers doesn't help, then hold the disk when it's powered on. If it doesn't spin up then you have a power problem or dead disk. If it does spin up but the BIOS doesn't see it then you may have a cable issue (there were a few types of PATA cables produced, IIRC UDMA/66 or something required a nicer cable... older cables were incapable, and your drive firmware maybe doesn't like the fact .. pretty long shot here). Safest option is always a dedicated cable to each PATA device. - Walter -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/ms-tnef Size: 3274 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100322/319b64bd/attachment-0001.bin From bap at shrdlu.com Mon Mar 22 06:54:15 2010 From: bap at shrdlu.com (Bernard Peek) Date: Mon, 22 Mar 2010 06:54:15 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269213573.27507.9.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> <4BA6A89D.30601@getaroundtoit.co.uk> <1269213573.27507.9.camel@bossclass> Message-ID: <4BA71417.3080507@shrdlu.com> On 21/03/10 23:19, Adrian McMenamin wrote: > On Mon, 2010-03-22 at 12:15 +1300, David L Neil wrote: > >> On 22/03/10 11:54, Adrian McMenamin wrote: >> ... >> >> >>> Actually, not recognise is not the right word - not see at all is >>> better. >>> >> When you say "not see" are you talking about the Linux level or the BIOS >> level? >> >> Check the BIOS is including/prioritising parallel drives... >> > It only reports the DVD drive as present. When I boot the thing with > Knoppix it reports that the DVD is there as /dev/hdb - there is > no /dev/hda (is this a clue does anyone think?). > This just means that the DVD is configured as the slave device, it's not relevant to this particular issue. Have you tried using a different ribbon cable, or using that drive in a different box? -- Bernard Peek bap at shrdlu.com From jjllmmss at googlemail.com Mon Mar 22 08:36:24 2010 From: jjllmmss at googlemail.com (JLMS) Date: Mon, 22 Mar 2010 08:36:24 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269212059.27507.6.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> Message-ID: On Sun, Mar 21, 2010 at 10:54 PM, Adrian McMenamin wrote: > The good news (that makes this ever-so-slightly less than a spam email) > is that now I have finally acted on my oft repeated promise to the kids > to get/build them a new/faster computer they have both agreed that they > no longer want Windows on it and will be happier with Linux. > > The bad news is that, having bought a high spec Mobo (Asus P7H55-M Pro) > for them, it seems unwilling to recognise their old 200G IDE (PATA) > disk, though it will recognise the DVD drive on the same cable. > > Actually, not recognise is not the right word - not see at all is > better. > > I cannot google anything to suggest that I have hit a known problem so > am coming to believe the drive has been killed in some way, though don't > know how. > > I have to test that obviously, but before I unscrew everything again, > anybody ever had a similar problem and know of some heuristics I might > employ to fix this in a software/BIOS sort of way. > > Long shot, I know... Try using an external USB enclosure. You can get them for little on Ebay From james.dutton at gmail.com Mon Mar 22 09:35:22 2010 From: james.dutton at gmail.com (James Courtier-Dutton) Date: Mon, 22 Mar 2010 09:35:22 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269212059.27507.6.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> Message-ID: On 21 March 2010 22:54, Adrian McMenamin wrote: > > The bad news is that, having bought a high spec Mobo (Asus P7H55-M Pro) > for them, it seems unwilling to recognise their old 200G IDE (PATA) > disk, though it will recognise the DVD drive on the same cable. > > Actually, not recognise is not the right word - not see at all is > better. > Try checking the dip switches on the HD. Unplug the DVD, and configure the HD to be slave. It is quite possible that being an old HD, just moving it to the new machine has caused some damage. I know a customer who had a server with 5 HDs in it running RAID. They powered it down to move it to a different room. When they powered it up again, 4 out of the 5 HDs had failed. They raised a support call with IBM and the response from IBM is that they did not believe the customer. Apparently failure of IBM disks was rare, but failing 4 out of 5 on the same day was so unlikely that they did not believe the customer. The disks were only 1 month old. IBM promptly sent an engineer out to the customer for free to investigate. Only after the engineer confirmed the fault, did IBM finally agree with the customer. This was a long time ago now, so I am not taking a swipe at IBM specifically, it is just an isolated example. From iain at shihad.org Mon Mar 22 10:28:47 2010 From: iain at shihad.org (Iain M Conochie) Date: Mon, 22 Mar 2010 10:28:47 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269212059.27507.6.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> Message-ID: <4BA7465F.9030103@shihad.org> Adrian McMenamin wrote: > > > The bad news is that, having bought a high spec Mobo (Asus P7H55-M Pro) > for them, it seems unwilling to recognise their old 200G IDE (PATA) > disk, though it will recognise the DVD drive on the same cable. > Have you tried the disk on the cable by itself? Most new machines expect only one drive on the PATA cables, and also usually expect it to be a CD / DVD ROM device. There may be BIOS features to turn off this annoyance but I have not reaaly found any Good luck Iain From itsbruce at workshy.org Mon Mar 22 14:44:49 2010 From: itsbruce at workshy.org (Bruce Richardson) Date: Mon, 22 Mar 2010 14:44:49 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269213573.27507.9.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> <4BA6A89D.30601@getaroundtoit.co.uk> <1269213573.27507.9.camel@bossclass> Message-ID: <20100322144449.GA4523@phaistos.bruce> On Sun, Mar 21, 2010 at 11:19:33PM +0000, Adrian wrote: > On Mon, 2010-03-22 at 12:15 +1300, David L Neil wrote: > > On 22/03/10 11:54, Adrian McMenamin wrote: > > ... > > > > > Actually, not recognise is not the right word - not see at all is > > > better. > > > > When you say "not see" are you talking about the Linux level or the BIOS > > level? > > > > Check the BIOS is including/prioritising parallel drives... > > It only reports the DVD drive as present. When I boot the thing with > Knoppix it reports that the DVD is there as /dev/hdb - there is > no /dev/hda (is this a clue does anyone think?). Have you gone into the BIOS set-up to see if the BIOS can see the drive, or are you going entirely by what Knoppix reports? -- Bruce I object to intellect without discipline. I object to power without constructive purpose. -- Spock -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100322/16c61f65/attachment.pgp From damerell at chiark.greenend.org.uk Mon Mar 22 15:11:10 2010 From: damerell at chiark.greenend.org.uk (David Damerell) Date: Mon, 22 Mar 2010 15:11:10 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: References: <1269212059.27507.6.camel@bossclass> <4BA6A89D.30601@getaroundtoit.co.uk> <1269213573.27507.9.camel@bossclass> Message-ID: <19367.34958.646750.686998@chiark.greenend.org.uk> On Monday, 22 Mar 2010, Walter Stanish wrote: >(there were a few types of PATA cables produced, IIRC UDMA/66 or >something required a nicer cable... older cables were incapable, >and your drive firmware maybe doesn't like the fact .. pretty >long shot here). UDMA/66 cables typically require explicitly that the master and slave be in specific positions. I would start by attaching the disc only in each position and checking that the disc physically spins up at boot time (if you hold it in your hand the sensation is unmistakable). -- David Damerell flcl? Today is Second Leicesterday, March. Tomorrow will be Second Brieday, March. From richard.leger at gmail.com Mon Mar 22 16:15:40 2010 From: richard.leger at gmail.com (Richard LEGER) Date: Mon, 22 Mar 2010 16:15:40 +0000 Subject: [Gllug] =?iso-8859-1?q?Network_Simulation_Tools_=28netkit=2C_cloo?= =?iso-8859-1?q?nix-net=29_to_the_attention_of_John=2C_No=EBl_or_an?= =?iso-8859-1?q?y_other_interested=2E=2E=2E?= Message-ID: Ciaran, Thank you for organisation of last thursday GLLUG event, the libguest presentation was really intersting... Noël, John, Looking forward to the wireshark presentation at next month GLLUG event... would you have more information about it (such as date, time of event), let us know... John, Following our first meeting at the GLLUG event last thursday, as we discussed, find below information about tools for making network simulation via Virtual Machines... Netkit (http://wiki.netkit.org ) that I have been using for my current training Cloonix-Net (http://cloonix.net) an interesting one that I may test later on... and from which you can now simulate network based on OpenWrt KVM machines (new feature since 2009-12 apparently) They seems to be other as well that I have not yet looked into in detail but that seem providing similar features... Vde Vnuml Marionnet See you Richard -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100322/2bbb6663/attachment.htm From general.mooney at googlemail.com Mon Mar 22 16:41:01 2010 From: general.mooney at googlemail.com (=?UTF-8?Q?Ciar=C3=A1n_Mooney?=) Date: Mon, 22 Mar 2010 16:41:01 +0000 Subject: [Gllug] =?utf-8?q?Network_Simulation_Tools_=28netkit=2C_cloonix-n?= =?utf-8?q?et=29_to_the_attention_of_John=2C_No=C3=ABl_or_any_other?= =?utf-8?q?_interested=2E=2E=2E?= In-Reply-To: References: Message-ID: <3e4e51a81003220941w42de7f1aw17f0e7a7c768dd6a@mail.gmail.com> > Thank you for organisation of last thursday GLLUG event, the libguest > presentation was really intersting... It was a pleasure. > Looking forward to the wireshark presentation at next month GLLUG event... > would you have more information about it (such as date, time of event), let > us know... I am organising this one too, just waiting on conformation of the venue (Large Lecture Theatre) then I'll be putting out an announcement. Regards, Ciarán From sta296 at astradyne.co.uk Mon Mar 22 20:49:29 2010 From: sta296 at astradyne.co.uk (Tethys) Date: Mon, 22 Mar 2010 20:49:29 +0000 Subject: [Gllug] ATA errors Message-ID: <201003222049.o2MKnTaI003340@leto.astradyne.corp> My backup failed last night, because the device was read only: > building file list ... done > rsync: delete_file: unlink "/backup/local/shared/camera/2009-08-24--foq/img_0507.jpg" failed: Read-only file system (30) > rsync: delete_file: unlink "/backup/local/shared/camera/2009-08-24--foq/.xvpics/img_0507.jpg" failed: Read-only file system (30) That's never a good sign. Looking through the logs, I see a number of ATA errors, starting with a timeout, a device error, and a bunch of HSM violations. A few examples: > Mar 22 07:56:03 riva kernel: ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen > Mar 22 07:56:03 riva kernel: ata2.00: cmd 25/00:00:87:d1:86/00:02:03:00:00/e0 tag 0 dma 262144 in > Mar 22 07:56:03 riva kernel: res 40/00:00:00:00:00/ff:ff:ff:ff:ff/00 Emask 0x4 (timeout) > Mar 22 07:56:03 riva kernel: ata2.00: status: { DRDY } > Mar 22 07:56:03 riva kernel: ata2: hard resetting link > Mar 22 07:56:03 riva kernel: ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310) > Mar 22 07:56:03 riva kernel: ata2.00: configured for UDMA/33 > Mar 22 07:56:03 riva kernel: ata2: EH complete > Mar 22 07:56:03 riva kernel: SCSI device sdb: 1465149168 512-byte hdwr sectors (750156 MB) > Mar 22 07:56:03 riva kernel: sdb: Write Protect is off > Mar 22 07:56:03 riva kernel: SCSI device sdb: drive cache: write back > Mar 22 07:59:05 riva kernel: ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0 > Mar 22 07:59:05 riva kernel: ata2.00: BMDMA2 stat 0x6d0009 > Mar 22 07:59:05 riva kernel: ata2.00: cmd 25/00:00:df:ce:87/00:02:03:00:00/e0 tag 0 dma 262144 in > Mar 22 07:59:05 riva kernel: res 51/04:80:5f:cf:87/00:01:03:00:00/e0 Emask 0x1 (device error) > Mar 22 07:59:05 riva kernel: ata2.00: status: { DRDY ERR } > Mar 22 07:59:05 riva kernel: ata2.00: error: { ABRT } > Mar 22 07:59:05 riva kernel: ata2.00: configured for UDMA/33 > Mar 22 07:59:05 riva kernel: ata2: EH complete > Mar 22 07:59:05 riva kernel: SCSI device sdb: 1465149168 512-byte hdwr sectors (750156 MB) > Mar 22 07:59:05 riva kernel: sdb: Write Protect is off > Mar 22 07:59:05 riva kernel: SCSI device sdb: drive cache: write back > Mar 22 08:01:15 riva kernel: ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen > Mar 22 08:01:15 riva kernel: ata2.00: cmd 25/00:00:ff:1e:88/00:02:03:00:00/e0 tag 0 dma 262144 in > Mar 22 08:01:15 riva kernel: res ff/ff:ff:ff:ff:ff/ff:ff:ff:ff:ff/ff Emask 0x2 (HSM violation) > Mar 22 08:01:15 riva kernel: ata2.00: status: { Busy } > Mar 22 08:01:15 riva kernel: ata2.00: error: { ICRC UNC IDNF ABRT } > Mar 22 08:01:15 riva kernel: ata2: hard resetting link > Mar 22 08:01:15 riva kernel: ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310) > Mar 22 08:01:15 riva kernel: ata2.00: configured for UDMA/33 > Mar 22 08:01:15 riva kernel: ata2: EH complete > Mar 22 08:01:15 riva kernel: SCSI device sdb: 1465149168 512-byte hdwr sectors (750156 MB) > Mar 22 08:01:15 riva kernel: sdb: Write Protect is off > Mar 22 08:01:15 riva kernel: SCSI device sdb: drive cache: write back SMART doesn't seem to be much help here, with the device refusing to run anything but the mandatory offline test: > riva:~# smartctl --test=offline /dev/sdb > smartctl version 5.36 [i686-redhat-linux-gnu] Copyright (C) 2002-6 Bruce Allen > Home page is http://smartmontools.sourceforge.net/ > > Default Self Test Successful > riva:~# smartctl --test=short /dev/sdb > smartctl version 5.36 [i686-redhat-linux-gnu] Copyright (C) 2002-6 Bruce Allen > Home page is http://smartmontools.sourceforge.net/ > > Short offline self test failed [unsupported field in scsi command] > riva:~# smartctl --test=long /dev/sdb > smartctl version 5.36 [i686-redhat-linux-gnu] Copyright (C) 2002-6 Bruce Allen > Home page is http://smartmontools.sourceforge.net/ > > Long (extended) offline self test failed [unsupported field in scsi command] It's a Seagate drive: > Host: scsi1 Channel: 00 Id: 00 Lun: 00 > Vendor: ATA Model: ST3750528AS Rev: CC38 > Type: Direct-Access ANSI SCSI revision: 05 I'm running 2.6.18-128.4.1.el5 on CentOS 5. Any ideas? Does anyone know enough about the ATA spec to tell me what these errors actually mean? Is this a genuine hardware failure? If so, where's it likely to be? Drive? Controller? Cable? The drive is only a few weeks old, so while sometimes shit happens, I want to investigate the (probably more likely) alternatives as well. Tet From addw at phcomp.co.uk Mon Mar 22 20:56:16 2010 From: addw at phcomp.co.uk (Alain Williams) Date: Mon, 22 Mar 2010 20:56:16 +0000 Subject: [Gllug] ATA errors In-Reply-To: <201003222049.o2MKnTaI003340@leto.astradyne.corp> References: <201003222049.o2MKnTaI003340@leto.astradyne.corp> Message-ID: <20100322205616.GS4083@phcomp.co.uk> On Mon, Mar 22, 2010 at 08:49:29PM +0000, Tethys wrote: > > My backup failed last night, because the device was read only: > > > building file list ... done > > rsync: delete_file: unlink "/backup/local/shared/camera/2009-08-24--foq/img_0507.jpg" failed: Read-only file system (30) > > rsync: delete_file: unlink "/backup/local/shared/camera/2009-08-24--foq/.xvpics/img_0507.jpg" failed: Read-only file system (30) > > That's never a good sign. Looking through the logs, I see a number of > ATA errors, starting with a timeout, a device error, and a bunch of > HSM violations. A few examples: > > > Mar 22 07:56:03 riva kernel: ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen > > Mar 22 07:56:03 riva kernel: ata2.00: cmd 25/00:00:87:d1:86/00:02:03:00:00/e0 tag 0 dma 262144 in > > Mar 22 07:56:03 riva kernel: res 40/00:00:00:00:00/ff:ff:ff:ff:ff/00 Emask 0x4 (timeout) > > Mar 22 07:56:03 riva kernel: ata2.00: status: { DRDY } > > Mar 22 07:56:03 riva kernel: ata2: hard resetting link > > Mar 22 07:56:03 riva kernel: ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310) > > Mar 22 07:56:03 riva kernel: ata2.00: configured for UDMA/33 > > Mar 22 07:56:03 riva kernel: ata2: EH complete > > Mar 22 07:56:03 riva kernel: SCSI device sdb: 1465149168 512-byte hdwr sectors (750156 MB) > > Mar 22 07:56:03 riva kernel: sdb: Write Protect is off > > Mar 22 07:56:03 riva kernel: SCSI device sdb: drive cache: write back I had a lot of problems about a year ago with a client machine, some of it discussed on this list. Long story short: the end result was that the hardware at each end of the SATA link was negotiating too high a speed, this resulted in occasional errors, most of which recovered (after longish delays) and then a device would go off line. It was fixed by the h/ware people inserting a tiny jumper on the disks to cap the data rate ... the disks have been well behaved since. I don't know if this is what is happening here. -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php Past chairman of UKUUG: http://www.ukuug.org/ #include From S.Tohill at westminster.ac.uk Mon Mar 22 21:47:00 2010 From: S.Tohill at westminster.ac.uk (sean) Date: Mon, 22 Mar 2010 21:47:00 +0000 Subject: [Gllug] OT Robin Milner Message-ID: <4BA7E554.1030906@wmin.ac.uk> Robin Milner died on Saturday. Not linux related i know, but he was a great computer scientist who contributed much to the area. regards sean -- The University of Westminster is a charity and a company limited by guarantee. Registration number: 977818 England. Registered Office: 309 Regent Street, London W1B 2UW, UK. From rich at annexia.org Mon Mar 22 21:54:37 2010 From: rich at annexia.org (Richard Jones) Date: Mon, 22 Mar 2010 21:54:37 +0000 Subject: [Gllug] Is this a RAM error? Message-ID: <20100322215437.GA7905@annexia.org> This is a "machine check error", the output of mcelog. It's a bit confusing. Can someone confirm that this is a RAM problem? Rich. -- Richard Jones Red Hat -------------- next part -------------- MCE 0 Mon Mar 22 19:26:21 2010 HARDWARE ERROR. This is *NOT* a software problem! Please contact your hardware vendor CPU 3 BANK 3 ADDR 114ea4470 MCG status: MCi status: Error enabled MCi_ADDR register valid MCA: Generic CACHE Level-2 Generic Error STATUS 942001080c01010a MCGSTATUS 0 CPUID Vendor Intel Family 6 Model 15 No DIMM found for 114ea4470 in SMBIOS MCE 1 Mon Mar 22 19:26:59 2010 HARDWARE ERROR. This is *NOT* a software problem! Please contact your hardware vendor CPU 3 BANK 3 ADDR 1244e4b68 MCG status: MCi status: Error enabled MCi_ADDR register valid MCA: Generic CACHE Level-2 Generic Error STATUS 942000c80e01010a MCGSTATUS 0 CPUID Vendor Intel Family 6 Model 15 No DIMM found for 1244e4b68 in SMBIOS MCE 2 Mon Mar 22 19:27:17 2010 HARDWARE ERROR. This is *NOT* a software problem! Please contact your hardware vendor CPU 3 BANK 3 MCG status: MCi status: Error enabled MCA: corrected filtering (some unreported errors in same region) Level-2 Generic memory hierarchy error STATUS 902000481020100e MCGSTATUS 0 CPUID Vendor Intel Family 6 Model 15 MCE 3 Mon Mar 22 19:27:27 2010 HARDWARE ERROR. This is *NOT* a software problem! Please contact your hardware vendor CPU 3 BANK 3 MCG status: MCi status: Error enabled MCA: corrected filtering (some unreported errors in same region) Level-2 Generic memory hierarchy error STATUS 902000081120100e MCGSTATUS 0 CPUID Vendor Intel Family 6 Model 15 MCE 4 Mon Mar 22 19:28:00 2010 HARDWARE ERROR. This is *NOT* a software problem! Please contact your hardware vendor CPU 3 BANK 3 MCG status: MCi status: Error enabled MCA: corrected filtering (some unreported errors in same region) Level-2 Generic memory hierarchy error STATUS 902000081220100e MCGSTATUS 0 CPUID Vendor Intel Family 6 Model 15 From rich at annexia.org Mon Mar 22 21:56:30 2010 From: rich at annexia.org (Richard Jones) Date: Mon, 22 Mar 2010 21:56:30 +0000 Subject: [Gllug] OT Robin Milner In-Reply-To: <4BA7E554.1030906@wmin.ac.uk> References: <4BA7E554.1030906@wmin.ac.uk> Message-ID: <20100322215630.GB7905@annexia.org> On Mon, Mar 22, 2010 at 09:47:00PM +0000, sean wrote: > Robin Milner died on Saturday. Not linux related i know, but he was a > great computer scientist who contributed much to the area. He coinvented type inference about 30 years ago. Even now this most useful technique is not widely used, which is a great shame ... Rich. -- Richard Jones Red Hat From adrian at newgolddream.dyndns.info Mon Mar 22 22:53:55 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Mon, 22 Mar 2010 22:53:55 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <4BA7465F.9030103@shihad.org> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> Message-ID: <1269298435.9762.2.camel@bossclass> On Mon, 2010-03-22 at 10:28 +0000, Iain M Conochie wrote: > Adrian McMenamin wrote: > > > > > > > The bad news is that, having bought a high spec Mobo (Asus P7H55-M Pro) > > for them, it seems unwilling to recognise their old 200G IDE (PATA) > > disk, though it will recognise the DVD drive on the same cable. > > > > Have you tried the disk on the cable by itself? Most new machines expect > only one drive on the PATA cables, and also usually expect it to be a CD > / DVD ROM device. There may be BIOS features to turn off this annoyance > but I have not reaaly found any > Thanks for all the help guys. It was a cable/jumper problem. As is suggested above the mobo was rather more pernickety about this sort of thing that the old socket 478 job it was replacing. Now I have installed Ubuntu on the box but it freezes when I do more or less anything in X. I fear I am in for a long evening tomorrow as I struggle to get the thing working. Oh well From hearnsj at googlemail.com Mon Mar 22 23:42:08 2010 From: hearnsj at googlemail.com (John Hearns) Date: Mon, 22 Mar 2010 23:42:08 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269298435.9762.2.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> Message-ID: <9f8092cc1003221642m2c873227hdd4dec5306b81eaa@mail.gmail.com> On 22 March 2010 22:53, Adrian McMenamin wrote: > Now I have installed Ubuntu on the box but it freezes when I do more or > less anything in X. I fear I am in for a long evening tomorrow as I > struggle to get the thing working. Oh well Do you want your children to grow up as weak, lily-livered slaves of Bill? Why would they need X anyway? Treat your children to long, invigorating cross-country runs, cold showers in the morning, and a healthy dose of lynx, irssi, mutt and vi. They will thank you in the end. From gvimrc at googlemail.com Tue Mar 23 00:13:51 2010 From: gvimrc at googlemail.com (gvim) Date: Tue, 23 Mar 2010 00:13:51 +0000 Subject: [Gllug] Mercurial or Subversion for single user? Message-ID: <4BA807BF.5050106@googlemail.com> I'm looking to start using a revision control system for my work which usually involves: - Work at home on code project - rsync to laptop - Take project away on laptop at weekend - Return and rsync back to desktop Subversion, Git and Mercurial seem to be the most popular but a cursory glance at Git suggests it's well beyond what I need so that leaves Subversion and Mercurial. Subversion is client/server whereas Mercurial is distributed but the distinction may not be relevant in a single-user scenario. I read Subversion has/had some renaming issues but didn't delve into it too deeply. So, what would be the best option. I'm inclined to go with Mercurial at this stage. gvim From martin at hinterlands.org Tue Mar 23 00:28:08 2010 From: martin at hinterlands.org (Martin A. Brooks) Date: Tue, 23 Mar 2010 00:28:08 -0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA807BF.5050106@googlemail.com> References: <4BA807BF.5050106@googlemail.com> Message-ID: On Tue, March 23, 2010 00:13, gvim wrote: > So, what would be the best option. I'm inclined to go with Mercurial at > this stage. By far the best option is to use whichever of these suits your needs best. I strongly recommend using the piece of software that works most appropriately. From gvimrc at googlemail.com Tue Mar 23 00:42:10 2010 From: gvimrc at googlemail.com (gvim) Date: Tue, 23 Mar 2010 00:42:10 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: References: <4BA807BF.5050106@googlemail.com> Message-ID: <4BA80E62.9020103@googlemail.com> On 23/03/2010 00:28, Martin A. Brooks wrote: > On Tue, March 23, 2010 00:13, gvim wrote: >> So, what would be the best option. I'm inclined to go with Mercurial at >> this stage. > > By far the best option is to use whichever of these suits your needs best. > > I strongly recommend using the piece of software that works most > appropriately. > That's why I asked. I have no experience with either so wanted to know if one is more suited to a single-user scenario which isn't addressed in the literature. gvim From martin at hinterlands.org Tue Mar 23 01:12:42 2010 From: martin at hinterlands.org (Martin A. Brooks) Date: Tue, 23 Mar 2010 01:12:42 -0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA80E62.9020103@googlemail.com> References: <4BA807BF.5050106@googlemail.com> <4BA80E62.9020103@googlemail.com> Message-ID: <5a61ad50a5baee0df251d73dda697635.squirrel@olga.hinterlands.org> On Tue, March 23, 2010 00:42, gvim wrote: > On 23/03/2010 00:28, Martin A. Brooks wrote: >> I strongly recommend using the piece of software that works most >> appropriately. >> > > That's why I asked. I have no experience with either so wanted to know if > one is more suited to a single-user scenario which isn't addressed in the > literature. I may be being too simplistic, but have you considered simply trying out each piece of software for a period if time? What works perfectly for me may be utterly useless for you. I can tell you "t3h git r0x!! hg su> References: <4BA807BF.5050106@googlemail.com> <4BA80E62.9020103@googlemail.com> <5a61ad50a5baee0df251d73dda697635.squirrel@olga.hinterlands.org> Message-ID: <4BA8161A.1090203@googlemail.com> On 23/03/2010 01:12, Martin A. Brooks wrote: > On Tue, March 23, 2010 00:42, gvim wrote: > > I may be being too simplistic, but have you considered simply trying out > each piece of software for a period if time? > > What works perfectly for me may be utterly useless for you. I can tell > you "t3h git r0x!! hg su> I'd rather spend half the time learning the right one if one is more suited to a single-user scenario. If they're both the same in this respect then, yes, I may have to simply try both. gvim From itsbruce at workshy.org Tue Mar 23 04:40:52 2010 From: itsbruce at workshy.org (Bruce Richardson) Date: Tue, 23 Mar 2010 04:40:52 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269298435.9762.2.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> Message-ID: <20100323044052.GA5115@phaistos.bruce> On Mon, Mar 22, 2010 at 10:53:55PM +0000, Adrian wrote: > Now I have installed Ubuntu on the box but it freezes when I do more or > less anything in X. How powerful is your PSU? Did you upgrade the CPU along with the mobo? One possibility is that when you go into X, it overtaxes the system - altbhough I'd expect it to shut down rather than just freeze, if that were the issue. The board you chose is designed for overclockers, by the way. One thing to know about ASUS's range of overclocking boards is that they can be very fussy about the memory you install on them; it's worth researching them a bit to see what the user community has found that works best. -- Bruce What would Edward Woodward do? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100323/2894f1e7/attachment-0001.pgp From itsbruce at workshy.org Tue Mar 23 05:23:55 2010 From: itsbruce at workshy.org (Bruce Richardson) Date: Tue, 23 Mar 2010 05:23:55 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA807BF.5050106@googlemail.com> References: <4BA807BF.5050106@googlemail.com> Message-ID: <20100323052355.GA5228@phaistos.bruce> On Tue, Mar 23, 2010 at 12:13:51AM +0000, gvim wrote: > Subversion, Git and Mercurial seem to be the most popular but a > cursory glance at Git suggests it's well beyond what I need so that > leaves Subversion and Mercurial. Subversion is client/server whereas > Mercurial is distributed but the distinction may not be relevant in a > single-user scenario. Quite. Until you're sharing a repository with others, it really isn't going to affect you. > I read Subversion has/had some renaming issues > but didn't delve into it too deeply. It's not recommended to rename and modify a file in the same commit. So don't do that. There are some edge cases that can bite you when merging branches - I'm guessing you won't be doing much of that yet. > > So, what would be the best option. I'm inclined to go with Mercurial > at this stage. Your (currently) modest requirements are suited by either. If you work with free software much, you're going to encounter both. Subversion is more commonly used, if that makes any difference. On the other hand, Mercurial is probably a little easier to set up, initially, and doesn't litter your filesystem with hidden directories (subversion hides status information in .svn directories which it adds to any directories you check out). Both have subtle issues waiting to bite the unwary regarding tags and branches. In the long run, though, you'll want to become familiar with more than one VCS and both of your candidates are functional. If you toss a coin to choose between them, it won't do you any harm. -- Bruce I unfortunately do not know how to turn cheese into gold. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100323/88bf52bb/attachment.pgp From chrisbell at 3966.ukfsn.org Tue Mar 23 07:20:14 2010 From: chrisbell at 3966.ukfsn.org (Chris Bell) Date: Tue, 23 Mar 2010 07:20:14 +0000 (GMT) Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269298435.9762.2.camel@bossclass> Message-ID: On Mon 22 Mar, Adrian McMenamin wrote: > > > Now I have installed Ubuntu on the box but it freezes when I do more or > less anything in X. I fear I am in for a long evening tomorrow as I > struggle to get the thing working. Oh well > I had a problem with a total hang when starting X during the boot sequence with recent Debian kernels which did not happen when I selected an earlier kernel, sorted by adding more RAM. -- Chris Bell www.chrisbell.org.uk (was www.overview.demon.co.uk) Microsoft sells you Windows ... Linux gives you the whole house. From andrew at osmosoft.com Tue Mar 23 07:31:58 2010 From: andrew at osmosoft.com (Andrew Back) Date: Tue, 23 Mar 2010 07:31:58 +0000 Subject: [Gllug] Event: Open Source in Telecommunications - 22/04. Message-ID: <20100323073158.GC18604@rhys.osmosoft.com> Hello, The British Computer Society Open Source Specialist Group and BT are organising an event on open source in telecommunications, on Thursday 22nd April at the BCS offices in Covent Garden. This is free to attend and open to all. Talks include a presentation on Vyatta - an open source IP routing platform and alternative to the likes of Cisco, LiMo Foundation, OpenNMS, Asterisk, and one that as a radio geek I'm particularly looking forward to, OpenBTS (open source GSM platform) and its use in a pilot on the pacific island of Niue. The full agenda and registration details are available at: http://is.gd/aTVRu Cheers, Andrew -- Andrew Back mailto:andrew at osmosoft.com http://carrierdetect.com From adrian at newgolddream.dyndns.info Tue Mar 23 08:29:02 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Tue, 23 Mar 2010 08:29:02 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <20100323044052.GA5115@phaistos.bruce> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> <20100323044052.GA5115@phaistos.bruce> Message-ID: <1269332942.5512.3.camel@bossclass> On Tue, 2010-03-23 at 04:40 +0000, Bruce Richardson wrote: > On Mon, Mar 22, 2010 at 10:53:55PM +0000, Adrian wrote: > > Now I have installed Ubuntu on the box but it freezes when I do more or > > less anything in X. > > How powerful is your PSU? Did you upgrade the CPU along with the mobo? > One possibility is that when you go into X, it overtaxes the system - > altbhough I'd expect it to shut down rather than just freeze, if that > were the issue. I got a new PSU and memory for the board too. I haven't really taken over anything except the case, the DVD drive and the PATA disk. I had problems booting into Linux off the install cd until I fiddled with the kernel command line, so I suspect that is where I will have to start on the installed kernel. I deliberated picked a fast board as the girls had moaned about how slow their old machine was. Hopefully this one will last them a few years... From adrian at newgolddream.dyndns.info Tue Mar 23 08:32:46 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Tue, 23 Mar 2010 08:32:46 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269332942.5512.3.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> <20100323044052.GA5115@phaistos.bruce> <1269332942.5512.3.camel@bossclass> Message-ID: <1269333166.5512.5.camel@bossclass> On Tue, 2010-03-23 at 08:29 +0000, Adrian McMenamin wrote: > > I had problems booting into Linux off the install cd until I fiddled > with the kernel command line, so I suspect that is where I will have to > start on the installed kernel. > Should add that XP appeared to be running ok before I wiped it - so I think the hardware all works and it's software only now. > I deliberated picked a fast board as the girls had moaned about how slow > their old machine was. Hopefully this one will last them a few years... > Deliberated? errr, deliberately From matthew.king at monnsta.net Tue Mar 23 09:10:42 2010 From: matthew.king at monnsta.net (Matthew King) Date: Tue, 23 Mar 2010 09:10:42 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA807BF.5050106@googlemail.com> (gvim's message of "Tue, 23 Mar 2010 00:13:51 +0000") References: <4BA807BF.5050106@googlemail.com> Message-ID: <87iq8nbdyl.fsf@knight.monnsta.net> gvim writes: > I'm looking to start using a revision control system for my work which usually involves: > > - Work at home on code project > - rsync to laptop > - Take project away on laptop at weekend > - Return and rsync back to desktop > > Subversion, Git and Mercurial seem to be the most popular but a > cursory glance at Git suggests it's well beyond what I need so that > leaves Subversion and Mercurial. Subversion is client/server whereas > Mercurial is distributed but the distinction may not be relevant in a > single-user scenario. I read Subversion has/had some renaming issues > but didn't delve into it too deeply. > > So, what would be the best option. I'm inclined to go with Mercurial > at this stage. If you've already discounted git, then stick with Mercurial. Subversion is awful. And before any subversion fans give me grief, Linus says Subversion is awful and Linus is God, therefore subversion is awful. Right? Personally I'd go with git though because 1) git is simple, 2) it has a nice name, and 3) it's what I know. Git has its own method for sharing code among multiple locations, but if you're careful only to develop on the latest version (which you'd have to be with any RCS anyway) then rsync is suitable. Personally I'd use git's own code sharing mechanisms because I don't trust myself to remember to rsync when I'm supposed to but YMMV. That said, there are only a few necessary commands: git init (Start. Only used once.) git commit - git commit -a Commit everything - git commit -m "foo" Don't ask you to edit a commit message - use foo That's the bare minimum, really. Other useful commands: git mv foo bar Analog of mv foo bar git rm foo Analog of, you guessed it, rm foo git add bar Opposite of git rm git status and git diff are also a handy way to see what's going on. Then there's branches, rollbacks, partial commits, etc. Still simple but by no means essential. gitk is also an immeasurably useful tool to provides a visual indication of your repository's history. Matthew -- I must take issue with the term "a mere child", for it has been my invariable experience that the company of a mere child is infinitely preferable to that of a mere adult. -- Fran Lebowitz From sta296 at astradyne.co.uk Tue Mar 23 09:17:13 2010 From: sta296 at astradyne.co.uk (Tethys) Date: Tue, 23 Mar 2010 09:17:13 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: Your message of "Tue, 23 Mar 2010 09:10:42 GMT." <87iq8nbdyl.fsf@knight.monnsta.net> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> Message-ID: <201003230917.o2N9HD5m005656@leto.astradyne.corp> -------- Matthew King writes: >If you've already discounted git, then stick with Mercurial. Subversion >is awful. And before any subversion fans give me grief, Linus says >Subversion is awful and Linus is God, therefore subversion is awful. > >Right? Yes. I was beginning to wonder if anyone was going to get off the fence and make a firm recommendation. Out of those two, Mercurial is the clear winner. >Personally I'd go with git though because 1) git is simple, 2) it has a >nice name, and 3) it's what I know. There we disagree. I like git. I use git. But I couldn't come close to calling it simple. It's simple until a merge fails due to conflicts. And then it's extremely unintuitive, particularly if you're merging into a dirty tree (which seems to be a common workflow for me). If you want simple, use darcs[1] Tet [1] Which has its own set of problems, but by and large is more suitable for a VCS novice. From andy at andymillar.co.uk Tue Mar 23 09:17:50 2010 From: andy at andymillar.co.uk (Andy Millar) Date: Tue, 23 Mar 2010 09:17:50 -0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: References: <4BA807BF.5050106@googlemail.com> Message-ID: <1269335870.6177.28.camel@millaralpt.co.tradefair> On Tue, 2010-03-23 at 00:28 +0000, Martin A. Brooks wrote: > By far the best option is to use whichever of these suits your needs best. > > I strongly recommend using the piece of software that works most > appropriately. I use subversion. A colleague swears by git. I suspect you can even find someone who swears by CVS. They all (broadly) do the same thing. Andy -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 5162 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100323/d0e09218/attachment-0001.bin From matthew.king at monnsta.net Tue Mar 23 10:10:18 2010 From: matthew.king at monnsta.net (Matthew King) Date: Tue, 23 Mar 2010 10:10:18 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <201003230917.o2N9HD5m005656@leto.astradyne.corp> (Tethys's message of "Tue, 23 Mar 2010 09:17:13 +0000") References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <201003230917.o2N9HD5m005656@leto.astradyne.corp> Message-ID: <87eijbbb79.fsf@knight.monnsta.net> Tethys writes: > Matthew King writes: > >>Personally I'd go with git though because 1) git is simple, 2) it has a >>nice name, and 3) it's what I know. > > There we disagree. I like git. I use git. But I couldn't come close > to calling it simple. It's simple until a merge fails due to conflicts. Perhaps I should have added the caveat that it's simple *for the specified use case*. I've had my own merging fun too... Matthew -- I must take issue with the term "a mere child", for it has been my invariable experience that the company of a mere child is infinitely preferable to that of a mere adult. -- Fran Lebowitz From indigojo at blogistan.co.uk Tue Mar 23 09:50:11 2010 From: indigojo at blogistan.co.uk (Matthew Smith) Date: Tue, 23 Mar 2010 09:50:11 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA807BF.5050106@googlemail.com> References: <4BA807BF.5050106@googlemail.com> Message-ID: <4BA88ED3.9020306@blogistan.co.uk> On 23/03/2010 00:13, gvim wrote: > I'm looking to start using a revision control system for my work which usually involves: > > - Work at home on code project > - rsync to laptop > - Take project away on laptop at weekend > - Return and rsync back to desktop > > Subversion, Git and Mercurial seem to be the most popular but a cursory glance at Git suggests it's well beyond what I need so that leaves Subversion and Mercurial. Subversion is client/server whereas Mercurial is distributed but the distinction may not be relevant in a single-user scenario. I read Subversion has/had some renaming issues but didn't delve into it too deeply. > > So, what would be the best option. I'm inclined to go with Mercurial at this stage. I've been using Mercurial (Hg) for my own project (QTM) since about 2008 I think (before that, I used SVN). It's very easy to transfer from one to another, but the biggest advantage is that the repositories are very lightweight and you aren't dependent on any central server. You can use branches, for example, on one copy of the project without it affecting the master repo, which you can't do on SVN (actually, some people use Git or Hg to do that with SVN). With SVN, tags and branches are based on copying the entire source to a separate directory, which doesn't happen in Hg (or any other distributed VCS); tags in particular are just that: names on a particular change commit. Hg also has good online documentation. I would recommend it over SVN for new projects. Regards, Matt Smith -- http://www.blogistan.co.uk/qt/ From itsbruce at workshy.org Tue Mar 23 10:27:55 2010 From: itsbruce at workshy.org (Bruce Richardson) Date: Tue, 23 Mar 2010 10:27:55 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA88ED3.9020306@blogistan.co.uk> References: <4BA807BF.5050106@googlemail.com> <4BA88ED3.9020306@blogistan.co.uk> Message-ID: <20100323102755.GA5657@phaistos.bruce> On Tue, Mar 23, 2010 at 09:50:11AM +0000, Matthew wrote: > branches, for example, on one copy of the project without it affecting > the master repo, which you can't do on SVN (actually, some people use > Git or Hg to do that with SVN). There is also svk http://svk.bestpractical.com/view/HomePage -- Bruce Hierophant: someone who remembers, when you are on the way down, everything you did to them on the way up. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100323/5d7ebb7c/attachment.pgp From dave at dave.org.uk Tue Mar 23 10:34:22 2010 From: dave at dave.org.uk (Dave Cross) Date: Tue, 23 Mar 2010 10:34:22 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100323102755.GA5657@phaistos.bruce> References: <4BA807BF.5050106@googlemail.com> <4BA88ED3.9020306@blogistan.co.uk> <20100323102755.GA5657@phaistos.bruce> Message-ID: <4BA8992E.3020405@dave.org.uk> On 03/23/2010 10:27 AM, Bruce Richardson wrote: > On Tue, Mar 23, 2010 at 09:50:11AM +0000, Matthew wrote: >> branches, for example, on one copy of the project without it affecting >> the master repo, which you can't do on SVN (actually, some people use >> Git or Hg to do that with SVN). > > There is also svk http://svk.bestpractical.com/view/HomePage Which was end-of-lifed almost a year ago. http://blog.bestpractical.com/2009/05/the-future-of-svk.html Dave... From martin at hinterlands.org Tue Mar 23 11:25:25 2010 From: martin at hinterlands.org (Martin A. Brooks) Date: Tue, 23 Mar 2010 11:25:25 -0000 Subject: [Gllug] [OTish] Social in Manchester today Message-ID: Hi On the offchance some GLLUGers are in or around Manchester today, please consider yourselves invited to attend the UKUUG Social this evening. http://spring2010.ukuug.org/SocialProgramme I am (with my http://antibodyMX.net/ hat on) buying the first £150 of drinks. Hope to see some of you there Mart. From proggaprogga at gmail.com Tue Mar 23 12:33:17 2010 From: proggaprogga at gmail.com (Progga) Date: Tue, 23 Mar 2010 12:33:17 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA807BF.5050106@googlemail.com> References: <4BA807BF.5050106@googlemail.com> Message-ID: <20100323123317.GA6706@ju.localhost> On Tue, Mar 23, 2010 at 12:13:51AM +0000, gvim wrote: > So, what would be the best option. I'm inclined to go with Mercurial at this stage. I use both and won't say that hg is way more useful than svn. People who do lots of merging really like hg, but then I don't do lots of merging :-) If you do want to go for hg at the end then check Joel Spolsky's hg tutorial at http://hginit.com From general_email at technicalbloke.com Tue Mar 23 13:56:35 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Tue, 23 Mar 2010 13:56:35 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100323123317.GA6706@ju.localhost> References: <4BA807BF.5050106@googlemail.com> <20100323123317.GA6706@ju.localhost> Message-ID: <4BA8C893.8000301@technicalbloke.com> Progga wrote: > On Tue, Mar 23, 2010 at 12:13:51AM +0000, gvim wrote: > > >> So, what would be the best option. I'm inclined to go with Mercurial at this stage. >> > > I use both and won't say that hg is way more useful than svn. People who do > lots of merging really like hg, but then I don't do lots of merging :-) > > If you do want to go for hg at the end then check Joel Spolsky's hg tutorial at > http://hginit.com > > > SVN is fine for a single user as long as you don't want to be branching and merging a lot. Linus' big complaint is how badly it scales and that's not a factor for you. The thing it had that the others didn't a couple of years back was good integration with Eclipse but maybe that has changed now too. Roger. From general_email at technicalbloke.com Tue Mar 23 14:03:15 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Tue, 23 Mar 2010 14:03:15 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <1269333166.5512.5.camel@bossclass> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> <20100323044052.GA5115@phaistos.bruce> <1269332942.5512.3.camel@bossclass> <1269333166.5512.5.camel@bossclass> Message-ID: <4BA8CA23.1050805@technicalbloke.com> Adrian McMenamin wrote: > On Tue, 2010-03-23 at 08:29 +0000, Adrian McMenamin wrote: > > > >> I had problems booting into Linux off the install cd until I fiddled >> with the kernel command line, so I suspect that is where I will have to >> start on the installed kernel. >> >> Have you tried any other distro's live CDs? That might help you narrow down what it's choking on and whether it's kernel or distro specific. Roger. From richardlewis at fastmail.co.uk Tue Mar 23 14:14:54 2010 From: richardlewis at fastmail.co.uk (Richard Lewis) Date: Tue, 23 Mar 2010 14:14:54 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <87iq8nbdyl.fsf@knight.monnsta.net> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> Message-ID: <878w9j6s69.wl%richard.lewis@gold.ac.uk> At Tue, 23 Mar 2010 09:10:42 +0000, Matthew King wrote: > > 1) git is simple, Huh? > 2) it has a nice name, Hehe. > and 3) it's what I know. Haha! -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Richard Lewis ISMS, Computing Goldsmiths, University of London Tel: +44 (0)20 7078 5134 Skype: richardjlewis JID: ironchicken at jabber.earth.li http://www.richard-lewis.me.uk/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- +----------------------------------------------+ | Support open access to scholarship | | http://freeculture.org/ http://www.doaj.org/ | +----------------------------------------------+ From richardlewis at fastmail.co.uk Tue Mar 23 14:15:59 2010 From: richardlewis at fastmail.co.uk (Richard Lewis) Date: Tue, 23 Mar 2010 14:15:59 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <201003230917.o2N9HD5m005656@leto.astradyne.corp> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <201003230917.o2N9HD5m005656@leto.astradyne.corp> Message-ID: <877hp36s4g.wl%richard.lewis@gold.ac.uk> At Tue, 23 Mar 2010 09:17:13 +0000, Tethys wrote: > > If you want simple, use darcs[1] > +1 for darcs. My colleagues can even get undergraduates to use it. -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Richard Lewis ISMS, Computing Goldsmiths, University of London Tel: +44 (0)20 7078 5134 Skype: richardjlewis JID: ironchicken at jabber.earth.li http://www.richard-lewis.me.uk/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- +----------------------------------------------+ | Support open access to scholarship | | http://freeculture.org/ http://www.doaj.org/ | +----------------------------------------------+ From rich at annexia.org Tue Mar 23 16:30:26 2010 From: rich at annexia.org (Richard Jones) Date: Tue, 23 Mar 2010 16:30:26 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <1269335870.6177.28.camel@millaralpt.co.tradefair> References: <4BA807BF.5050106@googlemail.com> <1269335870.6177.28.camel@millaralpt.co.tradefair> Message-ID: <20100323163026.GA8103@annexia.org> On Tue, Mar 23, 2010 at 09:17:50AM -0000, Andy Millar wrote: > On Tue, 2010-03-23 at 00:28 +0000, Martin A. Brooks wrote: > > By far the best option is to use whichever of these suits your needs best. > > > > I strongly recommend using the piece of software that works most > > appropriately. > > I use subversion. > > A colleague swears by git. > > I suspect you can even find someone who swears by CVS. > > They all (broadly) do the same thing. This is simply not true any more. Git's rebasing features are well beyond what CVS is capable of. Rich. -- Richard Jones Red Hat From rich at annexia.org Tue Mar 23 16:31:45 2010 From: rich at annexia.org (Richard Jones) Date: Tue, 23 Mar 2010 16:31:45 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA807BF.5050106@googlemail.com> References: <4BA807BF.5050106@googlemail.com> Message-ID: <20100323163145.GB8103@annexia.org> On Tue, Mar 23, 2010 at 12:13:51AM +0000, gvim wrote: > So, what would be the best option. I'm inclined to go with Mercurial > at this stage. Git, basically, although it's a little harder to learn than Mercurial. Run a mile from Subversion. Rich. -- Richard Jones Red Hat From rich at annexia.org Tue Mar 23 16:40:39 2010 From: rich at annexia.org (Richard Jones) Date: Tue, 23 Mar 2010 16:40:39 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <878w9j6s69.wl%richard.lewis@gold.ac.uk> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <878w9j6s69.wl%richard.lewis@gold.ac.uk> Message-ID: <20100323164039.GC8103@annexia.org> On Tue, Mar 23, 2010 at 02:14:54PM +0000, Richard Lewis wrote: > At Tue, 23 Mar 2010 09:10:42 +0000, > Matthew King wrote: > > > > 1) git is simple, > > Huh? git's underlying concept is simple. It's highly unintuitive on top of that until you understand what's going on at the low level. I'd recommend reading this: http://lwn.net/Articles/145194/ Then do: mkdir test cd test git init touch README git add README git commit -m "Add README" cd .git cat HEAD cat refs/heads/master find -name objects (Note that the object files themselves are zlib-compressed so you can't read them directly. See: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=509942 for more information) Rich. -- Richard Jones Red Hat From mick at dcs.bbk.ac.uk Tue Mar 23 18:46:04 2010 From: mick at dcs.bbk.ac.uk (Mick Farmer) Date: Tue, 23 Mar 2010 18:46:04 +0000 Subject: [Gllug] Netgear DG834G ADSL Router Message-ID: <201003231846.o2NIk4nA000994@zeus.dcs.bbk.ac.uk> Dear GLLUGers, I've been using one of these for a number of years and it's been rock steady. However, I have a query about its connection to my private LAN. There's a display called "Attached Devices". Most of the time it shows nothing. After a number of Refresh hits it will show my main computer which has a fixed IP address. This can disappear on subsequent Refresh hits. It never shows my other machines which use DHCP for their addresses. Does it treat them differently? I can't make contact with the other machines on the LAN. Not even ping works. Does this imply that ARP is not implemented on the router? Could this be because I'm using wireless connections and not the Ethernet ports? Any guidance much appreciated. Regards, Mick /"\ \ / Linux Registered X ASCII Ribbon Campaign User #287765 / \ Against HTML Mail From adrian at newgolddream.dyndns.info Tue Mar 23 20:13:34 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Tue, 23 Mar 2010 20:13:34 -0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <4BA8CA23.1050805@technicalbloke.com> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> <20100323044052.GA5115@phaistos.bruce> <1269332942.5512.3.camel@bossclass> <1269333166.5512.5.camel@bossclass> <4BA8CA23.1050805@technicalbloke.com> Message-ID: On Tue, March 23, 2010 2:03 pm, general_email at technicalbloke.com wrote: > Adrian McMenamin wrote: >> On Tue, 2010-03-23 at 08:29 +0000, Adrian McMenamin wrote: >> >> >> >>> I had problems booting into Linux off the install cd until I fiddled >>> with the kernel command line, so I suspect that is where I will have to >>> start on the installed kernel. >>> >>> > > Have you tried any other distro's live CDs? That might help you narrow > down what it's choking on and whether it's kernel or distro specific. > First guess was noapic - didn't work Second was acpi=off which worked Though whether it will work for the update kernel I am now loading, I don't yet know From john at sinodun.org.uk Tue Mar 23 20:21:30 2010 From: john at sinodun.org.uk (John Winters) Date: Tue, 23 Mar 2010 20:21:30 +0000 Subject: [Gllug] Netgear DG834G ADSL Router In-Reply-To: <201003231846.o2NIk4nA000994@zeus.dcs.bbk.ac.uk> References: <201003231846.o2NIk4nA000994@zeus.dcs.bbk.ac.uk> Message-ID: <4BA922CA.5020405@sinodun.org.uk> Mick Farmer wrote: > Dear GLLUGers, > > I've been using one of these for a number of years and it's > been rock steady. > However, I have a query about its connection to my private > LAN. There's a display called "Attached Devices". Most of > the time it shows nothing. After a number of Refresh hits > it will show my main computer which has a fixed IP address. > This can disappear on subsequent Refresh hits. Mine seems a little vague on this front too. To start with it doesn't show everything, then after a refresh it does. The machine name for each device seems to be derived from Windows file sharing. Machines running Samba show a name; machines not running Samba are "UNKNOWN". > It never shows my other machines which use DHCP for their > addresses. Does it treat them differently? All my machines except one use DHCP. The one which doesn't is the DHCP server. They all show up in the list. Some are wired and some use WiFi. > I can't make contact with the other machines on the LAN. > Not even ping works. Does this imply that ARP is not > implemented on the router? Do you mean none of the machines on the LAN can communicate with any of the others? If so it doesn't sound like you have a working LAN. If not, is there a particular machine which can't see any others? AFAIK, to your LAN the Netgear box functions just like a switch, which IIRC means it doesn't do ARP at all. I've had a couple of glasses of wine, and it's been a while but as I remember it the ARP protocol runs between the two communicating computers. The switch's job is just to pass on packets as appropriate. Please clarify your statement that your machines can't contact each other. Cheers, John From james.dutton at gmail.com Tue Mar 23 20:52:41 2010 From: james.dutton at gmail.com (James Courtier-Dutton) Date: Tue, 23 Mar 2010 20:52:41 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <4BA807BF.5050106@googlemail.com> References: <4BA807BF.5050106@googlemail.com> Message-ID: On 23 March 2010 00:13, gvim wrote: > I'm looking to start using a revision control system for my work which usually involves: > > - Work at home on code project > - rsync to laptop > - Take project away on laptop at weekend > - Return and rsync back to desktop > > Subversion, Git and Mercurial seem to be the most popular but a cursory glance at Git suggests it's well beyond what I need so that leaves Subversion and Mercurial. Subversion is client/server whereas Mercurial is distributed but the distinction may not be relevant in a single-user scenario. I read Subversion has/had some renaming issues but didn't delve into it too deeply. > > So, what would be the best option. I'm inclined to go with Mercurial at this stage. > They are all aimed at a slightly different audience. svn is old now, and therefore works well with tools like eclipse and windows. svn is a good improvement over cvs. svn is client/server model. git and mercurial are aimed at a more collaborative audience with multiple developers all contributing so they have better methods for resolving conflicts. mercurial is aimed more at the developer, whereas git is aimed more at the manager. So, I think mercurial is probably your best bet. The major bonus that git gives is it lets a manager pick and choose patches from a large pool of potential patches whereas, for example svn, one has to patch in the correct order or things go wrong. Merging is much more difficult with svn than git. I have used all three a lot, and tend to prefer git now, but only because a majority of the people I work with also use git. git does have a very nice simple integrity assurance method. The filename of the patch is the sha1 hash of the patch itself. So, it is very easy to ensure that no bit flips have happened anywhere in the code base or that some email program has modified the line feeds of the patch. I don't think there is a lot of difference really with both git and mercurial. I watched Linus and Matt side by side challenging each other to do certain tasks. git was quicker at some things, hg was quicker at others. But if I had to choose between svn and mercurial, hg would win even for a single developer project. So, in summary, I would support your choice of using HG. Kind Regards James From nix at esperi.org.uk Wed Mar 24 00:16:28 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:16:28 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <9f8092cc1003221642m2c873227hdd4dec5306b81eaa@mail.gmail.com> (John Hearns's message of "Mon, 22 Mar 2010 23:42:08 +0000") References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> <9f8092cc1003221642m2c873227hdd4dec5306b81eaa@mail.gmail.com> Message-ID: <87wrx2wp43.fsf@spindle.srvr.nix> On 22 Mar 2010, John Hearns told this: > showers in the morning, and a healthy dose of lynx, irssi, mutt and > vi. They will thank you in the end. vi? You use vi? I use sed, and for complex jobs 'emacs -batch'. Anything else leads to moral decay and the marriage of men and dogs. From nix at esperi.org.uk Wed Mar 24 00:18:06 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:18:06 +0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: (Adrian McMenamin's message of "Tue, 23 Mar 2010 20:13:34 -0000") References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> <20100323044052.GA5115@phaistos.bruce> <1269332942.5512.3.camel@bossclass> <1269333166.5512.5.camel@bossclass> <4BA8CA23.1050805@technicalbloke.com> Message-ID: <87sk7qwp1d.fsf@spindle.srvr.nix> On 23 Mar 2010, Adrian McMenamin said: > Second was acpi=off which worked ACPI really really should work on a modern board: with it off a lot of things won't work, from temperature regulation and the power button to (on some boards) even things like CD-ROM ejecting and the sound card (!) Maybe a BIOS upgrade is in order? From nix at esperi.org.uk Wed Mar 24 00:24:57 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:24:57 +0000 Subject: [Gllug] ATA errors In-Reply-To: <201003222049.o2MKnTaI003340@leto.astradyne.corp> (Tethys's message of "Mon, 22 Mar 2010 20:49:29 +0000") References: <201003222049.o2MKnTaI003340@leto.astradyne.corp> Message-ID: <87ociewopy.fsf@spindle.srvr.nix> On 22 Mar 2010, Tethys outgrape: > > My backup failed last night, because the device was read only: > >> building file list ... done >> rsync: delete_file: unlink "/backup/local/shared/camera/2009-08-24--foq/img_0507.jpg" failed: Read-only file system (30) >> rsync: delete_file: unlink "/backup/local/shared/camera/2009-08-24--foq/.xvpics/img_0507.jpg" failed: Read-only file system (30) > > That's never a good sign. Looking through the logs, I see a number of > ATA errors, starting with a timeout, a device error, and a bunch of > HSM violations. A few examples: From nix at esperi.org.uk Wed Mar 24 00:26:01 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:26:01 +0000 Subject: [Gllug] OT Robin Milner In-Reply-To: <4BA7E554.1030906@wmin.ac.uk> (sean's message of "Mon, 22 Mar 2010 21:47:00 +0000") References: <4BA7E554.1030906@wmin.ac.uk> Message-ID: <87k4t2woo6.fsf@spindle.srvr.nix> On 22 Mar 2010, sean outgrape: > Robin Milner died on Saturday. Not linux related i know, but he was a > great computer scientist who contributed much to the area. Oh, that's a shame :( From nix at esperi.org.uk Wed Mar 24 00:28:15 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:28:15 +0000 Subject: [Gllug] Is this a RAM error? In-Reply-To: <20100322215437.GA7905@annexia.org> (Richard Jones's message of "Mon, 22 Mar 2010 21:54:37 +0000") References: <20100322215437.GA7905@annexia.org> Message-ID: <87fx3qwokg.fsf@spindle.srvr.nix> On 22 Mar 2010, Richard Jones uttered the following: > This is a "machine check error", the output of mcelog. It's a bit > confusing. Can someone confirm that this is a RAM problem? It claims: > CPU 3 BANK 3 ADDR 114ea4470 > MCG status: > MCi status: > Error enabled > MCi_ADDR register valid > MCA: Generic CACHE Level-2 Generic Error > STATUS 942001080c01010a MCGSTATUS 0 ... to be an L2 cache problem. > CPUID Vendor Intel Family 6 Model 15 > No DIMM found for 114ea4470 in SMBIOS (that's not much of a surprise, these tables are as reliable as anything else written by BIOS vendors, i.e., utter shit). IIRC, anyway. I'm really not an expert in this area. Maybe the L2 cache error is reported for *all* RAM faults for all I know, but unless you have ECCRAM you're not going to get MCEs for faults in main memory... From nix at esperi.org.uk Wed Mar 24 00:30:00 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:30:00 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <1269335870.6177.28.camel@millaralpt.co.tradefair> (Andy Millar's message of "Tue, 23 Mar 2010 09:17:50 -0000") References: <4BA807BF.5050106@googlemail.com> <1269335870.6177.28.camel@millaralpt.co.tradefair> Message-ID: <87bpeewohj.fsf@spindle.srvr.nix> On 23 Mar 2010, Andy Millar outgrape: > I suspect you can even find someone who swears by CVS. ITYM 'at', HTH. > They all (broadly) do the same thing. Listen to Richard. He's right. git is somewhat arcane but can do astounding things. (I have too little experience with Mercurial to compare the two: whenever I try to use it I start out swearing at it and end up shooting my computer and burning down my coworkers). From caroline.ford.work at googlemail.com Wed Mar 24 00:35:22 2010 From: caroline.ford.work at googlemail.com (Caroline Ford) Date: Wed, 24 Mar 2010 00:35:22 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <87bpeewohj.fsf@spindle.srvr.nix> References: <4BA807BF.5050106@googlemail.com> <1269335870.6177.28.camel@millaralpt.co.tradefair> <87bpeewohj.fsf@spindle.srvr.nix> Message-ID: Well we (tuxpaint) still use CVS, FWIW. Caroline Sent from a mobile device. On 24 Mar 2010, at 00:30, Nix wrote: > On 23 Mar 2010, Andy Millar outgrape: >> I suspect you can even find someone who swears by CVS. > > ITYM 'at', HTH. > >> They all (broadly) do the same thing. > > Listen to Richard. He's right. git is somewhat arcane but can do > astounding things. (I have too little experience with Mercurial to > compare the two: whenever I try to use it I start out swearing at it > and > end up shooting my computer and burning down my coworkers). > -- > Gllug mailing list - Gllug at gllug.org.uk > http://lists.gllug.org.uk/mailman/listinfo/gllug From nix at esperi.org.uk Wed Mar 24 00:44:08 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:44:08 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <87iq8nbdyl.fsf@knight.monnsta.net> (Matthew King's message of "Tue, 23 Mar 2010 09:10:42 +0000") References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> Message-ID: <877hp2wntz.fsf@spindle.srvr.nix> On 23 Mar 2010, Matthew King stated: > git mv foo bar > Analog of mv foo bar > > git rm foo > Analog of, you guessed it, rm foo > > git add bar > Opposite of git rm No! No! No! It's not the opposite at *all*, and getting this wrong leads to great confusion down the road. It adds *the actual file*, i.e. the *actual content* to the 'index' (also known as the 'cache' or 'staging area', which is a sort of limbo files go into before commit. Change the file after 'git add' then do a 'git commit', and the thing you added, *without that change*, is what gets committed, because what gets added is what's in the index, and the index hasn't changed since you did a 'git add'. 'git commit -a' is the same as 'git add .' (which adds everything in the working tree which is in a file that git already knows about) followed by 'git commit'. The opposite of 'git add foo.c' is 'git reset -- foo.c'. I find this really useful. I can do lots of work in a big change, then split it into commits via a bunch of 'git add's followed by 'git commit's: and if I later decide they're in the wrong order or some are just wrong, I can use 'git commit --amend' and 'git rebase -i' to reshuffle and tweak them to my heart's content. (If the changes touch some of the same files, this won't work quite so easily, but 'git add -p' lets you add only *chunks* of changed files, similar to darcs.) > Then there's branches, rollbacks, partial commits, etc. Still simple but > by no means essential. gitk is also an immeasurably useful tool to > provides a visual indication of your repository's history. Yes indeed. (Though I'm not sure what rollbacks are, I don't think the git docs use the term. Do you mean 'git commit --amend', or detached heads?) (and yes, 'detached HEAD' *is* the single silliest name for an obvious concept in the entirety of the version control world.) From nix at esperi.org.uk Wed Mar 24 00:45:39 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:45:39 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100323164039.GC8103@annexia.org> (Richard Jones's message of "Tue, 23 Mar 2010 16:40:39 +0000") References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <878w9j6s69.wl%richard.lewis@gold.ac.uk> <20100323164039.GC8103@annexia.org> Message-ID: <8739zqwnrg.fsf@spindle.srvr.nix> On 23 Mar 2010, Richard Jones said: > (Note that the object files themselves are zlib-compressed so you > can't read them directly. You want 'git cat-file' in the (very rare) situation that you ever want to do that (outside scripts, anyway: git cat-file is quite useful there.) 'git show' is more likely to be useful in real-world situations. From nix at esperi.org.uk Wed Mar 24 00:50:36 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:50:36 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100323163145.GB8103@annexia.org> (Richard Jones's message of "Tue, 23 Mar 2010 16:31:45 +0000") References: <4BA807BF.5050106@googlemail.com> <20100323163145.GB8103@annexia.org> Message-ID: <87y6hiv8yr.fsf@spindle.srvr.nix> On 23 Mar 2010, Richard Jones verbalised: > Run a mile from Subversion. Run *screaming* from CVS. I'm scared whenever I do a commit in CVS (a thankfully rare event) and actually had nightmares about doing large branches in it, because if you do anything wrong (and it's so *easy* to do things wrong with branches) you're utterly screwed: you have to hand-edit the repository files, or restore from a backup, to get back to where you came from. With git, reversing a mistaken *anything* is one command away: git checkout '@{1 hour ago}' to get back to an earlier location, or git reset --hard 'HEAD^' or, perhaps git reset --hard 'HEAD^{4}' (or some other number) to make a clapped-out commit (or series of them) vanish. (As long as you haven't pushed them to other people, that is. Once you've done that, you have to live with your mistakes in the history and just 'git revert' them.) From nix at esperi.org.uk Wed Mar 24 00:53:09 2010 From: nix at esperi.org.uk (Nix) Date: Wed, 24 Mar 2010 00:53:09 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: (James Courtier-Dutton's message of "Tue, 23 Mar 2010 20:52:41 +0000") References: <4BA807BF.5050106@googlemail.com> Message-ID: <87tys6v8ui.fsf@spindle.srvr.nix> On 23 Mar 2010, James Courtier-Dutton uttered the following: > git does have a very nice simple integrity assurance method. The > filename of the patch is the sha1 hash of the patch itself. Git does not use patches (that's darcs, and the idea is, alas, broken: nobody has ever figured out a way to make a patch algebra work with merges). Git uses snapshots of the entire tree, chained together into a directed graph of commits over time. > So, it is very easy to ensure that no bit flips have happened anywhere > in the code base or that some email program has modified the line > feeds of the patch. More than that, because a commit includes the sha1sums of its parents and its tree, and a tree includes the sha1sums of all its component parts, each commit is a proof that the commit itself has not been modified and that *neither has its entire history*. From phil at hands.com Wed Mar 24 09:08:55 2010 From: phil at hands.com (Philip Hands) Date: Wed, 24 Mar 2010 09:08:55 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <87tys6v8ui.fsf@spindle.srvr.nix> References: <4BA807BF.5050106@googlemail.com> <87tys6v8ui.fsf@spindle.srvr.nix> Message-ID: <20100324090855.13cccae2@palm.hands.com> On Wed, 24 Mar 2010 00:53:09 +0000 Nix wrote: > On 23 Mar 2010, James Courtier-Dutton uttered the following: ... > Git uses snapshots of the entire tree, chained together into a > directed graph of commits over time. In fact, the directed graph bit is reasonably independent of the objects, so you have a big bucket of versions, and then outside that there's the graph of the history that gave rise to these versions. I fell in love with git after having utterly misunderstood what I was doing when trying to rewrite history only to discover that I'd lobotomised the repository (by using --force a few times when it wouldn't do what I thought I wanted -- Doh!), and the versions I cared most about had disappeared -- you might think that's an odd point at which to fall in love with it, but having taken a machine-shotgun to my entire leg, I fairly soon discovered: git fsck --unreachable This lists versions that exist in the repository but are not recorded in history -- iI fairly rapidly worked out that doing something like: gitk --all $(git fsck --unreachable | cut -d\ -f3) allows you to visualise where the orphans fit when compared with the rest of the archive's history -- I had my lost version back in minutes, and realised that you have to be considerably more stupid than I'd been to actually lose data with git (I could have managed that by running the garbage collector after being stupid). It is rather important to understand something of the underlying model in git though -- it's not difficult to do that, and will ensure that you don't just feel confused all the time -- I'd recommend reading: http://www.newartisans.com/2008/04/git-from-the-bottom-up.html So, don't dismiss git as being overkill, it's really very good for all sorts of odd tasks (I use it with etckeeper to track the contents of /etc on all my systems, for instance) and if you're liable to do any collaboration on free software you're going to bump into it eventually. Once you're addicted to git, the pain of going back to svn collaboration can be completely eliminated with git-svn. If you're fed up with having to care what sort of repository you're sitting in, Joey Hess has a wrapper that will drive all of them called 'mr' Cheers, Phil. -- |)| Philip Hands [+44 (0)20 8530 9560] http://www.hands.com/ |-| HANDS.COM Ltd. http://www.uk.debian.org/ |(| 10 Onslow Gardens, South Woodford, London E18 1NE ENGLAND From gllug at getaroundtoit.co.uk Wed Mar 24 09:53:56 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Wed, 24 Mar 2010 22:53:56 +1300 Subject: [Gllug] Survey: FOSS use in New Zealand Message-ID: <4BA9E134.3070806@getaroundtoit.co.uk> FYI, Regards, =dn "Survey finds support available for open desktop shift Proprietary document management systems are a barrier to the open source desktop By Computerworld staff | Auckland | Monday, 22 March, 2010 | 1 Comment New Zealand Open Source Society president Don Christie says the results of a local survey show there is a "rich choice" of service providers ready to help organisations move to open source desktops. The society and and Victoria University conducted their Public Sector Remix IT Vendor Capability Survey (pdf) after government agencies identified the availability of third party support as an essential enabler of free desktop software adoption." ... http://computerworld.co.nz/news.nsf/news/survey-finds-support-available-for-open-desktop-shift From gllug at getaroundtoit.co.uk Wed Mar 24 09:54:37 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Wed, 24 Mar 2010 22:54:37 +1300 Subject: [Gllug] More ACTA from New Zealand Message-ID: <4BA9E15D.5020106@getaroundtoit.co.uk> FYI, Regards, =dn "NZ Govt pressing for greater transparency in ACTA talks However, NZ has to respect the views of partner countries, says commerce minister Power By James Heffield | PC World, Auckland | Tuesday, 23 March, 2010 | 1 Comment New Zealand officials are "pressing for greater transparency" in future Anti-Counterfeiting Trade Agreement (ACTA) negotiations, says commerce minister Simon Power. The eighth round of negotiations looms in Wellington from April 12-16 and many critics, including Labour communications spokesperson Clare Curran, have questioned the lack of publicly available information." ... http://computerworld.co.nz/news.nsf/news/nz-govt-pressing-for-greater-transparency-in-acta-negotiations From gllug at getaroundtoit.co.uk Wed Mar 24 10:28:30 2010 From: gllug at getaroundtoit.co.uk (David L Neil) Date: Wed, 24 Mar 2010 23:28:30 +1300 Subject: [Gllug] Formal reaction to Australia's plans to censor the internet Message-ID: <4BA9E94E.1070007@getaroundtoit.co.uk> Another report from down-under, FYI. Regards, =dn "Conroy's internet censorship agenda slammed by tech giants ASHER MOSES March 23, 2010 Australia's biggest technology companies, communications academics and many lobby groups have delivered a withering critique of the government's plans to censor the internet. The government today published most of the 174 submissions it received relating to improving the transparency and accountability measures of its internet filtering policy." ... http://www.smh.com.au/technology/technology-news/conroys-internet-censorship-agenda-slammed-by-tech-giants-20100323-qt83.html From tim at seacon.co.uk Wed Mar 24 11:31:07 2010 From: tim at seacon.co.uk (t.clarke) Date: Wed, 24 Mar 2010 11:31:07 +0000 Subject: [Gllug] Port forwarding with IP-tables Message-ID: Hi, I wonder if anyone who is proficient with iptables can advise what commands I need to use to set up the following: I have a setup currently with a router connected to a Debian box on 192.168.1.2 which is also connected to an internal 192.168.123.0 network. Occasionally when the ADSL connection plays up it is necessary to reboot the router remotely. Currently to do this I have to telnet from home to the main server (on the internal network), then telnet to the Debian box, then telnet from there to to the router and then login to the router as 'root' and issue the reboot command. This works, although is tedious. But the router immediately drops the telnet connection and I can't gracefully exit the 'chain' of telnets. What I would like to do is simply telnet from the main server to the router by setting up port-forwarding on the Debian box so that, for example, I can telnet to say port 4567 on the Debian box which port-forwards to port 23 on 192.168.1.2 Any quick how-to-do guide would be most gratefully received Tim -------------------------------------------------------------------------------- This E-Mail (and any files transmitted with it) is intended solely for the use of the individual or entity to whom it is addressed. If you have received it in error please notify the sender and delete the message. Seacon Terminals Ltd, Company No. 1547396, registered in England and Wales Registered Office: Tower Wharf, Northfleet, Kent, DA11 9BD, England From addw at phcomp.co.uk Wed Mar 24 11:35:31 2010 From: addw at phcomp.co.uk (Alain Williams) Date: Wed, 24 Mar 2010 11:35:31 +0000 Subject: [Gllug] Port forwarding with IP-tables In-Reply-To: References: Message-ID: <20100324113531.GG4083@phcomp.co.uk> On Wed, Mar 24, 2010 at 11:31:07AM +0000, t.clarke wrote: > Hi, > > I wonder if anyone who is proficient with iptables can advise what commands I > need to use to set up the following: > > I have a setup currently with a router connected to a Debian box on 192.168.1.2 > which is also connected to an internal 192.168.123.0 network. Occasionally > when the ADSL connection plays up it is necessary to reboot the router remotely. > Currently to do this I have to telnet from home to the main server (on the > internal network), then telnet to the Debian box, then telnet from there to > to the router and then login to the router as 'root' and issue the reboot > command. This works, although is tedious. But the router immediately drops > the telnet connection and I can't gracefully exit the 'chain' of telnets. > > What I would like to do is simply telnet from the main server to the router > by setting up port-forwarding on the Debian box so that, for example, I > can telnet to say port 4567 on the Debian box which port-forwards to port 23 > on 192.168.1.2 > > Any quick how-to-do guide would be most gratefully received iptables -t nat -A PREROUTING -i eth0 -p tcp --dport 4567 -j DNAT --to-destination 192.168.1.2:23-23 -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 http://www.phcomp.co.uk/ Parliament Hill Computers Ltd. Registration Information: http://www.phcomp.co.uk/contact.php Past chairman of UKUUG: http://www.ukuug.org/ #include From social-bounces at gllug.org.uk Wed Mar 24 10:38:08 2010 From: social-bounces at gllug.org.uk (social-bounces at gllug.org.uk) Date: Wed, 24 Mar 2010 10:38:08 +0000 Subject: [Gllug] =?iso-8859-1?q?Forward_of_moderated_message?= Message-ID: An embedded message was scrubbed... From: =?ISO-8859-1?Q?L=E9on_Brocard?= Subject: [ANNOUNCE] London Perl Mongers Technical Meeting 12th April 2010 Date: Tue, 23 Mar 2010 16:13:03 +0000 Size: 5366 Url: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100324/f5ae58d7/attachment.eml From mick at dcs.bbk.ac.uk Wed Mar 24 12:24:21 2010 From: mick at dcs.bbk.ac.uk (Mick Farmer) Date: Wed, 24 Mar 2010 12:24:21 +0000 Subject: [Gllug] Netgear DG834G ADSL Router In-Reply-To: Message from John Winters on Tue, 23 Mar 2010 20:21:30 +0000 .<4BA922CA.5020405@sinodun.org.uk> Message-ID: <201003241224.o2OCOLod014626@zeus.dcs.bbk.ac.uk> Dear John, et. al., I've delved a little deeper and think my problem may come down to the routing table. Destination Gateway 195.166.128.80 0.0.0.0 192.168.7.0 0.0.0.0 127.0.0.0 0.0.0.0 239.0.0.0 0.0.0.0 0.0.0.0 195.166.128.80 My private LAN is 192.168.7.0 and the router's address on this LAN is 192.168.7.245. My router can ping the machine with a fixed IP address, 192.168.7.5, but none of the DHCP-connected machines. My external address is 212.159.72.229. 195.166.128.80 is a mchine on my ISP's network, lo0-plusnet.thn-ag3.plus.net. There's no documentation on address 0.0.0.0. Shouldn't the gateway to 192.168.7.0 be 192.168.7.245? Regards, Mick /"\ \ / Linux Registered X ASCII Ribbon Campaign User #287765 / \ Against HTML Mail From tim at seacon.co.uk Wed Mar 24 13:47:38 2010 From: tim at seacon.co.uk (t.clarke) Date: Wed, 24 Mar 2010 13:47:38 +0000 Subject: [Gllug] Port forwarding with IP-tables In-Reply-To: <20100324113531.GG4083@phcomp.co.uk> References: <20100324113531.GG4083@phcomp.co.uk> Message-ID: Alain thanks for the response works fine !! Tim From ba1020 at homie.homelinux.net Wed Mar 24 14:48:27 2010 From: ba1020 at homie.homelinux.net (Juergen Schinker) Date: Wed, 24 Mar 2010 14:48:27 +0000 Subject: [Gllug] Netgear DG834G ADSL Router In-Reply-To: <201003231846.o2NIk4nA000994@zeus.dcs.bbk.ac.uk> References: <201003231846.o2NIk4nA000994@zeus.dcs.bbk.ac.uk> Message-ID: <4BAA263B.9040609@homie.homelinux.net> Mick Farmer wrote: > Dear GLLUGers, > > I've been using one of these for a number of years and it's > been rock steady. > > However, I have a query about its connection to my private > LAN. There's a display called "Attached Devices". Most of > the time it shows nothing. After a number of Refresh hits > it will show my main computer which has a fixed IP address. > This can disappear on subsequent Refresh hits. > > It never shows my other machines which use DHCP for their > addresses. Does it treat them differently? > > I can't make contact with the other machines on the LAN. > Not even ping works. Does this imply that ARP is not > implemented on the router? > > Could this be because I'm using wireless connections and not > the Ethernet ports? > > Any guidance much appreciated. > > Regards, > > Mick /"\ > \ / > Linux Registered X ASCII Ribbon Campaign > User #287765 / \ Against HTML Mail well the wireless clients could be wireless isolated there is a wireless isolation in the Wireless Settings.. in the attached devices you should see all clients no matter how attached... do not post your public IP ! this Router can do ARP ! J From rich at annexia.org Wed Mar 24 17:01:26 2010 From: rich at annexia.org (Richard Jones) Date: Wed, 24 Mar 2010 17:01:26 +0000 Subject: [Gllug] Is this a RAM error? In-Reply-To: <87fx3qwokg.fsf@spindle.srvr.nix> References: <20100322215437.GA7905@annexia.org> <87fx3qwokg.fsf@spindle.srvr.nix> Message-ID: <20100324170125.GA10995@annexia.org> On Wed, Mar 24, 2010 at 12:28:15AM +0000, Nix wrote: > ... to be an L2 cache problem. > > > CPUID Vendor Intel Family 6 Model 15 > > No DIMM found for 114ea4470 in SMBIOS > > (that's not much of a surprise, these tables are as reliable as anything > else written by BIOS vendors, i.e., utter shit). Ugh ... I *don't* want to replace the processor :-( Rich. -- Richard Jones Red Hat From rich at annexia.org Wed Mar 24 17:09:30 2010 From: rich at annexia.org (Richard Jones) Date: Wed, 24 Mar 2010 17:09:30 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <877hp2wntz.fsf@spindle.srvr.nix> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <877hp2wntz.fsf@spindle.srvr.nix> Message-ID: <20100324170930.GB10995@annexia.org> On Wed, Mar 24, 2010 at 12:44:08AM +0000, Nix wrote: > It adds *the actual file*, i.e. the *actual content* to the 'index' > (also known as the 'cache' or 'staging area', which is a sort of limbo > files go into before commit. Change the file after 'git add' then do a > 'git commit', and the thing you added, *without that change*, is what > gets committed, because what gets added is what's in the index, and the > index hasn't changed since you did a 'git add'. Yup, the index is an odd and oddly great concept. It does take a bit of getting used to, and is quite unlike any other VCS (AFAIK?) > (If the changes touch some of the same files, this won't work quite so > easily, but 'git add -p' lets you add only *chunks* of changed files, > similar to darcs.) Don't forget the awesome power of the 's' key (in git add -p mode), which splits up the hunks to be smaller and smaller. This means you can split up even closely related changes. And (I discovered the other day) you can even edit the hunks manually before they get added. What I actually do if I'm randomly hacking on different features, is to commit it all and frequently (using --amend), then: git reset --mixed HEAD^ to completely undo the whole commit, and add/commit it in logical chunks. Rebasing / rewriting history has completely changed how I think and present patches. Then there's 'git reflog', which means as long as you committed the change at some point, you're never going to lose it, short of a hard disk failure. Rich. -- Richard Jones Red Hat From dan at berrange.com Wed Mar 24 17:22:08 2010 From: dan at berrange.com (Daniel P. Berrange) Date: Wed, 24 Mar 2010 17:22:08 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100324170930.GB10995@annexia.org> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <877hp2wntz.fsf@spindle.srvr.nix> <20100324170930.GB10995@annexia.org> Message-ID: <20100324172208.GA25711@berrange.com> On Wed, Mar 24, 2010 at 05:09:30PM +0000, Richard Jones wrote: > On Wed, Mar 24, 2010 at 12:44:08AM +0000, Nix wrote: > > It adds *the actual file*, i.e. the *actual content* to the 'index' > > (also known as the 'cache' or 'staging area', which is a sort of limbo > > files go into before commit. Change the file after 'git add' then do a > > 'git commit', and the thing you added, *without that change*, is what > > gets committed, because what gets added is what's in the index, and the > > index hasn't changed since you did a 'git add'. > > Yup, the index is an odd and oddly great concept. It does take a bit > of getting used to, and is quite unlike any other VCS (AFAIK?) Perforce (closed source) also had a pretty similar concept to this. Regards, Daniel -- |: http://berrange.com/ -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://freshmeat.net/~danielpb/ -o- http://gtk-vnc.sourceforge.net :| -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100324/1c06f3a2/attachment.pgp From tethys at gmail.com Wed Mar 24 17:46:35 2010 From: tethys at gmail.com (- Tethys) Date: Wed, 24 Mar 2010 17:46:35 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100324170930.GB10995@annexia.org> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <877hp2wntz.fsf@spindle.srvr.nix> <20100324170930.GB10995@annexia.org> Message-ID: <264d99b01003241046w6be640fdjeac699636733b9e7@mail.gmail.com> On Wed, Mar 24, 2010 at 5:09 PM, Richard Jones wrote: > Don't forget the awesome power of the 's' key (in git add -p mode), > which splits up the hunks to be smaller and smaller.  This means you > can split up even closely related changes. At work, we went from darcs to git, adn now back to darcs again. The single thing I miss most about git is the 's' key in "add -p" mode. > And (I discovered the other day) you can even edit the hunks > manually before they get added. How? Even splitting patches wasn't always fine-grained enough for me, and it only let me split the patch into multiple hunks as defined by diff. I sometimes wanted to further subdivide an individual hunk, when multiple logical changes affected a single block of code, but couldn't see a way to do it. Tet -- ?It seems intuitively obvious to me, which means that it might be wrong.? -- Chris Torek From rich at annexia.org Wed Mar 24 18:17:13 2010 From: rich at annexia.org (Richard Jones) Date: Wed, 24 Mar 2010 18:17:13 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <264d99b01003241046w6be640fdjeac699636733b9e7@mail.gmail.com> References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <877hp2wntz.fsf@spindle.srvr.nix> <20100324170930.GB10995@annexia.org> <264d99b01003241046w6be640fdjeac699636733b9e7@mail.gmail.com> Message-ID: <20100324181713.GA13878@annexia.org> On Wed, Mar 24, 2010 at 05:46:35PM +0000, - Tethys wrote: > On Wed, Mar 24, 2010 at 5:09 PM, Richard Jones wrote: > > > Don't forget the awesome power of the 's' key (in git add -p mode), > > which splits up the hunks to be smaller and smaller.  This means you > > can split up even closely related changes. > > At work, we went from darcs to git, adn now back to darcs again. The > single thing I miss most about git is the 's' key in "add -p" mode. > > > And (I discovered the other day) you can even edit the hunks > > manually before they get added. > > How? Even splitting patches wasn't always fine-grained enough for me, > and it only let me split the patch into multiple hunks as defined by > diff. I sometimes wanted to further subdivide an individual hunk, > when multiple logical changes affected a single block of code, but > couldn't see a way to do it. Press 'e' (edit) during add -p: # Manual hunk edit mode -- see bottom for a quick guide @@ -1,4 +1,2 @@ a -b -c d # --- # To remove '-' lines, make them ' ' lines (context). # To remove '+' lines, delete them. # Lines starting with # will be removed. # # If the patch applies cleanly, the edited hunk will immediately be # marked for staging. If it does not apply cleanly, you will be given # an opportunity to edit again. If all lines of the hunk are removed, # then the edit is aborted and the hunk is left unchanged. Rich. -- Richard Jones Red Hat From adrian at newgolddream.dyndns.info Wed Mar 24 19:29:26 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Wed, 24 Mar 2010 19:29:26 -0000 Subject: [Gllug] [OT] Hardware problem In-Reply-To: <87sk7qwp1d.fsf@spindle.srvr.nix> References: <1269212059.27507.6.camel@bossclass> <4BA7465F.9030103@shihad.org> <1269298435.9762.2.camel@bossclass> <20100323044052.GA5115@phaistos.bruce> <1269332942.5512.3.camel@bossclass> <1269333166.5512.5.camel@bossclass> <4BA8CA23.1050805@technicalbloke.com> <87sk7qwp1d.fsf@spindle.srvr.nix> Message-ID: On Wed, March 24, 2010 12:18 am, Nix wrote: > On 23 Mar 2010, Adrian McMenamin said: >> Second was acpi=off which worked > > ACPI really really should work on a modern board: with it off a lot of > things won't work, from temperature regulation and the power button to > (on some boards) even things like CD-ROM ejecting and the sound card (!) > > Maybe a BIOS upgrade is in order? It appears to be only an issue with the old (amd64) kernel the Ubuntu 9.10 install CD used. Using the up to date kernel with acpi seems to work fine. From john at cornerstonelinux.co.uk Wed Mar 24 19:53:55 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Wed, 24 Mar 2010 19:53:55 +0000 Subject: [Gllug] [ANN] Meeting TONIGHT - libguestfs In-Reply-To: <20100318222933.GA1380@annexia.org> References: <3e4e51a81003180248r3fbf5ce6wbb79ea80502a0dc6@mail.gmail.com> <20100318222933.GA1380@annexia.org> Message-ID: <20100324195355.GG16801@cornerstonelinux.co.uk> On Thu, Mar 18, 2010 at 10:29:33PM +0000, Richard Jones wrote: > > Hope everyone enjoyed it. I think we had a good turn out for a > weekday evening. Thanks very much for a very good talk. -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100324/85c11670/attachment-0001.pgp From general.mooney at googlemail.com Thu Mar 25 08:53:02 2010 From: general.mooney at googlemail.com (=?UTF-8?Q?Ciar=C3=A1n_Mooney?=) Date: Thu, 25 Mar 2010 08:53:02 +0000 Subject: [Gllug] [ANN] GLLUG Talk - 15th April - University of Westminster Message-ID: <3e4e51a81003250153q3e0f72ebue4218089d85905f6@mail.gmail.com> Hi, The next talk is titled "Wire Shark and the basics of network traffic monitoring." and will be given by Nahual Marisi and John Sharp. We have a slight change of venue, we will still be meeting at the University of Westminster (Cavendish Campus), but in Room C1.04[1]. If people could arrive for 1900, ready for the talk to begin at 1930. I will try and get some signs made up to direct people to room C1.04 Hopefully there will be a trip to a local pub for social and technical chit-chat. I feel the "I'm going" email's were unpopular and Sean had a sign-in sheet organised on the day anyway. So I think we should stick people just turning up and signing in. (ie you don't have to email the list or me explicitly stating you are going, unless you *really really* want us to know.) Lastly thanks to all those that attended the last meeting and to Sean for organising a place for us to meet. I really enjoyed meeting you all and listening to what Richard had to say. Regards, Ciarán [1] http://www.openstreetmap.org/index.html?mlat=51.520918&mlon=-0.140103&zoom=14 PS. We still have no speaker for the 20th May meeting. If you have something interesting to say please propose it on the list or email me directly. From npauli at st-johns.org.uk Thu Mar 25 18:52:22 2010 From: npauli at st-johns.org.uk (N. Pauli) Date: Thu, 25 Mar 2010 18:52:22 +0000 (GMT) Subject: [Gllug] Hitachi SMS-100 Message-ID: Dear All, We're currently considering one of these as a SAN to replace our current directly attached storage RAID5 file server. We're a school with about 500 user accounts. Next year we would like to virtualise some of our current servers [mainly lightly loaded database servers but also one handling WSUS and acting as a BDC] which means they too would be using the SMS-100. Is anyone out there using an SMS-100 as their main fileserver and/or can tell me whether it would be up to the job? Regards, Nigel -- Nigel Pauli I.T. Manager St. John's School, Northwood http://www.st-johns.org.uk/ From hearnsj at googlemail.com Thu Mar 25 22:46:03 2010 From: hearnsj at googlemail.com (John Hearns) Date: Thu, 25 Mar 2010 22:46:03 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: References: Message-ID: <9f8092cc1003251546l1aca9c47j3b05d1f67052434@mail.gmail.com> On 25 March 2010 18:52, N. Pauli wrote: > Dear All, > > We're currently considering one of these as a SAN to replace our current directly attached storage RAID5 file server. Having a quick look at that, it has either fibrechannel or iSCSI connectivity. If FC, do you REALLY REALLY want to start coping with implementing a new FC SAN in a high school environment? And before anyone calls bullshit, I've personally replaced two SR GBICs, one MR GBIC and one LR GBIC in the last week, plus recently installed umpty Tbytes of SATA disk and extended an FC SAN. From npauli at st-johns.org.uk Fri Mar 26 08:17:06 2010 From: npauli at st-johns.org.uk (N. Pauli) Date: Fri, 26 Mar 2010 08:17:06 +0000 (GMT) Subject: [Gllug] Hitachi SMS-100 Message-ID: On Thu, 25 Mar, John Hearns wrote: > On 25 March 2010 18:52, N. Pauli wrote: > > Dear All, > > > > We're currently considering one of these as a SAN to replace our current directly attached storage RAID5 file server. > > Having a quick look at that, it has either fibrechannel or iSCSI connectivity. > If FC, do you REALLY REALLY want to start coping with implementing a > new FC SAN in a high school environment? > > And before anyone calls bullshit, I've personally replaced two SR > GBICs, one MR GBIC and one LR GBIC in the last week, plus recently > installed umpty Tbytes of SATA disk and extended an FC SAN. > No, definitely not fibre channel, John. It would be the iSCSI connectivity I'd be going for. The thing is, we had a technical / sales fellow over from RM who sell regular SAN solutions and this Hitachi SMS-100 which seemed to me totally right-sized and appropriate for us. But he summarily dismissed the SMS-100 [at £8K] as too slow to be a file server and talked a full SAN at double the price and requiring £1350 p.a. of support fees. As you can see I'm feeling slightly annoyed and extremely cynical about this; BUT I could have the wrong end of the stick. Looking at http://www.hds.com/uk/products/storage-systems/simple-modular-storage.html and various reviews it seems to be well spec'd and genuinely aimed at non-specialist IT support staff at a small to medium sized business which, at the end of the day, is pretty well what we are at a prep school. What do you all think? Regards, Nigel -- Nigel Pauli I.T. Manager St. John's School, Northwood http://www.st-johns.org.uk/ From rw at shadowrobot.com Fri Mar 26 09:21:18 2010 From: rw at shadowrobot.com (Rich Walker) Date: Fri, 26 Mar 2010 09:21:18 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: References: Message-ID: <4BAC7C8E.2030904@shadowrobot.com> On 2010-03-26 08:17, N. Pauli wrote: > The thing is, we had a technical / sales fellow over from RM who sell > regular SAN solutions and this Hitachi SMS-100 which seemed to me > totally right-sized and appropriate for us. But he summarily > dismissed the SMS-100 [at £8K] as too slow to be a file server and > talked a full SAN at double the price and requiring £1350 p.a. of > support fees. > Wouldn't "Guy from RM dismisses it" translate as "Must-Buy" in everyone else's world? cheers, Rich. -- rich walker | Shadow Robot Company | rw at shadowrobot.com managing director 251 Liverpool Road | skype: rich_at_shadow need a Hand? London N1 1LX | +44 20 7700 2487 http://www.shadowrobot.com/hand/ @shadowrobot From essuu at ourshack.com Fri Mar 26 09:52:59 2010 From: essuu at ourshack.com (Simon Wilcox) Date: Fri, 26 Mar 2010 09:52:59 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: References: Message-ID: <4BAC83FB.5010406@ourshack.com> Hi Nigel, On 25/3/10 18:52, N. Pauli wrote: > We're currently considering one of these as a SAN to replace our > current directly attached storage RAID5 file server. Looks quite nice. Have you got more detailed specs on it as there's nothing much in the marketing blurb. Apologies if you've already asked this but I would be asking questions like: How does it compare to your raid-5 file server in terms of disks and raid configuration ? How are you using the raid-5 system now in terms of IOPS, data rate, read/write ratio etc ? They should be able to give you expected performance for given r/w ratios and disk configurations although they probably won't publish that anywhere (it's often considered proprietary information). > We're a school with about 500 user accounts. Next year we would like > to virtualise some of our current servers [mainly lightly loaded > database servers but also one handling WSUS and acting as a BDC] > which means they too would be using the SMS-100. Which virtualisation technology do you plan on using ? Make sure the Hitachi management software allows you to monitor the read/write ratio and the IOPS performance of the SAN. Use perfmon to measure the disk activity on the physical Windows boxes. This will give you a good idea whether you have enough capacity in the SAN to take the extra workload. > Is anyone out there using an SMS-100 as their main fileserver and/or > can tell me whether it would be up to the job? You need to understand what your current workload is before you can say whether it will be suitable. Also, of course, make sure you allow for backup in your solution and have a contingency plan for when this single point of failure fails. Because it *will* fail at some point. Rgds, Simon. From hearnsj at googlemail.com Fri Mar 26 10:21:56 2010 From: hearnsj at googlemail.com (John Hearns) Date: Fri, 26 Mar 2010 10:21:56 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: References: Message-ID: <9f8092cc1003260321j253d81a9r29caaf57d5be48bc@mail.gmail.com> On 26 March 2010 08:17, N. Pauli wrote: >>> > We're currently considering one of these as a SAN to replace our current directly attached storage RAID5 file server. >> > The thing is, we had a technical / sales fellow over from RM who sell regular SAN solutions and this Hitachi SMS-100 which seemed to me totally right-sized and appropriate for us. But he summarily dismissed the SMS-100 [at £8K] as too slow to be a file server and talked a full SAN at double the price and requiring £1350 p.a. of support fees. > > As you can see I'm feeling slightly annoyed and extremely cynical about this; You are quite right to be cynical. Keep that cynicism. the buzzword at the moment is unified networking - specially at the scale you are talking about. Get yourself a reliable core network switch - heck, think abotu 10gig ethernet for this iSCSI server (I admit this is probably not an option) Me, I would ask for a demo unit for a month onsite - see if it does what you want it to. The company I work for recently went for Hitachi storage - I'm not directly involved, so I don;t know what exact kit they have. From npauli at st-johns.org.uk Fri Mar 26 11:31:05 2010 From: npauli at st-johns.org.uk (N. Pauli) Date: Fri, 26 Mar 2010 11:31:05 +0000 (GMT) Subject: [Gllug] Hitachi SMS-100 Message-ID: On Fri, 26 Mar, Rich Walker wrote: > On 2010-03-26 08:17, N. Pauli wrote: > > The thing is, we had a technical / sales fellow over from RM who sell > > regular SAN solutions and this Hitachi SMS-100 which seemed to me > > totally right-sized and appropriate for us. But he summarily > > dismissed the SMS-100 [at £8K] as too slow to be a file server and > > talked a full SAN at double the price and requiring £1350 p.a. of > > support fees. > > Wouldn't "Guy from RM dismisses it" translate as "Must-Buy" in everyone > else's world? > > cheers, Rich. > RM aren't too bad, Rich. It being a recession and all I expect I can't blame people seeing us as likely to have some money and deciding to do a bit of over-selling. Regards, Nigel -- Nigel Pauli I.T. Manager St. John's School, Northwood http://www.st-johns.org.uk/ From npauli at st-johns.org.uk Fri Mar 26 11:31:10 2010 From: npauli at st-johns.org.uk (N. Pauli) Date: Fri, 26 Mar 2010 11:31:10 +0000 (GMT) Subject: [Gllug] Hitachi SMS-100 Message-ID: On Fri, 26 Mar, Simon Wilcox wrote: > Hi Nigel, > > On 25/3/10 18:52, N. Pauli wrote: > > We're currently considering one of these as a SAN to replace our > > current directly attached storage RAID5 file server. > > Looks quite nice. Have you got more detailed specs on it as there's nothing much in the marketing blurb. > > Apologies if you've already asked this but I would be asking questions like: > > How does it compare to your raid-5 file server in terms of disks and raid configuration ? > > How are you using the raid-5 system now in terms of IOPS, data rate, read/write ratio etc ? > > They should be able to give you expected performance for given r/w ratios and disk configurations although they probably won't publish that anywhere (it's often considered proprietary information). > > > We're a school with about 500 user accounts. Next year we would like > > to virtualise some of our current servers [mainly lightly loaded > > database servers but also one handling WSUS and acting as a BDC] > > which means they too would be using the SMS-100. > > Which virtualisation technology do you plan on using ? > > Make sure the Hitachi management software allows you to monitor the read/write ratio and the IOPS performance of the SAN. > > Use perfmon to measure the disk activity on the physical Windows boxes. This will give you a good idea whether you have enough capacity in the SAN to take the extra workload. > > > Is anyone out there using an SMS-100 as their main fileserver and/or > > can tell me whether it would be up to the job? > > You need to understand what your current workload is before you can say whether it will be suitable. > > Also, of course, make sure you allow for backup in your solution and have a contingency plan for when this single point of failure fails. Because it *will* fail at some point. > Simon, Thanks for those very useful questions to ask. We're pretty sure that we can better what we have in terms of access speeds and I've done as much due diligence as I can in terms of right-sizing and manageability. Most usefully I've found some other schools who've got them. Regards, Nigel -- Nigel Pauli I.T. Manager St. John's School, Northwood http://www.st-johns.org.uk/ From npauli at st-johns.org.uk Fri Mar 26 11:31:15 2010 From: npauli at st-johns.org.uk (N. Pauli) Date: Fri, 26 Mar 2010 11:31:15 +0000 (GMT) Subject: [Gllug] Hitachi SMS-100 Message-ID: On Fri, 26 Mar, John Hearns wrote: > On 26 March 2010 08:17, N. Pauli wrote: > >>> > We're currently considering one of these as a SAN to replace our current directly attached storage RAID5 file server. > >> > > The thing is, we had a technical / sales fellow over from RM who sell regular SAN solutions and this Hitachi SMS-100 which seemed to me totally right-sized and appropriate for us. But he summarily > dismissed the SMS-100 [at £8K] as too slow to be a file server and talked a full SAN at double the price and requiring £1350 p.a. of support fees. > > > > As you can see I'm feeling slightly annoyed and extremely cynical about this; > You are quite right to be cynical. Keep that cynicism. the buzzword at > the moment is unified networking - specially at the scale you are > talking about. > Get yourself a reliable core network switch - heck, think abotu 10gig > ethernet for this iSCSI server (I admit this is probably not an > option) > > Me, I would ask for a demo unit for a month onsite - see if it does > what you want it to. > > The company I work for recently went for Hitachi storage - I'm not > directly involved, so I don;t know what exact kit they have. Thanks for the useful advice. To start we're going to use reasonably priced 1gig but that won't preclude upgrading as you suggest when it becomes an option. Regards, Nigel -- Nigel Pauli I.T. Manager St. John's School, Northwood http://www.st-johns.org.uk/ From martin at hinterlands.org Fri Mar 26 11:55:17 2010 From: martin at hinterlands.org (Martin A. Brooks) Date: Fri, 26 Mar 2010 11:55:17 -0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: References: Message-ID: <6c377ad6da8ec44f9141475665bc9ac0.squirrel@olga.hinterlands.org> On Thu, March 25, 2010 18:52, N. Pauli wrote: > Is anyone out there using an SMS-100 as their main fileserver and/or can > tell me whether it would be up to the job? One of my filtering customers uses Hitachi SAN kit and is very happy with it indeed. I have recently got to play with 3Par's SAN offering and I have to say that's it's a seriously impressive piece of kit. From benjamin at py-soft.co.uk Fri Mar 26 12:22:46 2010 From: benjamin at py-soft.co.uk (Benjamin Donnachie) Date: Fri, 26 Mar 2010 12:22:46 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: <6c377ad6da8ec44f9141475665bc9ac0.squirrel@olga.hinterlands.org> References: <6c377ad6da8ec44f9141475665bc9ac0.squirrel@olga.hinterlands.org> Message-ID: <-7927473572802250836@unknownmsgid> On 26 Mar 2010, at 11:55, "Martin A. Brooks" wrote: > I have recently got to play with 3Par's SAN offering and I > have to say that's it's a seriously impressive piece of kit. My (soon to be ex-) company recently invested in some 3par SANs and we are very pleased. But you get what you pay for - I think we paid in the region if $250k for the pair. Ben From ba1020 at homie.homelinux.net Fri Mar 26 15:14:19 2010 From: ba1020 at homie.homelinux.net (Juergen Schinker) Date: Fri, 26 Mar 2010 15:14:19 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: References: Message-ID: <4BACCF4B.2040405@homie.homelinux.net> N. Pauli wrote: > Dear All, > > We're currently considering one of these as a SAN to replace our current directly attached storage RAID5 file server. > > We're a school with about 500 user accounts. Next year we would like to virtualise some of our current servers [mainly lightly loaded database servers but also one handling WSUS and acting as a BDC] which means they too would be using the SMS-100. > > Is anyone out there using an SMS-100 as their main fileserver and/or can tell me whether it would be up to the job? > > Regards, > Nigel > I recommend you buy a Terrastation from Buffalo... connected via iSCSI and a reasonable fast Gigabit switch should be good enough for the job From general_email at technicalbloke.com Fri Mar 26 15:53:01 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Fri, 26 Mar 2010 15:53:01 +0000 Subject: [Gllug] Cygwin confusion... Message-ID: <4BACD85D.6020509@technicalbloke.com> I have customers HD which is on it's last legs, SMART is reporting it's used up its reallocated sector allowance. There looks to be a fair bit of logical damage so I can't just copy off the pertinent folders I'm going to have to image it. No problem under normal circumstances however I can't boot into Linux with this device attached, the kernel panics and I get no further, this happens with Ubuntu 8.04, 9.04 and Mint 9.04. It's an IDE drive so I can't start the machine then plug it in. Happily however XP can boot with it attached, although the logical damage is such it cannot read the volume and I don't want to let it run scandisk on it until I have an image of the whole drive. I have tried dd for windows but this is missing the "error=no" option and because there are bad sectors it can only get so far before falling over. So I set about finding a build of ddrescue for windows but the only one I could find requires cygwin: http://ftp.uni-kl.de/pub/windows/cygwin/release/ddrescue/ Fair enough, I installed that and copied the cygwin1.dll to the same folder as the .exe and it runs BUT... A) I'm not sure how to reference the drives in windows, fdisk -l doesn't work as cygwin doesn't seem to have fdisk and ls /dev at the cygwin prompt only shows half a dozen devices, none of them a hard drive: "fd mqueue shm stderr stdin stdout". Do I need to use the \\?\Device\HardDiskVolume3 style syntax here? B) I'd like to keep the 55Gb (of the 80Gb disk) dd was able to do before falling over as it took about 15hrs to get that far and this drive is clearly living on borrowed time. Is there a way I can get it to extend the partial image I already have? Thanks, Roger. From andy at andymillar.co.uk Fri Mar 26 16:49:53 2010 From: andy at andymillar.co.uk (Andy Millar) Date: Fri, 26 Mar 2010 16:49:53 -0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: <4BACCF4B.2040405@homie.homelinux.net> References: <4BACCF4B.2040405@homie.homelinux.net> Message-ID: <1269622193.6177.98.camel@millaralpt.co.tradefair> On Fri, 2010-03-26 at 15:14 +0000, Juergen Schinker wrote: > I recommend you buy a Terrastation from Buffalo... > > connected via iSCSI and a reasonable fast Gigabit switch > > should be good enough for the job Given their limit of 4 disks, I really doubt you'll get significantly decent performance out of them. We've recently bought some (15?) Thecus 8800s as cheap VM storage, they're dirt cheap and perform relatively well. (https://www.andymillar.co.uk/blog/index.php/2010/02/24/work/thecus-8800-sequential-io-performance/) They support ISCSI and up to 8 disks, getting the Pro model will allow you to fit an additional quad port GigE or dual port 10GigE card if you want more than 1 gbit connectivity to your device. Dell Equallogics are also pretty good for "cheap" storage. The Dell support is very good. Andy -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/x-pkcs7-signature Size: 5162 bytes Desc: not available Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100326/ecb5d625/attachment.bin From mail-lists at karan.org Fri Mar 26 17:00:46 2010 From: mail-lists at karan.org (Karanbir Singh) Date: Fri, 26 Mar 2010 17:00:46 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: References: Message-ID: <4BACE83E.1020801@karan.org> Hi, On 03/25/2010 06:52 PM, N. Pauli wrote: > We're currently considering one of these as a SAN to replace our current directly attached storage RAID5 file server. I'm guessing the raid05 box isn't doing all you want. Do you have some metrics on exactly where you are now and where you'd like to be ? I've worked with various HDS solutions over the years, ranging from rock bottom mass produced stuff to their very large scale USP-v based kit - and its all good. They are, imho, one of most effective solutions around at the moment. What this also translates to is that usually there are better bang-for-the-buck solutions to be hard from other vendors, specially in the lower end of the spectrum. - KB From chrisbell at 3966.ukfsn.org Fri Mar 26 19:56:08 2010 From: chrisbell at 3966.ukfsn.org (Chris Bell) Date: Fri, 26 Mar 2010 19:56:08 +0000 (GMT) Subject: [Gllug] Cygwin confusion... In-Reply-To: <4BACD85D.6020509@technicalbloke.com> Message-ID: On Fri 26 Mar, general_email at technicalbloke.com wrote: > > Fair enough, I installed that and copied the cygwin1.dll to the same > folder as the .exe and it runs BUT... > > A) I'm not sure how to reference the drives in windows, fdisk -l doesn't > work as cygwin doesn't seem to have fdisk and ls /dev at the cygwin > prompt only shows half a dozen devices, none of them a hard drive: "fd > mqueue shm stderr stdin stdout". Do I need to use the > \\?\Device\HardDiskVolume3 style syntax here? > > B) I'd like to keep the 55Gb (of the 80Gb disk) dd was able to do before > falling over as it took about 15hrs to get that far and this drive is > clearly living on borrowed time. Is there a way I can get it to extend > the partial image I already have? > > Thanks, > > Roger. I would not even consider doing fsck until all the information possible had been backed up to another disk. I would disconnect the drive until able to do a backup, either using dd or a file copy to another disc. If the computer will not boot from Knoppix or similar I would connect the drive to another computer which I could boot and do the backup there. Reliability would be far more important than speed, so I would be happy to use an older computer for the job. -- Chris Bell www.chrisbell.org.uk (was www.overview.demon.co.uk) Microsoft sells you Windows ... Linux gives you the whole house. From nix at esperi.org.uk Fri Mar 26 19:59:18 2010 From: nix at esperi.org.uk (Nix) Date: Fri, 26 Mar 2010 19:59:18 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100324170930.GB10995@annexia.org> (Richard Jones's message of "Wed, 24 Mar 2010 17:09:30 +0000") References: <4BA807BF.5050106@googlemail.com> <87iq8nbdyl.fsf@knight.monnsta.net> <877hp2wntz.fsf@spindle.srvr.nix> <20100324170930.GB10995@annexia.org> Message-ID: <871vf6svl5.fsf@spindle.srvr.nix> On 24 Mar 2010, Richard Jones uttered the following: > On Wed, Mar 24, 2010 at 12:44:08AM +0000, Nix wrote: >> It adds *the actual file*, i.e. the *actual content* to the 'index' >> (also known as the 'cache' or 'staging area', which is a sort of limbo >> files go into before commit. Change the file after 'git add' then do a >> 'git commit', and the thing you added, *without that change*, is what >> gets committed, because what gets added is what's in the index, and the >> index hasn't changed since you did a 'git add'. > > Yup, the index is an odd and oddly great concept. It does take a bit > of getting used to, and is quite unlike any other VCS (AFAIK?) I just wish the naming wasn't so damn inconsistent. Some commands have an --index option, some have --staging, some have --cache, and not only do some of these options do the same, in other commands identically- named options do different things. It's a bloody mess, and I'm half-inclined to commit a local patch cleaning it up (though Junio has already rejected the idea of fixing it upstream, gah). Equally, the opposite of 'git add' *should* obviously be named something like 'git remove', but isn't. This at least is easy to fix with two lines in .gitconfig: [alias] remove = reset -- >> (If the changes touch some of the same files, this won't work quite so >> easily, but 'git add -p' lets you add only *chunks* of changed files, >> similar to darcs.) > > Don't forget the awesome power of the 's' key (in git add -p mode), > which splits up the hunks to be smaller and smaller. This means you > can split up even closely related changes. And (I discovered the > other day) you can even edit the hunks manually before they get added. Ooo. I never noticed that feature, somehow, even though it's documented. That makes it even more useful :) > What I actually do if I'm randomly hacking on different features, is > to commit it all and frequently (using --amend), then: > > git reset --mixed HEAD^ > > to completely undo the whole commit, and add/commit it in logical > chunks. I tend to do a bunch of commits, then redo it in a sane fashion on another branch. I suspect your approach is easier, and will adopt it from now on, now that I know about the s key :) > Rebasing / rewriting history has completely changed how I think and > present patches. Quite so. No longer is 'commit' a heavyweight 'be terrified' option; it's something you do because you want to try something else, or because you're getting a coffee, or for any other reason that grabs your whim. It's pushing that you should now be terrified of :) > Then there's 'git reflog', which means as long as you committed the > change at some point, you're never going to lose it, short of a hard > disk failure. Well, if you revert something and then need it many weeks later, the reflog might have expired: but if you catch yourself doing that a lot you can turn off reflog expiry at the cost of some disk space (OK, if you do a lot of big rebases, quite a lot of disk space). From nix at esperi.org.uk Fri Mar 26 20:40:27 2010 From: nix at esperi.org.uk (Nix) Date: Fri, 26 Mar 2010 20:40:27 +0000 Subject: [Gllug] Mercurial or Subversion for single user? In-Reply-To: <20100324090855.13cccae2@palm.hands.com> (Philip Hands's message of "Wed, 24 Mar 2010 09:08:55 +0000") References: <4BA807BF.5050106@googlemail.com> <87tys6v8ui.fsf@spindle.srvr.nix> <20100324090855.13cccae2@palm.hands.com> Message-ID: <87vdcirf44.fsf@spindle.srvr.nix> On 24 Mar 2010, Philip Hands said: > which to fall in love with it, but having taken a machine-shotgun to my > entire leg, I fairly soon discovered: > > git fsck --unreachable > > This lists versions that exist in the repository but are not recorded in > history -- iI fairly rapidly worked out that doing something like: I've done this too. > gitk --all $(git fsck --unreachable | cut -d\ -f3) Oo, that's nice. I used git show instead, which is much less useful, because you only get single unreachable commits, not the whole parental history of each one. > allows you to visualise where the orphans fit when compared with the > rest of the archive's history -- I had my lost version back in minutes, > and realised that you have to be considerably more stupid than I'd been > to actually lose data with git (I could have managed that by running the > garbage collector after being stupid). I think you'll find that you'd *also* need to have waited two weeks first, or explicitly reset the gc.pruneExpire config option, or explicitly passed --prune=now or something like that to 'git gc'. > It is rather important to understand something of the underlying model > in git though -- it's not difficult to do that, and will ensure that > you don't just feel confused all the time -- I'd recommend reading: Lots of people I know avoided git because they weren't willing to put in the small amount of effort to understand this (they wanted it to 'just work', which is I suppose a reasonable position). You don't need to understand it anymore to do simple things, but if you want to really make it dance you have to understand about the commit DAG... > So, don't dismiss git as being overkill, it's really very good for all Oh, it *is* overkill... but you don't have to use all its features. > sorts of odd tasks (I use it with etckeeper to track the contents > of /etc on all my systems, for instance) and if you're liable to do any I kept on meaning to do that and never dared. I might set up a VM and experiment there... or just, I dunno, experiment on /tmp instead of /etc to start with :) > Once you're addicted to git, the pain of going back to svn collaboration > can be completely eliminated with git-svn. Now if only there were decent git<->hg and git<->bzr bidirectional gateways... anything that could stop me from ever again having to go near bloody bzr would be very welcome. > If you're fed up with having to care what sort of repository you're > sitting in, Joey Hess has a wrapper that will drive all of them called > 'mr' oo, nice (and essentially ungoogleable!) From nix at esperi.org.uk Fri Mar 26 20:41:56 2010 From: nix at esperi.org.uk (Nix) Date: Fri, 26 Mar 2010 20:41:56 +0000 Subject: [Gllug] Is this a RAM error? In-Reply-To: <20100324170125.GA10995@annexia.org> (Richard Jones's message of "Wed, 24 Mar 2010 17:01:26 +0000") References: <20100322215437.GA7905@annexia.org> <87fx3qwokg.fsf@spindle.srvr.nix> <20100324170125.GA10995@annexia.org> Message-ID: <87r5n6rf1n.fsf@spindle.srvr.nix> On 24 Mar 2010, Richard Jones said: > On Wed, Mar 24, 2010 at 12:28:15AM +0000, Nix wrote: >> ... to be an L2 cache problem. >> >> > CPUID Vendor Intel Family 6 Model 15 >> > No DIMM found for 114ea4470 in SMBIOS >> >> (that's not much of a surprise, these tables are as reliable as anything >> else written by BIOS vendors, i.e., utter shit). > > Ugh ... I *don't* want to replace the processor :-( I'd ask on l-k. The acknowledged expert regarding MCEs appears to be Andi Kleen. From nix at esperi.org.uk Fri Mar 26 20:46:51 2010 From: nix at esperi.org.uk (Nix) Date: Fri, 26 Mar 2010 20:46:51 +0000 Subject: [Gllug] Hitachi SMS-100 In-Reply-To: <4BAC7C8E.2030904@shadowrobot.com> (Rich Walker's message of "Fri, 26 Mar 2010 09:21:18 +0000") References: <4BAC7C8E.2030904@shadowrobot.com> Message-ID: <87mxxuretg.fsf@spindle.srvr.nix> On 26 Mar 2010, Rich Walker spake thusly: > On 2010-03-26 08:17, N. Pauli wrote: >> The thing is, we had a technical / sales fellow over from RM who sell >> regular SAN solutions and this Hitachi SMS-100 which seemed to me >> totally right-sized and appropriate for us. But he summarily >> dismissed the SMS-100 [at £8K] as too slow to be a file server and >> talked a full SAN at double the price and requiring £1350 p.a. of >> support fees. > > Wouldn't "Guy from RM dismisses it" translate as "Must-Buy" in everyone > else's world? I'd say so. I'm astonished that the education system still uses them. I mean, they were awful when I was at school, what, twenty years ago: if they're still there after twenty years, I suspect they think they have a sinecure. From general_email at technicalbloke.com Fri Mar 26 22:00:34 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Fri, 26 Mar 2010 22:00:34 +0000 Subject: [Gllug] Cygwin confusion... In-Reply-To: References: Message-ID: <4BAD2E82.5040008@technicalbloke.com> Chris Bell wrote: > On Fri 26 Mar, general_email at technicalbloke.com wrote: > >> Fair enough, I installed that and copied the cygwin1.dll to the same >> folder as the .exe and it runs BUT... >> >> A) I'm not sure how to reference the drives in windows, fdisk -l doesn't >> work as cygwin doesn't seem to have fdisk and ls /dev at the cygwin >> prompt only shows half a dozen devices, none of them a hard drive: "fd >> mqueue shm stderr stdin stdout". Do I need to use the >> \\?\Device\HardDiskVolume3 style syntax here? >> >> B) I'd like to keep the 55Gb (of the 80Gb disk) dd was able to do before >> falling over as it took about 15hrs to get that far and this drive is >> clearly living on borrowed time. Is there a way I can get it to extend >> the partial image I already have? >> >> Thanks, >> >> Roger. >> > > I would not even consider doing fsck until all the information possible > had been backed up to another disk. I would disconnect the drive until able > to do a backup, either using dd or a file copy to another disc. If the > computer will not boot from Knoppix or similar I would connect the drive to > another computer which I could boot and do the backup there. Reliability > would be far more important than speed, so I would be happy to use an older > computer for the job. > > Well that's exactly where I am now, I wouldn't dream of fscking a drive in this condition before backing up what I could of it, I think you must have missed the first half of my post somehow. The problem I have is that I'm not sure of the syntax for specifying which drives to use in ddrescue (on windows/cygwin, I haven't found a linux that will boot at all with this drive attached) and, more importantly, how to fire it up so it adds to the partial image file I got from using straight ahead windows dd as I don't want to put more hours on the disk than I absolutely have to. Cheers, Roger. From farnsaw at stonedoor.com Sat Mar 27 01:11:48 2010 From: farnsaw at stonedoor.com (Andrew Farnsworth) Date: Fri, 26 Mar 2010 21:11:48 -0400 Subject: [Gllug] Cygwin confusion... In-Reply-To: <4BAD2E82.5040008@technicalbloke.com> References: <4BAD2E82.5040008@technicalbloke.com> Message-ID: Chris, One thing that no one has mentioned yet is the possibility of sending the drive off to a recovery firm. It is not cheap, but if the data is important enough it is worth it. I have used Drive Savers [ http://www.drivesaversdatarecovery.com/] whom you can reach via 0-800-89-0011. I am NOT affiliated with them in any way, however, they did save my butt about 11 years ago with a customer's HD. Good luck, Andy -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100326/47953222/attachment.htm From chrisbell at 3966.ukfsn.org Sat Mar 27 06:45:09 2010 From: chrisbell at 3966.ukfsn.org (Chris Bell) Date: Sat, 27 Mar 2010 06:45:09 +0000 (GMT) Subject: [Gllug] Cygwin confusion... In-Reply-To: <4BAD2E82.5040008@technicalbloke.com> Message-ID: On Fri 26 Mar, general_email at technicalbloke.com wrote: > > Well that's exactly where I am now, I wouldn't dream of fscking a drive > in this condition before backing up what I could of it, I think you must > have missed the first half of my post somehow. The problem I have is > that I'm not sure of the syntax for specifying which drives to use in > ddrescue (on windows/cygwin, I haven't found a linux that will boot at > all with this drive attached) and, more importantly, how to fire it up > so it adds to the partial image file I got from using straight ahead > windows dd as I don't want to put more hours on the disk than I > absolutely have to. > > Cheers, > > > Roger. > That is why I suggested you disconnect the drive until you have found a way to boot a computer and are ready to retrieve the data. You may need to modify the BIOS settings to be able to boot from a CD, but I have seen very few PCs that will not boot from Knoppix. If the drive was originally Master on the first cable you may want to move it to the second IDE cable or change the drive jumper links to slave. You can of course pay a data recovery company, but I have computers with IDE drives in North Acton if you need to borrow one for the job. -- Chris Bell www.chrisbell.org.uk (was www.overview.demon.co.uk) Microsoft sells you Windows ... Linux gives you the whole house. From adrian at newgolddream.dyndns.info Sat Mar 27 11:21:15 2010 From: adrian at newgolddream.dyndns.info (Adrian McMenamin) Date: Sat, 27 Mar 2010 11:21:15 +0000 Subject: [Gllug] Flash 10 on amd64 Message-ID: <1269688875.19376.5.camel@bossclass> OK, I know I walking into a pit of hell by asking for advice on a piece of proprietary software ... but to continue the saga of the new computer I built for my daughters. I installed an amd64 kernel - from Ubuntu 9.10 and all now works swimmingly except Flash - which is important for them in terms of games etc Flash browser plugin seems randomly unresponsive to mouse clicks. This is a known problem but there seems to an infinite number of this works/no it doesn't solutions posted on the internet and so I thought I'd ask for quality advice from the list - and have put on the flame retardant underpants. I am using the non-free plugin - I have read that the free plugins are worse. Is that true? Anyone found a reliable work around From general_email at technicalbloke.com Sat Mar 27 13:16:16 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Sat, 27 Mar 2010 13:16:16 +0000 Subject: [Gllug] Cygwin confusion... In-Reply-To: References: Message-ID: <4BAE0520.5020807@technicalbloke.com> Chris Bell wrote: > On Fri 26 Mar, general_email at technicalbloke.com wrote: > > >> Well that's exactly where I am now, I wouldn't dream of fscking a drive >> in this condition before backing up what I could of it, I think you must >> have missed the first half of my post somehow. The problem I have is >> that I'm not sure of the syntax for specifying which drives to use in >> ddrescue (on windows/cygwin, I haven't found a linux that will boot at >> all with this drive attached) and, more importantly, how to fire it up >> so it adds to the partial image file I got from using straight ahead >> windows dd as I don't want to put more hours on the disk than I >> absolutely have to. >> >> Cheers, >> >> >> Roger. >> >> > That is why I suggested you disconnect the drive until you have found a > way to boot a computer and are ready to retrieve the data. You may need to > modify the BIOS settings to be able to boot from a CD, but I have seen very > few PCs that will not boot from Knoppix. If the drive was originally Master > on the first cable you may want to move it to the second IDE cable or change > the drive jumper links to slave. > You can of course pay a data recovery company, but I have computers with > IDE drives in North Acton if you need to borrow one for the job. > > > Dude thanks but I know all that, it is disconnected. I have fixed computers for a living for over 10 years now and routinely recover data from failing drives, I have myriad spare computers & my bios and jumpers are all set correctly. I would use Linux for this job but Linux WILL NOT BOOT with this drive attached where as windows WILL - that's why I'm using windows, the Linux kernel quite literally crashes during the boot process. I simply thought I'd ask if anyone knew the drive path syntax for ddrescue under cygwin and how to get it to append to an image rather than overwrite it as I have already imaged 80% of this drive. Thanks, Roger. From general_email at technicalbloke.com Sat Mar 27 13:23:02 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Sat, 27 Mar 2010 13:23:02 +0000 Subject: [Gllug] Flash 10 on amd64 In-Reply-To: <1269688875.19376.5.camel@bossclass> References: <1269688875.19376.5.camel@bossclass> Message-ID: <4BAE06B6.80703@technicalbloke.com> Adrian McMenamin wrote: > OK, I know I walking into a pit of hell by asking for advice on a piece > of proprietary software ... but to continue the saga of the new computer > I built for my daughters. > > I installed an amd64 kernel - from Ubuntu 9.10 and all now works > swimmingly except Flash - which is important for them in terms of games > etc > > How come you're not using Mint? Even when it works correctly Flash performance under Linux can be quite underwhelming esp as the video resolution creeps up. My P4x3000 really struggles to do fullscreen flash whereas it's silky smooth on my old PIII Windows laptop :( Roger. From jjllmmss at googlemail.com Sat Mar 27 13:56:18 2010 From: jjllmmss at googlemail.com (JLMS) Date: Sat, 27 Mar 2010 13:56:18 +0000 Subject: [Gllug] Flash 10 on amd64 In-Reply-To: <1269688875.19376.5.camel@bossclass> References: <1269688875.19376.5.camel@bossclass> Message-ID: On Sat, Mar 27, 2010 at 11:21 AM, Adrian McMenamin wrote: > OK, I know I walking into a pit of hell by asking for advice on a piece > of proprietary software ... but to continue the saga of the new computer > I built for my daughters. > > I installed an amd64 kernel - from Ubuntu 9.10 and all now works > swimmingly except Flash - which is important for them in terms of games > etc > > Flash browser plugin seems randomly unresponsive to mouse clicks. This > is a known problem but there seems to an infinite number of this > works/no it doesn't solutions posted on the internet and so I thought > I'd ask for quality advice from the list - and have put on the flame > retardant underpants. > > I am using the non-free plugin - I have read that the free plugins are > worse. Is that true? > > Anyone found a reliable work around > I think I did, I just cut and pasted, so I really don't understand the rationale of what is going on, but it worked for me: In the directory /usr/lib/nspluginwrapper/i386/linux rename npviewer.bin to npviewer.bin.real and then create an script called npviewer.bin that contains the following: ---- #!/bin/sh GDK_NATIVE_WINDOWS=true /usr/lib/nspluginwrapper/i386/linux/npviewer.bin.real $* ---- that seems to tell the binary to behave in the right way. Hope this helps. From john at cornerstonelinux.co.uk Sat Mar 27 14:14:32 2010 From: john at cornerstonelinux.co.uk (John Edwards) Date: Sat, 27 Mar 2010 14:14:32 +0000 Subject: [Gllug] Cygwin confusion... In-Reply-To: <4BAE0520.5020807@technicalbloke.com> References: <4BAE0520.5020807@technicalbloke.com> Message-ID: <20100327141432.GH16801@cornerstonelinux.co.uk> On Sat, Mar 27, 2010 at 01:16:16PM +0000, general_email at technicalbloke.com wrote: > I would use Linux for this job but Linux WILL NOT BOOT with this drive > attached where as windows WILL - that's why I'm using windows, the Linux > kernel quite literally crashes during the boot process. Does "quite literally crashes" mean that a kernel panic happens with an error message? On most Ubuntu systems you need to edit the boot options to remove "quiet" and "splash" to see all the error messages. Also have you tried putting the drive into a USB, Firewire, or eSATA enclosure and connecting it after the kernel has finished booting? Booting a BSD could be another way to get access to a UNIX system. > I simply thought > I'd ask if anyone knew the drive path syntax for ddrescue under cygwin > and how to get it to append to an image rather than overwrite it as I > have already imaged 80% of this drive. Here's the cygwin docs on /dev devices and the corresponding Windows device names: http://www.cygwin.com/cygwin-ug-net/using-specialnames.html -- #---------------------------------------------------------# | John Edwards Email: john at cornerstonelinux.co.uk | #---------------------------------------------------------# -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.gllug.org.uk/pipermail/gllug/attachments/20100327/fc43c02b/attachment.pgp From general_email at technicalbloke.com Sat Mar 27 19:33:01 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Sat, 27 Mar 2010 19:33:01 +0000 Subject: [Gllug] Cygwin confusion... In-Reply-To: <20100327141432.GH16801@cornerstonelinux.co.uk> References: <4BAE0520.5020807@technicalbloke.com> <20100327141432.GH16801@cornerstonelinux.co.uk> Message-ID: <4BAE5D6D.6070309@technicalbloke.com> John Edwards wrote: > On Sat, Mar 27, 2010 at 01:16:16PM +0000, general_email at technicalbloke.com wrote: > >> I would use Linux for this job but Linux WILL NOT BOOT with this drive >> attached where as windows WILL - that's why I'm using windows, the Linux >> kernel quite literally crashes during the boot process. >> > > Does "quite literally crashes" mean that a kernel panic happens with > an error message? Yep, can't remember the error message exactly but it definitely used the word "panic" and totally locked up. I want to minimize the hours this thing spends spun up though so I can't risk playing around trying to get a linux or BSD system to read it and needlessly power cycling the poor thing when windows seems quite able to see it. > On most Ubuntu systems you need to edit the boot > options to remove "quiet" and "splash" to see all the error messages. > > Also have you tried putting the drive into a USB, Firewire, or eSATA > enclosure and connecting it after the kernel has finished booting? > > Not yet actually, I've always got better recovery results from connecting drives directly (and it's nice to be able to see the SMART data too) so I haven't tried - good idea though, thanks. > Here's the cygwin docs on /dev devices and the corresponding > Windows device names: > http://www.cygwin.com/cygwin-ug-net/using-specialnames.html > > > Thanks, clearly I skimmed through that a bit too fast first time! :) Also found the --output-position= option in the ddrescue man pages so I reckon I'm set now. Cheers, Roger. From general_email at technicalbloke.com Sat Mar 27 21:29:42 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Sat, 27 Mar 2010 21:29:42 +0000 Subject: [Gllug] Cygwin confusion... In-Reply-To: <4BAE5D6D.6070309@technicalbloke.com> References: <4BAE0520.5020807@technicalbloke.com> <20100327141432.GH16801@cornerstonelinux.co.uk> <4BAE5D6D.6070309@technicalbloke.com> Message-ID: <4BAE78C6.7080000@technicalbloke.com> general_email at technicalbloke.com wrote: > John Edwards wrote: > >> Here's the cygwin docs on /dev devices and the corresponding >> Windows device names: >> http://www.cygwin.com/cygwin-ug-net/using-specialnames.html >> >> >> >> > > Thanks, clearly I skimmed through that a bit too fast first time! :) > > Also found the --output-position= option in the ddrescue man pages > so I reckon I'm set now. > > Cheers, > > Roger. > > Spoke a bit too soon there... This is what the using-specialnames bit of the docs reckons (I think)... $ ./ddrescue.exe -n -v --input-position 121506468b --input-position 121506468b if=\device\harddisk1 of=Z:\dee.img Z:\log_file.log ddrescue: cannot open input file: No such file or directory This is the same but with unix path separators... $ ./ddrescue.exe -n -v --input-position 121506468b --input-position 121506468b if=/device/harddisk1 of=Z:\dee.img Z:\log_file.log ddrescue: cannot open input file: No such file or directory This is the syntax that worked for dd... $ ./ddrescue.exe -r0 -v --input-position 121506468b --input-position 121506468b if=\\?\Device\Harddisk1 of=Z:\dee.img Z:\log_file.log cygwin warning: MS-DOS style path detected: if=\?DeviceHarddisk1 Preferred POSIX equivalent is: if=/?DeviceHarddisk1 CYGWIN environment variable option "nodosfilewarning" turns off this warning. Consult the user's guide for more details about POSIX paths: http://cygwin.com/cygwin-ug-net/using.html#using-pathnames ddrescue: cannot open input file: No such file or directory This is what the above error suggests... $ ./ddrescue.exe -r0 -v --input-position 121506468b --input-position 121506468b if=/?DeviceHarddisk1 of=Z:\dee.img Z:\log_file.log ddrescue: cannot open input file: No such file or directory Also if=/dev/hdb didn't work. I haven't tried windows drive letters as I have the drive unmounted so it hasn't got a drive letter, likewise df and cat /proc/partitions are no help. Any ideas? Thanks, Roger. From mblackmore at oxlug.org Sun Mar 28 00:50:59 2010 From: mblackmore at oxlug.org (Malcolm Blackmore) Date: Sun, 28 Mar 2010 00:50:59 +0000 Subject: [Gllug] Flash 10 on amd64 In-Reply-To: References: <1269688875.19376.5.camel@bossclass> Message-ID: <1269737459.2212.76.camel@ubuntu.clocksoft.dom> On Sat, 2010-03-27 at 13:56 +0000, JLMS wrote: > > I think I did, I just cut and pasted, Where did you get this from?? From robertc at boogdesign.com Sun Mar 28 11:23:52 2010 From: robertc at boogdesign.com (Rob Crowther) Date: Sun, 28 Mar 2010 11:23:52 +0100 Subject: [Gllug] Flash 10 on amd64 In-Reply-To: <1269688875.19376.5.camel@bossclass> References: <1269688875.19376.5.camel@bossclass> Message-ID: <4BAF2E38.8040907@boogdesign.com> On 27/03/10 11:21, Adrian McMenamin wrote: > Anyone found a reliable work around > I've been using the 64bit alpha on Fedora, seems to work OK: http://labs.adobe.com/technologies/flashplayer10/64bit.html Just download and put the file in ~/mozilla/.plugins Rob From nix at esperi.org.uk Sun Mar 28 17:13:40 2010 From: nix at esperi.org.uk (Nix) Date: Sun, 28 Mar 2010 17:13:40 +0100 Subject: [Gllug] Flash 10 on amd64 In-Reply-To: (JLMS's message of "Sat, 27 Mar 2010 13:56:18 +0000") References: <1269688875.19376.5.camel@bossclass> Message-ID: <87sk7kpgp7.fsf@spindle.srvr.nix> On 27 Mar 2010, JLMS told this: > ---- > #!/bin/sh > GDK_NATIVE_WINDOWS=true /usr/lib/nspluginwrapper/i386/linux/npviewer.bin.real $* > ---- > > that seems to tell the binary to behave in the right way. This tells Gtk's drawing layer to create native X windows for every Gtk GUI element, rather than (as Gtk 2.17+ otherwise does) rendering a giant bitmap and sending it to the server (which should be indistinguishable from using native windows except for a reduction in flicker). Normally, I'd expect setting GDK_NATIVE_WINDOWS to improve correctness for old programs that expect native windows to be there, but not to affect performance (actually I'd expect performance to be somewhat worse with native windows on). From nix at esperi.org.uk Sun Mar 28 15:24:34 2010 From: nix at esperi.org.uk (Nix) Date: Sun, 28 Mar 2010 15:24:34 +0100 Subject: [Gllug] Flash 10 on amd64 In-Reply-To: <4BAE06B6.80703@technicalbloke.com> (general's message of "Sat, 27 Mar 2010 13:23:02 +0000") References: <1269688875.19376.5.camel@bossclass> <4BAE06B6.80703@technicalbloke.com> Message-ID: <87wrwwplr1.fsf@spindle.srvr.nix> On 27 Mar 2010, general stated: > Even when it works correctly Flash performance under Linux can be quite > underwhelming esp as the video resolution creeps up. My P4x3000 really > struggles to do fullscreen flash whereas it's silky smooth on my old > PIII Windows laptop :( That's on account of Flash video output being such that almost no video cards can do hardware scaling (great choice, guys), and the software scaler in Adobe's Flash player being distinctly not brilliant (last time I looked it didn't even use SSE). From general_email at technicalbloke.com Sun Mar 28 21:52:04 2010 From: general_email at technicalbloke.com (general_email at technicalbloke.com) Date: Sun, 28 Mar 2010 21:52:04 +0100 Subject: [Gllug] Cygwin enlightenment In-Reply-To: <4BAE78C6.7080000@technicalbloke.com> References: <4BAE0520.5020807@technicalbloke.com> <20100327141432.GH16801@cornerstonelinux.co.uk> <4BAE5D6D.6070309@technicalbloke.com> <4BAE78C6.7080000@technicalbloke.com> Message-ID: <4BAFC174.2050803@technicalbloke.com> general_email at technicalbloke.com wrote: > general_email at technicalbloke.com wrote: > Spoke a bit too soon there... > > This is what the using-specialnames bit of the docs reckons (I think)... > $ ./ddrescue.exe -n -v --input-position 121506468b --input-position > 121506468b > if=\device\harddisk1 of=Z:\dee.img Z:\log_file.log > ddrescue: cannot open input file: No such file or directory > Pah, twas nothing to do with cygwin, forgot ddrescue doesn't use if= and of= Doh! All working fine now :) Roger. From jjllmmss at googlemail.com Sun Mar 28 22:00:52 2010 From: jjllmmss at googlemail.com (JLMS) Date: Sun, 28 Mar 2010 22:00:52 +0100 Subject: [Gllug] Flash 10 on amd64 In-Reply-To: <1269737459.2212.76.camel@ubuntu.clocksoft.dom> References: <1269688875.19376.5.camel@bossclass> <1269737459.2212.76.camel@ubuntu.clocksoft.dom> Message-ID: On Sun, Mar 28, 2010 at 1:50 AM, Malcolm Blackmore wrote: > On Sat, 2010-03-27 at 13:56 +0000, JLMS wrote: >> >> I think I did, I  just cut and pasted, > > Where did you get this from?? > Lots of googling, I was really fed up with Flash not working properly. From essuu at ourshack.com Tue Mar 30 11:58:26 2010 From: essuu at ourshack.com (Simon Wilcox) Date: Tue, 30 Mar 2010 11:58:26 +0100 Subject: [Gllug] Analysing Exim logs Message-ID: <4BB1D952.9070909@ourshack.com> This must surely exist but my Google-fu is failing me :-( We run exim and process around 15,000 emails per day. I'm familiar with eximstats and we use this to produce a daily summary report. A lot of the mail we send is from e-commerce applications and clients have asked for reports detailing delivery failures originating from specific addresses (e.g. orders at example.com) so that they are forewarned when someone has entered their details incorrectly. Does anyone know of a tool that can report by sender, or a tool that understands the exim format and can parse logs into a database for further analysis ? I'm looking at building a parser and plonking a reporting tool on the front but I feel sure this must be a solved problem. Can anyone recommend a suitable tool ? Many thanks, Simon. From james.dutton at gmail.com Tue Mar 30 12:28:13 2010 From: james.dutton at gmail.com (James Courtier-Dutton) Date: Tue, 30 Mar 2010 12:28:13 +0100 Subject: [Gllug] Analysing Exim logs In-Reply-To: <4BB1D952.9070909@ourshack.com> References: <4BB1D952.9070909@ourshack.com> Message-ID: On 30 March 2010 11:58, Simon Wilcox wrote: > This must surely exist but my Google-fu is failing me :-( > > We run exim and process around 15,000 emails per day. I'm familiar with > eximstats and we use this to produce a daily summary report. > > A lot of the mail we send is from e-commerce applications and clients > have asked for reports detailing delivery failures originating from > specific addresses (e.g. orders at example.com) so that they are forewarned > when someone has entered their details incorrectly. > smtp is an unreliable mail delivery protocol. It has a "failed delivery" option, but hardly anybody uses it for security against spam. So, even if you do parse the logs for failure messages, it is unlikely to be 100% reliable. There are other options for checking email address accuracy. User enters email address in web site. web site sends a special email to that email address. Until the user respond/reply to the special email, the email is assumed to be invalid. Much like the method most email mailing lists use. Kind Regards James From essuu at ourshack.com Tue Mar 30 12:42:06 2010 From: essuu at ourshack.com (Simon Wilcox) Date: Tue, 30 Mar 2010 12:42:06 +0100 Subject: [Gllug] Analysing Exim logs In-Reply-To: References: <4BB1D952.9070909@ourshack.com> Message-ID: <4BB1E38E.8030700@ourshack.com> Hi James, On 30/3/10 12:28, James Courtier-Dutton wrote: > smtp is an unreliable mail delivery protocol. It has a "failed > delivery" option, but hardly anybody uses it for security against > spam. Yes, I'm familiar with the problems with SMTP mail transport and get to spend quite some time explaining them to our clients. > So, even if you do parse the logs for failure messages, it is unlikely > to be 100% reliable. I don't really care what happens to it once it's off our servers (well I do but not for the purposes of this discussion) I just need to report failures against the originating address. > There are other options for checking email address accuracy. > User enters email address in web site. > web site sends a special email to that email address. > Until the user respond/reply to the special email, the email is > assumed to be invalid. > Much like the method most email mailing lists use. Well yes, but that doesn't help when their mailbox is full and we get a 'Quota Exceeded' rejection. I probably picked the wrong example in my previous email, apologies for that. S. From jasper at pointless.net Tue Mar 30 21:50:29 2010 From: jasper at pointless.net (Jasper Wallace) Date: Tue, 30 Mar 2010 21:50:29 +0100 (BST) Subject: [Gllug] Analysing Exim logs In-Reply-To: <4BB1D952.9070909@ourshack.com> References: <4BB1D952.9070909@ourshack.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Tue, 30 Mar 2010, Simon Wilcox wrote: > This must surely exist but my Google-fu is failing me :-( > > We run exim and process around 15,000 emails per day. I'm familiar with > eximstats and we use this to produce a daily summary report. > > A lot of the mail we send is from e-commerce applications and clients > have asked for reports detailing delivery failures originating from > specific addresses (e.g. orders at example.com) so that they are forewarned > when someone has entered their details incorrectly. > > Does anyone know of a tool that can report by sender, or a tool that > understands the exim format and can parse logs into a database for > further analysis ? > > I'm looking at building a parser and plonking a reporting tool on the > front but I feel sure this must be a solved problem. > > Can anyone recommend a suitable tool ? you might be better off with sending the emails with a VERP style return path (and Errors-To:), and then having any bounces fed to a script that can extract the recipents address (or customer id or however you want to tag the emails) and reporting that way. That would be realtime and not rely on periodic log parsing etc. - -- [http://pointless.net/] [0x2ECA0975] -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) iQEVAwUBS7JkFQCB+Qwuygl1AQLpfQf9FFQHn+9o/fJbC0yOpI18VD/OrrovJcq+ 5GKzaP8ec1a5oJUiKL1xntIBRKsHF032PAjwrYznl23/DaqUxma9z4PFk4Rmc2hj R7NkRDfIJCGAqg0QSkdeTSfI7+I+fQ/McB/UOrq79TAp4H1hcBTUWefzKjp2gp2K D5oeOk4DDsfse3jAlNzbDF2wWctGNrCL+r+kcXr5lAD1x3HA7Z6Em7BwmA0kXGB5 3L5cL33yZJDXsSQ2ZXMVB5NdX02D/9RgQnJVtq+H20GAYOwUtvkakdDIABFMPPMM tpewEQsNoeierXLCzN26U1F+1YRD1KUfvghdV74jB1uSyFV5mu72xw== =pPio -----END PGP SIGNATURE----- From td at bloogaloo.co.uk Wed Mar 31 23:17:41 2010 From: td at bloogaloo.co.uk (tid) Date: Wed, 31 Mar 2010 23:17:41 +0100 Subject: [Gllug] open-source workflow / process software? Message-ID: [slightly off-topic - sorry] I have a friend who is looking for some sort of software that can flag up events happening / happening at certain times or within certain constraints. He has a group of operators (!) looking after a selection of solaris/unix boxes and, gasp, openVMS. They need to get notification when a particular job hasn't arrived or hasn't completed. The likes of nagios / snmp-trap software don't feel right for this scenario as some of the tasks that come in aren't necessarily computer events, and he'd like something that could deal with an escalation procedure that adapts on a regular (potentially daily) basis. The scenario he sites is where a trade feed from an exchange runs too slow for 4 minutes then reboot the daemon, or if slow for 30 minutes then flag up that someone should phone the client. Currently, the operations team works off multiple check sheets, which is apparently error prone and fiddly. He's really looking anything that does process-flow - does anyone know of such an beast? Tid From walter.stanish at saffrondigital.com Wed Mar 31 23:33:23 2010 From: walter.stanish at saffrondigital.com (Walter Stanish) Date: Wed, 31 Mar 2010 23:33:23 +0100 Subject: [Gllug] open-source workflow / process software? References: Message-ID: You are really looking at two different types of systems. 1. Issue tracking There are many open source issue tracking systems, often with a plethora of features you don't need. Hacking your own may be quicker! (Eg: mantis, bugzilla, redmine.) 2. Real time messaging This sort of thing is often called 'enterprise message bus' but can basically be anything that sends messages about. HTTP, SMTP, instant messaging, etc. XMPP is an open protocol that would be a good option for integration here if you want to be able to support retries and push-based delivery to people while they are roaming about, on their mobile, etc. - Some are store/forward, some realtime - Some queue forever, some die after awhile failing to deliver (ala SMS from the GSM world) The above are just vague ideas based on the info you gave. Issues trackers are better for complex process flow. Something custom-hacked is better for simplistic scenarios where custom integration is the key and requirements are unlikely to change significantly over time (eg: bring in 2 departments of nontechnical personnel...) - Walter, in sunny .au -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.gllug.org.uk/pipermail/gllug/attachments/20100331/ee0c0c6c/attachment.htm