From matt at ucc.asn.au Sat Apr 2 10:35:11 2005 From: matt at ucc.asn.au (Matt Johnston) Date: Sat Apr 2 10:35:30 2005 Subject: [tech] martello speed In-Reply-To: References: Message-ID: <20050402023511.GA144901@morwong.ucc.gu.uwa.edu.au> On Wed, Mar 23, 2005 at 06:41:43PM +0800, James Andrewartha wrote: > The new SATA controller has been installed in martello, and I've been > benchmarking various RAID configurations with bonnie++. The results are > RAID 5 > ------ Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP martello 4G 38175 92 65903 27 20999 8 26989 60 58108 12 170.9 0 > > (fiddling with blockdev --setra didn't do much) *snip* > The bare disk result bears out the hdparm value of 55MB/s. The question is, > where's the performance improvement that should happend for RAID 1+0 and > RAID 5? The slow resync and hdparm result during array construction says to > me that the ports on the controller are not independent. > > A little searching finds > http://forums.storagereview.net/index.php?showtopic=17794&hl=promise which > claims "The linux driver sends only one command to the board at a time." We've been playing with various drivers and things over the past week with some results. The sx8.c driver in the kernel does indeed only queue one command at a time. There's a parameter which can be tweaked (CARM_MAX_Q), but we found that would cause corruption for particular values - knowing safe values seemed too hard. The author of the driver had only tested on prototype hardware, so didn't know what would be safe. Updating to the latest firmware didn't help either. There's a seperate open source driver available at promise.com, which has better performance than the sx8.c driver in the kernel, so we're using that one. The main downside is that it will probably bitrot more than the in-kernel sx8.c driver. In the process of testing we discovered a bug in Linux's ext2 driver, and hit another bug (already known and fixed) in the ext3 driver... These have now been patched on martello. I've played with --setra on the logical volume itself, and that does seem to give performance improvements. See below for some runs of bonnie++ with various settings. Anyway, the drives/controller/FS now seem fairly stable in their current configuration. Matt Single runs: ============ Plain RAID5, /dev/md0 Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP martello 4G 44153 96 71542 16 20943 7 34897 77 144465 30 447.8 0 LVM (default readahead of 256 for /dev/sd{b..e} and /dev/mapper/reliable-home Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP martello 4G 38025 96 71307 31 28362 12 31777 67 91284 17 378.2 1 LVM, readahead 512 for /dev/sd{b..e}, 2048 for /dev/mapper/reliable-home Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP martello 4G 37334 95 73596 37 30293 12 42510 92 173115 37 407.2 1 Concurrent runs (4x bonnie++ with -y flag): =========================================== RAID5 Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP martello 4G 7706 17 7740 1 5117 1 13818 30 15072 3 91.7 0 LVM with default readahead Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP martello 4G 7566 19 9762 5 5677 2 11928 26 13976 2 79.1 0 LVM with increased readahead Version 1.03 ------Sequential Output------ --Sequential Input- --Random- -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks-- Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP martello 4G 9474 24 9354 4 6870 2 19414 42 20497 4 79.7 0 From fryers at rcpt.to Mon Apr 4 06:54:42 2005 From: fryers at rcpt.to (Simon Fryer) Date: Mon Apr 4 06:54:50 2005 Subject: [tech] martello speed In-Reply-To: <20050402023511.GA144901@morwong.ucc.gu.uwa.edu.au> References: <20050402023511.GA144901@morwong.ucc.gu.uwa.edu.au> Message-ID: <20050403225442.GH71143@rcpt.to> All, > A while ago Matt Johnston tapped: [chomp] > Anyway, the drives/controller/FS now seem fairly stable in > their current configuration. Kewl! No the next logical question, and please excuse the management speak - what sort of timeframe are we looking at to move /home to martello? Simon -- ------------------------------------------------------------------------ "Well, an engineer is not concerned with the truth; that is left to philosophers and theologians: the prime concern of an engineer is the utility of the final product." Lectures on the Electrical Properties of Materials, L.Solymar, D.Walsh From zanchey at ucc.gu.uwa.edu.au Tue Apr 5 10:48:51 2005 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Tue Apr 5 10:49:03 2005 Subject: [tech] Cobbler's fan replaced Message-ID: Frenchie and I just replaced Cobbler's fan, so it runs (and oh-so-quietly, too). Windows is now booting, too, which is nice. Cheers, David Adam zanchey@ From zanchey at ucc.gu.uwa.edu.au Tue Apr 5 14:28:30 2005 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Tue Apr 5 14:28:52 2005 Subject: [tech] CC on Azure Message-ID: Hi all, Decided to look at why BZFlag won't compile on the architecture it was originally written for, and it turns out that the MIPS C Compiler on Azure has expired. Given that almost nobody uses Azure for, well, anything these days, if it's going to cost us money then I don't really think it's worth it. However, I get the impression that we might be able to work something out. If anyone knows anything about the licensing for Azure/Charteruse (sp?), can you send me an e-mail or something? Otherwise, I'll call the Freecall number for Australian SGI support tomorrow (probably from the clubroom phone (-: ) Relevant excerpts from the logs from BZFlag's configure script are included below; the full file is in /home/ucc/zanchey/bzflag-2.0.2.20050318/config.log (should be world-readable). Thanks, David Adam zanchey@ configure:2330: checking for C compiler default output file name configure:2333: cc conftest.c >&5 Feature has expired (-10,32) The MIPSpro C Compiler (license FEATURE string = cc) requires a license password. For license installation and trouble shooting information visit the web page: http://www.sgi.com/Support/Licensing/install_docs.html To obtain a Permanent license (proof of purchase required) or an Evaluation license please visit our license request web page: http://www.sgi.com/Products/license.html or send a blank email message to: license@sgi.com In North America, Silicon Graphics' customers may request Permanent licenses by sending a facsimile to: (650) 932-0537 or by calling our technical support hotline 1-800-800-4SGI If you are Outside of North America or you are not a Silicon Graphics support customer then contact your local support provider. cc ERROR: cannot exec /usr/lib32/cmplrs/fec configure:2336: $? = 2 configure: failed program was: | /* confdefs.h. */ | | #define PACKAGE_NAME "" | #define PACKAGE_TARNAME "" | #define PACKAGE_VERSION "" | #define PACKAGE_STRING "" | #define PACKAGE_BUGREPORT "" | #define PACKAGE "bzflag" | #define VERSION "2.0.2.20050405" | #define BZ_BUILD_OS "irix6.5" | #define BZ_BUILD_USER "zanchey" | #define BUILD_REGEX 1 | /* end confdefs.h. */ | | int | main () | { | | ; | return 0; | } configure:2374: error: C compiler cannot create executables From david at luyer.net Wed Apr 6 10:21:15 2005 From: david at luyer.net (David Luyer) Date: Wed Apr 6 10:21:39 2005 Subject: [tech] WAIX and Internet2 access working In-Reply-To: <20050323163955.GA2083@angrygoats.net> Message-ID: <200504060221.j362LGA6010162@mailout1.pacific.net.au> > There is a large iptables chain FREENETS on madako. This allows networks > to be matched based on whether they are reached via Grangenet or the > various non-byte charged bits of AARNET. I've set things up so that at > boot time, hosts that are limited to FREENETS have: > access to all of FREENETS (-A FORWARD -d 130.95.13.18 -j FREENETS) > no access to anything else (-A FORWARD -d 130.95.13.18 -j DROP) Does your code cover the case where a more-specific is charged (in the global BGP tables) but the less-specific is seen in the WAIX table? You need an explicit 'DROP' for such prefixes. (very annoying prefixes for all kinds of reasons) David. From nick at ucc.gu.uwa.edu.au Fri Apr 8 22:38:01 2005 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Fri Apr 8 22:38:12 2005 Subject: [tech] Ubuntised cobbler Message-ID: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> cobbler's booting a shiny new Ubuntu hoary install happily, now. NIS users and manbo:/space/home/away home directories. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal From trs80 at ucc.gu.uwa.edu.au Sat Apr 9 21:39:52 2005 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Mon Apr 11 12:40:52 2005 Subject: [tech] Ubuntised cobbler In-Reply-To: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> References: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> Message-ID: On Fri, 8 Apr 2005, Nick Bannon wrote: > cobbler's booting a shiny new Ubuntu hoary install happily, now. NIS > users and manbo:/space/home/away home directories. Which is horribly incsecure and nuked the existing homedirs :-( -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Wheel Member http://trs80.ucc.asn.au/ #| what squirrels do best | [ "There's nobody getting rich writing ]| -- Collect and hide your | [ software that I know of" -- Bill Gates, 1980 ]\ nuts." -- Acid Reflux #231 / From nick at ucc.gu.uwa.edu.au Sat Apr 9 23:19:53 2005 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Mon Apr 11 12:41:39 2005 Subject: [tech] Ubuntised cobbler In-Reply-To: References: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> Message-ID: <20050409151953.GA7481@ucc.gu.uwa.edu.au> On Sat, Apr 09, 2005 at 09:39:52PM +0800, James Andrewartha wrote: > Which is horribly incsecure and nuked the existing homedirs :-( Um. Yes. Sorry. Is being insecure a reason not to do it? I had thought that .../away directories were for that purpose, and we'd only try to keep /home vaguely physically secure. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal From grahame at angrygoats.net Mon Apr 11 13:32:33 2005 From: grahame at angrygoats.net (Grahame Bowland) Date: Mon Apr 11 13:32:44 2005 Subject: [tech] Ubuntised cobbler In-Reply-To: <20050409151953.GA7481@ucc.gu.uwa.edu.au> References: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> <20050409151953.GA7481@ucc.gu.uwa.edu.au> Message-ID: <20050411053233.GB16140@angrygoats.net> On Sat, Apr 09, 2005 at 11:19:53PM +0800, Nick Bannon wrote: > On Sat, Apr 09, 2005 at 09:39:52PM +0800, James Andrewartha wrote: > > Which is horribly incsecure and nuked the existing homedirs :-( > > Um. Yes. Sorry. > > Is being insecure a reason not to do it? I had thought that .../away > directories were for that purpose, and we'd only try to keep /home > vaguely physically secure. What is the insecurity being referred to... no lilo password? From nick at ucc.gu.uwa.edu.au Mon Apr 11 13:49:54 2005 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Mon Apr 11 13:50:08 2005 Subject: [tech] Ubuntised cobbler In-Reply-To: <20050411053233.GB16140@angrygoats.net> References: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> <20050409151953.GA7481@ucc.gu.uwa.edu.au> <20050411053233.GB16140@angrygoats.net> Message-ID: <20050411054954.GG10489@ucc.gu.uwa.edu.au> On Mon, Apr 11, 2005 at 01:32:33PM +0800, Grahame Bowland wrote: > What is the insecurity being referred to... no lilo password? No BIOS or GRUB password, open case, CD bootable, in the open clubroom with the 130.95.13.64/26 ethernet hanging out the back. It's using the away directories when it's booted into Windows, isn't it? Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal From grahame at angrygoats.net Mon Apr 11 15:12:33 2005 From: grahame at angrygoats.net (Grahame Bowland) Date: Mon Apr 11 15:12:47 2005 Subject: [tech] Ubuntised cobbler In-Reply-To: <20050411054954.GG10489@ucc.gu.uwa.edu.au> References: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> <20050409151953.GA7481@ucc.gu.uwa.edu.au> <20050411053233.GB16140@angrygoats.net> <20050411054954.GG10489@ucc.gu.uwa.edu.au> Message-ID: <20050411071233.GA16624@angrygoats.net> On Mon, Apr 11, 2005 at 01:49:54PM +0800, Nick Bannon wrote: > On Mon, Apr 11, 2005 at 01:32:33PM +0800, Grahame Bowland wrote: > > What is the insecurity being referred to... no lilo password? > > No BIOS or GRUB password, open case, CD bootable, in the open clubroom > with the 130.95.13.64/26 ethernet hanging out the back. > > It's using the away directories when it's booted into Windows, isn't it? Like it matters when you can just steal the ethernet out the back and mount /away anyway :-) I think trs80 is just grumpy ubuntu is clobbering Debian :-) From davyd at madeley.id.au Mon Apr 11 15:33:22 2005 From: davyd at madeley.id.au (Davyd Madeley) Date: Mon Apr 11 15:33:26 2005 Subject: [tech] Ubuntised cobbler In-Reply-To: <20050411071233.GA16624@angrygoats.net> References: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> <20050409151953.GA7481@ucc.gu.uwa.edu.au> <20050411053233.GB16140@angrygoats.net> <20050411054954.GG10489@ucc.gu.uwa.edu.au> <20050411071233.GA16624@angrygoats.net> Message-ID: <1113204802.8493.11.camel@pingu.madeley.id.au> On Mon, 2005-04-11 at 15:12 +0800, Grahame Bowland wrote: > Like it matters when you can just steal the ethernet out the back and > mount /away anyway :-) I think trs80 is just grumpy ubuntu is clobbering > Debian :-) Yeah. We shoulda installed SpaceLinux. It's the Linux that spacemen use. --d -- Davyd Madeley http://www.davyd.id.au/ PGP Fingerprint 08B0 341A 0B9B 08BB 2118 C060 2EDD BB4F 5191 6CDA From michal at ucc.asn.au Mon Apr 11 16:15:22 2005 From: michal at ucc.asn.au (Michal Gornisiewicz) Date: Mon Apr 11 16:15:28 2005 Subject: [tech] Ubuntised cobbler In-Reply-To: <20050411071233.GA16624@angrygoats.net> References: <20050408143800.GA473925@morwong.ucc.gu.uwa.edu.au> <20050409151953.GA7481@ucc.gu.uwa.edu.au> <20050411053233.GB16140@angrygoats.net> <20050411054954.GG10489@ucc.gu.uwa.edu.au> <20050411071233.GA16624@angrygoats.net> Message-ID: <20050411081522.GA523700@morwong.ucc.gu.uwa.edu.au> On Mon, Apr 11, 2005 at 03:12:33PM +0800, Grahame Bowland wrote: > Like it matters when you can just steal the ethernet out the back and > mount /away anyway :-) I think trs80 is just grumpy ubuntu is clobbering > Debian :-) Isn't Ubuntu Debian based? ie. just as crap! Mg From adrian at ucc.gu.uwa.edu.au Fri Apr 15 10:47:42 2005 From: adrian at ucc.gu.uwa.edu.au (Adrian Chadd) Date: Fri Apr 15 10:47:51 2005 Subject: [tech] LDAP! Message-ID: <20050415024741.GA6765@ucc.gu.uwa.edu.au> Grahame sent it to the wrong place. From: grahame@angrygoats.net (Grahame Bowland) To: tech@ucs.uwa.edu.au Subject: Migrating UCC to LDAP Created: 15/04/2005 01:19:08 Hi guys I have created a clean debian machine over at UCS in a Xen. That machine is running an LDAP server, and all local authentication is quite happily running from that server. The server has been populated with the UCC groups and accouts automatically generated from the NIS maps. My migration plan is as follows; * install LDAP server on martello * have LDAP server update from NIS periodically (easy) this involves creating posixAccount/shadowAccount and group entries. Already pretty much finished. * gradually move club room machines over to LDAP auth easy for Debian boxen. easy for MacOS X other machines on a case-by-case basis * once all machines are over, if we still need NIS make NIS slave from the LDAP config. This is just a simple script to parse a couple of LDAP queries and spit out /var/yp/passwd and /var/yp/group files, then run Make in that directory. Anyway, if all goes to plan mermaid will be doing LDAP authentication as of tomorrow night. If that works, I'll move over mussel, madako and other machines as I have the time. If anyone has comments? Also, as I'm going to be fiddling remotely at least some of the time please please let me know if you're fiddling as well. 'grahame' on austnet is a good way to grab me. Have fun Grahame .. then From: grahame@angrygoats.net (Grahame Bowland) To: tech@ucs.uwa.edu.au Subject: Mussel now doing LDAP Created: 15/04/2005 02:01:12 If you want to put it back to nis copy /etc/nsswitch.conf from the /etc/pre-ldap directory. Seems to work though :-) From dayta at ucc.gu.uwa.edu.au Fri Apr 15 10:52:15 2005 From: dayta at ucc.gu.uwa.edu.au (Leighton Haynes) Date: Fri Apr 15 10:52:19 2005 Subject: [tech] LDAP! In-Reply-To: <20050415024741.GA6765@ucc.gu.uwa.edu.au> References: <20050415024741.GA6765@ucc.gu.uwa.edu.au> Message-ID: <20050415025215.GD23425@ucc.gu.uwa.edu.au> If it's going to be LDAP, why not use an active directory server as master. Then the windows boxes will be happy too. Leighton.... On Fri, Apr 15, 2005 at 10:47:42AM +0800, Adrian Chadd wrote: > > Grahame sent it to the wrong place. > > > From: > grahame@angrygoats.net (Grahame Bowland) > > > To: > tech@ucs.uwa.edu.au > > > Subject: > Migrating UCC to LDAP > > > Created: > 15/04/2005 01:19:08 > > Hi guys > > I have created a clean debian machine over at UCS in a Xen. That machine > is running an LDAP server, and all local authentication is quite happily > running from that server. The server has been populated with the UCC > groups and accouts automatically generated from the NIS maps. > > My migration plan is as follows; > * install LDAP server on martello > * have LDAP server update from NIS periodically (easy) > this involves creating posixAccount/shadowAccount and > group entries. Already pretty much finished. > * gradually move club room machines over to LDAP auth > easy for Debian boxen. > easy for MacOS X > other machines on a case-by-case basis > * once all machines are over, if we still need NIS make > NIS slave from the LDAP config. This is just a simple script > to parse a couple of LDAP queries and spit out /var/yp/passwd > and /var/yp/group files, then run Make in that directory. > > Anyway, if all goes to plan mermaid will be doing LDAP authentication as > of tomorrow night. If that works, I'll move over mussel, madako and > other machines as I have the time. > > If anyone has comments? Also, as I'm going to be fiddling remotely at > least some of the time please please let me know if you're fiddling as > well. 'grahame' on austnet is a good way to grab me. > > Have fun > Grahame > > > > .. then > > > From: > grahame@angrygoats.net (Grahame Bowland) > > > To: > tech@ucs.uwa.edu.au > > > Subject: > Mussel now doing LDAP > > > Created: > 15/04/2005 02:01:12 > > If you want to put it back to nis copy /etc/nsswitch.conf from > the /etc/pre-ldap directory. Seems to work though :-) > -- #0421 113 305 - dayta@ucc.gu.uwa.edu.au "People demand freedom of speech as a compensation for the freedom of thought they never use." - Kierkegaard From trs80 at ucc.gu.uwa.edu.au Fri Apr 15 11:22:25 2005 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Fri Apr 15 11:22:32 2005 Subject: [tech] LDAP! In-Reply-To: <20050415024741.GA6765@ucc.gu.uwa.edu.au> References: <20050415024741.GA6765@ucc.gu.uwa.edu.au> Message-ID: On Fri, 15 Apr 2005, Adrian Chadd wrote: > Subject: > Mussel now doing LDAP > > > Created: > 15/04/2005 02:01:12 > > If you want to put it back to nis copy /etc/nsswitch.conf from > the /etc/pre-ldap directory. Seems to work though :-) Except for ssh.ucc.asn.au, which uses dropbear. -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Wheel Member http://trs80.ucc.asn.au/ #| what squirrels do best | [ "There's nobody getting rich writing ]| -- Collect and hide your | [ software that I know of" -- Bill Gates, 1980 ]\ nuts." -- Acid Reflux #231 / From trent at ucc.asn.au Tue Apr 19 16:22:37 2005 From: trent at ucc.asn.au (Trent Lloyd) Date: Tue Apr 19 16:22:55 2005 Subject: [tech] power cut + seven Message-ID: <4264BFCD.4080906@ucc.asn.au> Hi Guys, After the power cut, seven doesn't seem to have come back. (It's the desktop compaq machine on top of the sun) As I and others I would ask to look at it will be in canberra until sunday, could someone please take a look for me? It possibly just needs power cycling or something. TIA Cheers, Trent From zanchey at ucc.gu.uwa.edu.au Tue Apr 19 16:48:44 2005 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Tue Apr 19 16:48:49 2005 Subject: [tech] Power failure aftermath Message-ID: So. At about 10:15 this morning, the power to the primary machine room circuit failed, and every machine on it dropped off. After some stellar work by Matt Johnston, David Basden and James French, as well as guest appearances by Graheme Bowland and me, most things are now working again (whoever wired Piggery's power lines to the reset switch deserves a good kicking). The following things still do not work: - Dispense (won't run from Mussel, and is kinda screwy on Morwong). Coke members, expect to do a LOT of refunding: at this stage, there is only stuff left in slots 5 and 6, but the dispense client isn't reflecting this on Morwong. - Music/Webcam on Maroon: This machine won't start up. I'm looking at it now but I don't think it will be up until tomorrow, unless someone else comes in and has a look (if you do, can you let me know) - Mussel is a bit funky - it will randomly lock up for a brief period of time on my SSH sessions. Unreproducible, but it's happened three times while I've written this e-mail. - Snack machine is dead, but that's been the case for about a week now. I suspect the DEC server is having issues. If you find anything else that is broken, please reply on-list so one of our Wheel members who isn't schmoozing it up in Canberra (sorry guys, but I'm pretty sure that's what Canberra was -designed- for) can have a look. Cheers, David Adam UCC Wheel Member zanchey@ From zanchey at ucc.gu.uwa.edu.au Tue Apr 19 16:55:26 2005 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Tue Apr 19 16:55:42 2005 Subject: [tech] power cut + seven In-Reply-To: <4264BFCD.4080906@ucc.asn.au> References: <4264BFCD.4080906@ucc.asn.au> Message-ID: On Tue, 19 Apr 2005, Trent Lloyd wrote: > After the power cut, seven doesn't seem to have come back. (It's the > desktop compaq machine on top of the sun) I take it you mean the one that's not the IBM? The one with the Linux casebadge? > As I and others I would ask to look at it will be in canberra until > sunday, could someone please take a look for me? It possibly just needs > power cycling or something. It's on and up, responding to pings at rattle.bur.st and running DNS, by the looks of things. No SSH though, AFAICT. Any other suggestions? I'm locking the machine room now, so it might have to wait until tomorrow unless you can persuade a Wheel member with keys to come down tonight. Cheers, David Adam UCC Wheel Member zanchey@ From lathiat at bur.st Tue Apr 19 17:03:21 2005 From: lathiat at bur.st (Trent Lloyd) Date: Tue Apr 19 17:03:41 2005 Subject: [tech] power cut + seven In-Reply-To: References: <4264BFCD.4080906@ucc.asn.au> Message-ID: <4264C959.6000003@bur.st> David Adam wrote: > On Tue, 19 Apr 2005, Trent Lloyd wrote: > > >>After the power cut, seven doesn't seem to have come back. (It's the >>desktop compaq machine on top of the sun) > > > I take it you mean the one that's not the IBM? The one with the Linux > casebadge? No thats rattle.bur.st > > >>As I and others I would ask to look at it will be in canberra until >>sunday, could someone please take a look for me? It possibly just needs >>power cycling or something. > > > It's on and up, responding to pings at rattle.bur.st and running DNS, by > the looks of things. No SSH though, AFAICT. Nope, SSH doesn't work because its firewalled. Mine is a "Compaq Deskpro" case, horizontally orientated. > Any other suggestions? I'm locking the machine room now, so it might have > to wait until tomorrow unless you can persuade a Wheel member with keys to > come down tonight. > Check the right machine. :) Thanks for looking! Cheers, Trent > Cheers, > > David Adam > UCC Wheel Member > zanchey@ From lathiat at bur.st Tue Apr 19 17:08:37 2005 From: lathiat at bur.st (Trent Lloyd) Date: Tue Apr 19 17:08:44 2005 Subject: [tech] power cut + seven In-Reply-To: <4264C959.6000003@bur.st> References: <4264BFCD.4080906@ucc.asn.au> <4264C959.6000003@bur.st> Message-ID: <4264CA95.3080303@bur.st> My appologies, it may be in the middle cabinet and not where i said (i think it got moved by trs80) Cheers, Trent Trent Lloyd wrote: > > David Adam wrote: > >>On Tue, 19 Apr 2005, Trent Lloyd wrote: >> >> >> >>>After the power cut, seven doesn't seem to have come back. (It's the >>>desktop compaq machine on top of the sun) >> >> >>I take it you mean the one that's not the IBM? The one with the Linux >>casebadge? > > No thats rattle.bur.st > > >> >>>As I and others I would ask to look at it will be in canberra until >>>sunday, could someone please take a look for me? It possibly just needs >>>power cycling or something. >> >> >>It's on and up, responding to pings at rattle.bur.st and running DNS, by >>the looks of things. No SSH though, AFAICT. > > Nope, SSH doesn't work because its firewalled. > > Mine is a "Compaq Deskpro" case, horizontally orientated. > > >>Any other suggestions? I'm locking the machine room now, so it might have >>to wait until tomorrow unless you can persuade a Wheel member with keys to >>come down tonight. >> > > Check the right machine. :) > > Thanks for looking! > > Cheers, > Trent > > >>Cheers, >> >>David Adam >>UCC Wheel Member >>zanchey@ > > From zanchey at ucc.gu.uwa.edu.au Tue Apr 19 18:01:11 2005 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Tue Apr 19 18:01:15 2005 Subject: [tech] power cut + seven In-Reply-To: <4264CA95.3080303@bur.st> References: <4264BFCD.4080906@ucc.asn.au> <4264C959.6000003@bur.st> <4264CA95.3080303@bur.st> Message-ID: Trent, On Tue, 19 Apr 2005, Trent Lloyd wrote: > My appologies, it may be in the middle cabinet and not where i said (i > think it got moved by trs80) Right. Well, that explains it. I won't be around tomorrow until the late afternoon because of PROSH but I'm sure someone with machine room keys will have a look at it. Cheers, David Adam zanchey@ From mtearle at ucc.gu.uwa.edu.au Tue Apr 19 21:51:59 2005 From: mtearle at ucc.gu.uwa.edu.au (Mark Tearle) Date: Tue Apr 19 21:52:08 2005 Subject: [tech] Power failure aftermath In-Reply-To: References: Message-ID: On Tue, 19 Apr 2005, David Adam wrote: > - Snack machine is dead, but that's been the case for about a week now. I > suspect the DEC server is having issues. > > If you find anything else that is broken, please reply on-list so one of > our Wheel members who isn't schmoozing it up in Canberra (sorry guys, but > I'm pretty sure that's what Canberra was -designed- for) can have a look. > > Cheers, > > David Adam > UCC Wheel Member > zanchey@ Hmm, a few things likely to have happened there a) the daemon that drives it on mussel has gone AWOL. Bernard presently knows the magic to start it, VendServer.py in his home dir is a likely candidate b) the DEC server is actually really unhappy and many chickens might have to be sacrificied. I'll poke TRS80 tommorow to have a look at it. Mark "boozing" -- Mark Tearle - mtearle@ucc.gu.uwa.edu.au Fish! From davidb-0624 at rcpt.to Tue Apr 19 22:33:54 2005 From: davidb-0624 at rcpt.to (David Basden) Date: Tue Apr 19 22:34:18 2005 Subject: [tech] Power failure aftermath In-Reply-To: References: Message-ID: <20050419143354.GA21531@chastity.shikita.com.au> On Tue, Apr 19, 2005 at 09:51:59PM +0800, Mark Tearle wrote: > b) the DEC server is actually really unhappy and many chickens might > have to be sacrificied. I'll poke TRS80 tommorow to have a look at it. The one in the same rack as the switches, which I suspect has serial consoles on it, didn't come up. It seems to be fried. David From dagobah at ucc.gu.uwa.edu.au Wed Apr 20 09:18:24 2005 From: dagobah at ucc.gu.uwa.edu.au (Bernard Blackham) Date: Wed Apr 20 09:18:36 2005 Subject: [tech] Power failure aftermath In-Reply-To: References: Message-ID: <20050420011824.GA7162@ucc.gu.uwa.edu.au> On Tue, Apr 19, 2005 at 04:48:44PM +0800, David Adam wrote: > - Dispense (won't run from Mussel, and is kinda screwy on Morwong). Coke > members, expect to do a LOT of refunding: at this stage, there is only > stuff left in slots 5 and 6, but the dispense client isn't reflecting this > on Morwong. Should be fixed. Involves killing dispense on mermaid, running minicom, typing some random commands, and restarting dispense. > - Music/Webcam on Maroon: This machine won't start up. I'm looking at it > now but I don't think it will be up until tomorrow, unless someone else > comes in and has a look (if you do, can you let me know) Not sure about this. > - Snack machine is dead, but that's been the case for about a week now. I > suspect the DEC server is having issues. Should also be fixed. It no longer goes via the DEC, now plugs into the back of mermaid (perhaps via a patch lead to the patch panel IIRC). > If you find anything else that is broken, please reply on-list so one of > our Wheel members who isn't schmoozing it up in Canberra (sorry guys, but > I'm pretty sure that's what Canberra was -designed- for) can have a look. Yours sleepily, Bernard. From zanchey at ucc.gu.uwa.edu.au Wed Apr 20 17:16:56 2005 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Wed Apr 20 17:17:08 2005 Subject: [tech] Power failure aftermath In-Reply-To: <20050420011824.GA7162@ucc.gu.uwa.edu.au> References: <20050420011824.GA7162@ucc.gu.uwa.edu.au> Message-ID: On Wed, 20 Apr 2005, Bernard Blackham wrote: > On Tue, Apr 19, 2005 at 04:48:44PM +0800, David Adam wrote: > > - Dispense (won't run from Mussel, and is kinda screwy on Morwong). Coke > > members, expect to do a LOT of refunding: at this stage, there is only > > stuff left in slots 5 and 6, but the dispense client isn't reflecting this > > on Morwong. > > Should be fixed. Involves killing dispense on mermaid, running minicom, > typing some random commands, and restarting dispense. Are these documented anywhere? Just for eventualities like this? > > - Music/Webcam on Maroon: This machine won't start up. I'm looking at it > > now but I don't think it will be up until tomorrow, unless someone else > > comes in and has a look (if you do, can you let me know) > > Not sure about this. Fixed. Network cable not plugged in, and there was a disk in it labelled "Macintosh System 7 Disk Tools", which confused me no end. After several iatrogenic (i.e. MY BAD) issues involved in moving it from a low altitude to its usual high one, it now works all OK. > > - Snack machine is dead, but that's been the case for about a week now. I > > suspect the DEC server is having issues. > > Should also be fixed. It no longer goes via the DEC, now plugs into the > back of mermaid (perhaps via a patch lead to the patch panel IIRC). I cannot find this cable. I've tried, but Mermaid is in the cabinet, it's got another machine sitting on top of it, and nothing is labelled. And - CRAP - I think I screwed up the Coke machine again while I was fiddling. Will check now before I go home. Also: Trent, Seven is up, but it appears unhappy. Check dmesg - a lot of services apparently didn't start. It also doesn't boot without a keyboard, and I managed to screw up the KVM connection so there's now another keyboard in that cabinet. Didn't want to screw anything around, but you might want to look at setting the BIOS to boot without a keyboard. Thanks to Bernard, Shig and Frenchie for their assistance in getting the rest of the clubroom on-line. Trying to decide whether I've acheived more good than bad today. Possibly. David Adam UCC Wheel Member From zanchey at ucc.gu.uwa.edu.au Wed Apr 20 18:19:10 2005 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Wed Apr 20 18:19:14 2005 Subject: [tech] Power failure aftermath In-Reply-To: References: <20050420011824.GA7162@ucc.gu.uwa.edu.au> Message-ID: > got another machine sitting on top of it, and nothing is labelled. And - > CRAP - I think I screwed up the Coke machine again while I was fiddling. > Will check now before I go home. Alwyn engineered the lock open, and plugged the serial cable into the other port. Dispense works again now. Also, for anyone that's looking for it, meito's HDD is in the Machine Room. Snack machine is still a no-go. Work is in progress. Thanks, David Adam zanchey@ From lathiat at bur.st Wed Apr 20 20:58:29 2005 From: lathiat at bur.st (Trent Lloyd) Date: Wed Apr 20 20:58:46 2005 Subject: [tech] Power failure aftermath In-Reply-To: References: <20050420011824.GA7162@ucc.gu.uwa.edu.au> Message-ID: <426651F5.2000905@bur.st> > > Also: Trent, Seven is up, but it appears unhappy. Check dmesg - a lot of > services apparently didn't start. It also doesn't boot without a keyboard, > and I managed to screw up the KVM connection so there's now another > keyboard in that cabinet. Didn't want to screw anything around, but you > might want to look at setting the BIOS to boot without a keyboard. Thanks, its a pretty old legacy system and theres probably lots of broken crap that is nothing to worry about :) All the important stuff is running. > Trying to decide whether I've acheived more good than bad today. Possibly. My machine is fixed, thats good :) Cheers, Trent > David Adam > UCC Wheel Member From bernard at blackham.com.au Wed Apr 20 10:20:15 2005 From: bernard at blackham.com.au (Bernard Blackham) Date: Thu Apr 21 15:11:37 2005 Subject: [tech] Power failure aftermath In-Reply-To: References: Message-ID: <20050420022015.GA5016@blackham.com.au> On Tue, Apr 19, 2005 at 04:48:44PM +0800, David Adam wrote: > If you find anything else that is broken, please reply on-list so one of > our Wheel members who isn't schmoozing it up in Canberra (sorry guys, but > I'm pretty sure that's what Canberra was -designed- for) can have a look. Can somebody please hit the big blue button on migimaki (siting atop of mussel)? Ta. Bernard. -- Bernard Blackham From bernard at blackham.com.au Wed Apr 20 17:48:46 2005 From: bernard at blackham.com.au (Bernard Blackham) Date: Thu Apr 21 15:19:31 2005 Subject: [tech] Power failure aftermath In-Reply-To: References: <20050420011824.GA7162@ucc.gu.uwa.edu.au> Message-ID: <20050420094845.GA4484@blackham.com.au> On Wed, Apr 20, 2005 at 05:16:56PM +0800, David Adam wrote: > > > - Dispense (won't run from Mussel, and is kinda screwy on Morwong). Coke > > > members, expect to do a LOT of refunding: at this stage, there is only > > > stuff left in slots 5 and 6, but the dispense client isn't reflecting this > > > on Morwong. > > > > Should be fixed. Involves killing dispense on mermaid, running minicom, > > typing some random commands, and restarting dispense. > > Are these documented anywhere? Just for eventualities like this? The random commands? Not really. They're truly random. It takes a difference sequence of random commands each time to make it happy again. > > Should also be fixed. It no longer goes via the DEC, now plugs into the > > back of mermaid (perhaps via a patch lead to the patch panel IIRC). Ermmm. Read what I mean, not what I say :P ... s/mermaid/mussel/ > Trying to decide whether I've acheived more good than bad today. > Possibly. If not wheel, who would break stuff? ;) Bernard. -- Bernard Blackham From nick at ucc.gu.uwa.edu.au Sun Apr 24 17:30:47 2005 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Sun Apr 24 17:30:54 2005 Subject: [tech] Gigabit switch In-Reply-To: <200504240700.j3O700p0000142802@morwong.ucc.gu.uwa.edu.au> References: <200504240700.j3O700p0000142802@morwong.ucc.gu.uwa.edu.au> Message-ID: <20050424093047.GB4058@ucc.gu.uwa.edu.au> On Sun, Apr 24, 2005 at 03:00:00PM +0800, committee@ucc.gu.uwa.edu.au wrote: > The current agenda is: > grahame : Buy a gig switch. $485 for a 16 port D-Link from Harris Tech. [...] Which model's that? On http://www.ht.com.au/ I can see the unmanaged DGS-1016D's for $578.00 . The D-Link DGS-1216T Smart switch looks better, at $802.00 . The specs say "port based VLANs", but I'm not sure that means 802.1q . http://www.dlink.com.au/Default.aspx?ArticleID=1016 The DGS-1248T _does_ say 802.1q, but it's $2,080 - are you sure that the 16 port version does it as well? http://www.dlink.com.au/Default.aspx?ArticleID=200 Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal From grahame at angrygoats.net Sun Apr 24 21:53:23 2005 From: grahame at angrygoats.net (Grahame Bowland) Date: Sun Apr 24 21:53:43 2005 Subject: [tech] Gigabit switch In-Reply-To: <20050424093047.GB4058@ucc.gu.uwa.edu.au> References: <200504240700.j3O700p0000142802@morwong.ucc.gu.uwa.edu.au> <20050424093047.GB4058@ucc.gu.uwa.edu.au> Message-ID: <20050424135323.GA30821@angrygoats.net> On Sun, Apr 24, 2005 at 05:30:47PM +0800, Nick Bannon wrote: > On Sun, Apr 24, 2005 at 03:00:00PM +0800, committee@ucc.gu.uwa.edu.au wrote: > > The current agenda is: > > grahame : Buy a gig switch. $485 for a 16 port D-Link from Harris Tech. > [...] > > Which model's that? On http://www.ht.com.au/ I can see the unmanaged > DGS-1016D's for $578.00 . > > The D-Link DGS-1216T Smart switch looks better, at $802.00 . The specs > say "port based VLANs", but I'm not sure that means 802.1q . > http://www.dlink.com.au/Default.aspx?ArticleID=1016 > > The DGS-1248T _does_ say 802.1q, but it's $2,080 - are you sure that > the 16 port version does it as well? > http://www.dlink.com.au/Default.aspx?ArticleID=200 That price is the magic UWA price for a DGS-1016D. UWA staff can purchase from HT at that price. Port-based VLAN probably means the 1216T can do 802.1q based on the port, none of the MAC-based VLAN stuff. We wouldn't need that. Alex can probably give us pricing on a DGS-1216T.. Do we actually need VLANs on a UCC gig switch? The thing is almost twice the price; wouldn't all the things we'd want to connect at Gig be on the same (machine room) VLAN? From matt at ucc.asn.au Sun Apr 24 22:05:25 2005 From: matt at ucc.asn.au (Matt Johnston) Date: Sun Apr 24 22:05:29 2005 Subject: [tech] Gigabit switch In-Reply-To: <20050424135323.GA30821@angrygoats.net> References: <200504240700.j3O700p0000142802@morwong.ucc.gu.uwa.edu.au> <20050424093047.GB4058@ucc.gu.uwa.edu.au> <20050424135323.GA30821@angrygoats.net> Message-ID: <20050424140525.GG16816@ucc.gu.uwa.edu.au> On Sun, Apr 24, 2005 at 09:53:23PM +0800, Grahame Bowland wrote: > Do we actually need VLANs on a UCC gig switch? The thing is almost twice > the price; wouldn't all the things we'd want to connect at Gig be on > the same (machine room) VLAN? I could certainly see simple partitioning being useful, for connecting cobbler/cybium/other clubroom machines to the machineroom servers. Madako should be able to route at something a lot faster than 100mbit even if it can't do full 1000mbit. Matt From bogus@does.not.exist.com Sun Apr 24 23:15:09 2005 From: bogus@does.not.exist.com () Date: Sun Apr 24 23:15:59 2005 Subject: [tech] Gigabit switch In-Reply-To: <20050424140444.GF16816@ucc.gu.uwa.edu.au> References: <200504240700.j3O700p0000142802@morwong.ucc.gu.uwa.edu.au> <20050424093047.GB4058@ucc.gu.uwa.edu.au> <20050424135323.GA30821@angrygoats.net> <20050424140444.GF16816@ucc.gu.uwa.edu.au> Message-ID: <20050424151509.GA31699@angrygoats.net> On Sun, Apr 24, 2005 at 10:04:44PM +0800, Matt Johnston wrote: > On Sun, Apr 24, 2005 at 09:53:23PM +0800, Grahame Bowland wrote: > > Do we actually need VLANs on a UCC gig switch? The thing is almost twice > > the price; wouldn't all the things we'd want to connect at Gig be on > > the same (machine room) VLAN? > > I could certainly see simple partitioning being useful, for > connecting cobbler/cybium/other clubroom machines to the > machineroom servers. Madako should be able to route at > something a lot faster than 100mbit even if it can't do full > 1000mbit. Yep, but surely in that case buying two 16-port GigE switches and having one for each side is better than paying approximately the same price for 16 ports with VLANs? :) From acolyte at ucc.gu.uwa.edu.au Tue Apr 26 15:42:16 2005 From: acolyte at ucc.gu.uwa.edu.au (Andrew Bailey) Date: Tue Apr 26 15:42:29 2005 Subject: [tech] ssh.ucc.asn.au Message-ID: <20050426074216.GA223487@morwong.ucc.gu.uwa.edu.au> Hi, using charged ssh does not seem to work anymore. Is this on purpose? Andrew. -- "The hot dog eating contest is not only a beautiful display of athleticism, it is a fundamental way for citizens of all nations to display patriotism," - Wayne Norbitz From nick at ucc.gu.uwa.edu.au Tue Apr 26 15:52:52 2005 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Tue Apr 26 15:52:56 2005 Subject: [tech] ssh.ucc.asn.au In-Reply-To: <20050426074216.GA223487@morwong.ucc.gu.uwa.edu.au> References: <20050426074216.GA223487@morwong.ucc.gu.uwa.edu.au> Message-ID: <20050426075252.GD4909@ucc.gu.uwa.edu.au> On Tue, Apr 26, 2005 at 03:42:16PM +0800, Andrew Bailey wrote: > Hi, > using charged ssh does not seem to work anymore. Is this on purpose? It's not working with passwords right now, due to people trying to make it use LDAP. Matt was trying, someone else gave it a go as well. Any ideas? Keys are working, if you add (or get wheel to add) an entry to your ~acolyte/.ssh/authorized_keys . Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal From david at luyer.net Thu Apr 28 16:37:53 2005 From: david at luyer.net (David Luyer) Date: Thu Apr 28 16:38:10 2005 Subject: [tech] Gigabit switch In-Reply-To: <20050424135323.GA30821@angrygoats.net> Message-ID: <200504280837.j3S8bqJu011097@mailout1.pacific.net.au> > Port-based VLAN probably means the 1216T can do 802.1q based on the port, > none of the MAC-based VLAN stuff. We wouldn't need that. Alex can probably > give us pricing on a DGS-1216T.. Port-based VLAN usually means you can allocate eg. port 1,2,3=VLAN A; port 4,5,6=VLAN B but not dot1q trunk the VLANs. David.