From grahame at ucc.gu.uwa.edu.au Fri Nov 1 00:02:13 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:25:56 2004 Subject: [tech] Compiler on morwong Message-ID: <20021031160213.GA225208@morwong.ucc.gu.uwa.edu.au> We now have: Reading specs from /usr/local/lib/gcc-lib/alphaev5-dec-osf5.1/3.2.1/specs Configured with: ../configure --prefix=/usr/local Thread model: single gcc version 3.2.1 20021020 (prerelease) Seems to work, I've upgraded mutt, installed jed and various other things using it. The previous gcc had stopped producing binaries since the upgrade. gcj won't work, however this probably doesn't matter. It fails in the compile and I didn't have time to look into it, however the other compilers built fine. Cheers Grahame From grahame at ucc.gu.uwa.edu.au Fri Nov 1 00:05:18 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:25:57 2004 Subject: [tech] Sane mail Message-ID: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> Hi guys I'm running this plan by you guys, I'm willing to implement it if we want to do it. Basically the idea is: * remove all NFS mounts on mooneye * set up things so that mail lands on mooneye, mooneye then sends it on to morwong for delivery to mailboxes Advantages: * we don't lose our email when morwong vanishes * we're not relying on NFS locking to avoid corrupt mailboxes * seems just generally more sane Disadvantges: * ? Cheers, Grahame From mustang at ucc.gu.uwa.edu.au Fri Nov 1 00:08:41 2002 From: mustang at ucc.gu.uwa.edu.au (David Manchester) Date: Wed Oct 27 01:25:58 2004 Subject: [tech] Sane mail In-Reply-To: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021031160841.GA224954@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 12:05:18AM +0800, Grahame Bowland wrote: > Hi guys > > I'm running this plan by you guys, I'm willing to implement it if > we want to do it. > > Basically the idea is: > * remove all NFS mounts on mooneye > * set up things so that mail lands on mooneye, mooneye then > sends it on to morwong for delivery to mailboxes > > Advantages: > * we don't lose our email when morwong vanishes > * we're not relying on NFS locking to avoid corrupt mailboxes > * seems just generally more sane Do it. D. -- " I don't get mad.... I get stabby. " - William "Fat Tony" Williams. From grahame at ucc.gu.uwa.edu.au Fri Nov 1 04:01:14 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:25:58 2004 Subject: [tech] Microwave link down Message-ID: <20021031200114.GA243794@morwong.ucc.gu.uwa.edu.au> Waiting for the AAP men to let me know what's wrong with it... So I upgraded morwong to openssh-3.5p1. Seems to all work, the trick was to #define BROKEN_GETADDRINFO in config.h, then it compiled fine. It was also necessary to #define _OSF_SOURCE. I also upgraded mermaid to openssh 3.4p1 lest someone break into it. Cheers Grahame From mustang at ucc.gu.uwa.edu.au Fri Nov 1 06:59:37 2002 From: mustang at ucc.gu.uwa.edu.au (David Manchester) Date: Wed Oct 27 01:25:58 2004 Subject: [tech] Microwave link down In-Reply-To: <20021031200114.GA243794@morwong.ucc.gu.uwa.edu.au> References: <20021031200114.GA243794@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021031225937.GA29616@ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 04:01:14AM +0800, Grahame Bowland wrote: > Waiting for the AAP men to let me know what's wrong with it... > > So I upgraded morwong to openssh-3.5p1. Seems to all work, > the trick was to #define BROKEN_GETADDRINFO in config.h, then > it compiled fine. It was also necessary to #define _OSF_SOURCE. mustang@mermaid:~$ ssh mustang@morwong ssh_exchange_identification: Connection closed by remote host -- " I don't get mad.... I get stabby. " - William "Fat Tony" Williams. From grahame at ucc.gu.uwa.edu.au Fri Nov 1 07:27:09 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland (UCC)) Date: Wed Oct 27 01:25:59 2004 Subject: [tech] Microwave link down In-Reply-To: <20021031225937.GA29616@ucc.gu.uwa.edu.au> References: <20021031200114.GA243794@morwong.ucc.gu.uwa.edu.au> <20021031225937.GA29616@ucc.gu.uwa.edu.au> Message-ID: <20021031232709.GA28306@ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 06:59:37AM +0800, David Manchester wrote: > On Fri, Nov 01, 2002 at 04:01:14AM +0800, Grahame Bowland wrote: > > Waiting for the AAP men to let me know what's wrong with it... > > > > So I upgraded morwong to openssh-3.5p1. Seems to all work, > > the trick was to #define BROKEN_GETADDRINFO in config.h, then > > it compiled fine. It was also necessary to #define _OSF_SOURCE. > > mustang@mermaid:~$ ssh mustang@morwong > ssh_exchange_identification: Connection closed by remote host Telnet doesn't work either... the whole machine is not happy. I didn't do it (TM). ssh was definitely happy. From grahame at ucc.gu.uwa.edu.au Fri Nov 1 08:22:52 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:25:59 2004 Subject: [tech] Microwave link down In-Reply-To: <20021031232709.GA28306@ucc.gu.uwa.edu.au> References: <20021031200114.GA243794@morwong.ucc.gu.uwa.edu.au> <20021031225937.GA29616@ucc.gu.uwa.edu.au> <20021031232709.GA28306@ucc.gu.uwa.edu.au> Message-ID: <20021101002252.GA243016@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 07:27:09AM +0800, Grahame Bowland wrote: > On Fri, Nov 01, 2002 at 06:59:37AM +0800, David Manchester wrote: > > On Fri, Nov 01, 2002 at 04:01:14AM +0800, Grahame Bowland wrote: > > > Waiting for the AAP men to let me know what's wrong with it... > > > > > > So I upgraded morwong to openssh-3.5p1. Seems to all work, > > > the trick was to #define BROKEN_GETADDRINFO in config.h, then > > > it compiled fine. It was also necessary to #define _OSF_SOURCE. > > > > mustang@mermaid:~$ ssh mustang@morwong > > ssh_exchange_identification: Connection closed by remote host > > Telnet doesn't work either... the whole machine is not happy. > I didn't do it (TM). ssh was definitely happy. It's back now, anyone know what fixed it? From trs80 at ucc.gu.uwa.edu.au Fri Nov 1 13:39:12 2002 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Wed Oct 27 01:25:59 2004 Subject: [tech] morwong (inc. ssh) In-Reply-To: <20021101002252.GA243016@morwong.ucc.gu.uwa.edu.au> Message-ID: On Fri, 1 Nov 2002, Grahame Bowland wrote: > On Fri, Nov 01, 2002 at 07:27:09AM +0800, Grahame Bowland wrote: > > On Fri, Nov 01, 2002 at 06:59:37AM +0800, David Manchester wrote: > > > On Fri, Nov 01, 2002 at 04:01:14AM +0800, Grahame Bowland wrote: > > > > Waiting for the AAP men to let me know what's wrong with it... > > > > > > > > So I upgraded morwong to openssh-3.5p1. Seems to all work, > > > > the trick was to #define BROKEN_GETADDRINFO in config.h, then > > > > it compiled fine. It was also necessary to #define _OSF_SOURCE. > > > > > > mustang@mermaid:~$ ssh mustang@morwong > > > ssh_exchange_identification: Connection closed by remote host > > > > Telnet doesn't work either... the whole machine is not happy. > > I didn't do it (TM). ssh was definitely happy. > > It's back now, anyone know what fixed it? No idea, but I turned ssh X forwarding on, and updated the init.d script. prngd was running from ages ago, when I last tried to install a recent openssh. Do we want to bother with privsep on morwong? -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Treasurer 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 mustang at ucc.gu.uwa.edu.au Fri Nov 1 13:54:01 2002 From: mustang at ucc.gu.uwa.edu.au (David Manchester) Date: Wed Oct 27 01:26:00 2004 Subject: [tech] morwong (inc. ssh) In-Reply-To: References: <20021101002252.GA243016@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101055401.GB224954@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 01:39:12PM +0800, James Andrewartha wrote: > On Fri, 1 Nov 2002, Grahame Bowland wrote: > > > On Fri, Nov 01, 2002 at 07:27:09AM +0800, Grahame Bowland wrote: > > > On Fri, Nov 01, 2002 at 06:59:37AM +0800, David Manchester wrote: > > > > On Fri, Nov 01, 2002 at 04:01:14AM +0800, Grahame Bowland wrote: > > > > > Waiting for the AAP men to let me know what's wrong with it... > > > > > > > > > > So I upgraded morwong to openssh-3.5p1. Seems to all work, > > > > > the trick was to #define BROKEN_GETADDRINFO in config.h, then > > > > > it compiled fine. It was also necessary to #define _OSF_SOURCE. > > > > > > > > mustang@mermaid:~$ ssh mustang@morwong > > > > ssh_exchange_identification: Connection closed by remote host > > > > > > Telnet doesn't work either... the whole machine is not happy. > > > I didn't do it (TM). ssh was definitely happy. > > > > It's back now, anyone know what fixed it? > > No idea, but I turned ssh X forwarding on, and updated the init.d > script. prngd was running from ages ago, when I last tried to install a > recent openssh. Do we want to bother with privsep on morwong? Mermaid was resolving an IPV6 address for morwong, if that provides any more clues. D. -- " I don't get mad.... I get stabby. " - William "Fat Tony" Williams. From japester at ucc.gu.uwa.edu.au Fri Nov 1 14:14:44 2002 From: japester at ucc.gu.uwa.edu.au (Jean-Paul Blaquiere) Date: Wed Oct 27 01:26:00 2004 Subject: [tech] morwong and ssh identities Message-ID: <20021101061443.GA10507@ucc.gu.uwa.edu.au> over the last half an hour (till ~1400) ssh-ing to morwong has resulted in unknown host key prompts being shown, or connection refused. this from both other ucc machines and others. anyone got any ideas what's happening? ./jp -- Jean-Paul Blaquiere || Avatar of Computational japester at ucc.gu.uwa.edu.au || Thaumaturgy http://japester.ucc.asn.au || verum ipsum factum Questions are dangerous, for they have answers From nick at ucc.gu.uwa.edu.au Fri Nov 1 14:15:06 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:00 2004 Subject: [tech] morwong (inc. ssh) In-Reply-To: References: <20021101002252.GA243016@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101141506.F2941@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 01:39:12PM +0800, James Andrewartha wrote: > No idea, but I turned ssh X forwarding on, and updated the init.d > script. prngd was running from ages ago, when I last tried to install a > recent openssh. Do we want to bother with privsep on morwong? If it works, we may as well. Instructions are in README.privsep . A general request to anyone upgrading or reinstalling machines - please remember to backup and reuse the SSH host keys to minimise the "WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!"'s. I've mostly stopped saying "yes" blindly to those - you just can't trust the network these days. To cut down on confusion, I'm planning to run this command at 3pm ; morwong# mv /etc/ssh* /usr/local/etc/bkp20021101/_etc/ Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From grahame at ucc.gu.uwa.edu.au Fri Nov 1 14:22:47 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:00 2004 Subject: [tech] morwong (inc. ssh) In-Reply-To: <20021101141506.F2941@morwong.ucc.gu.uwa.edu.au> References: <20021101002252.GA243016@morwong.ucc.gu.uwa.edu.au> <20021101141506.F2941@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101062247.GA250404@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 02:15:06PM +0800, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 01:39:12PM +0800, James Andrewartha wrote: > > No idea, but I turned ssh X forwarding on, and updated the init.d > > script. prngd was running from ages ago, when I last tried to install a > > recent openssh. Do we want to bother with privsep on morwong? > > If it works, we may as well. Instructions are in README.privsep . > > A general request to anyone upgrading or reinstalling machines - please > remember to backup and reuse the SSH host keys to minimise the > "WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!"'s. I've mostly > stopped saying "yes" blindly to those - you just can't trust the > network these days. > > To cut down on confusion, I'm planning to run this command at 3pm ; > > morwong# mv /etc/ssh* /usr/local/etc/bkp20021101/_etc/ Don't bother with privilege seperation, it takes a bit of work to make it go on the different unixes and this hasn't been done for Tru64. I don't think it's worth the effort. (It didn't work when I tried it) From dunc-mail-1317F6D at rcpt.to Fri Nov 1 14:29:15 2002 From: dunc-mail-1317F6D at rcpt.to (Duncan Sargeant) Date: Wed Oct 27 01:26:01 2004 Subject: [tech] morwong (inc. ssh) In-Reply-To: <20021101141506.F2941@morwong.ucc.gu.uwa.edu.au> References: <20021101002252.GA243016@morwong.ucc.gu.uwa.edu.au> <20021101141506.F2941@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101062915.GA14875@optusnet.com.au> Nick Bannon wrote on Fri November 01, at 14:15 +0800: > On Fri, Nov 01, 2002 at 01:39:12PM +0800, James Andrewartha wrote: > > No idea, but I turned ssh X forwarding on, and updated the init.d > > script. prngd was running from ages ago, when I last tried to install a > > recent openssh. Do we want to bother with privsep on morwong? > > If it works, we may as well. Instructions are in README.privsep . ICBW, but about the only two things that privsep buys you is protection from any undisclosed security flaws, and plain text pass{words/phrases} appearing in an strace. Given that we can turn it on quickly when warned of the former, I think it would be a bad idea to enable it in our many-cooks environment. ,dunc From nick at ucc.gu.uwa.edu.au Fri Nov 1 14:56:34 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:01 2004 Subject: [tech] Sane mail In-Reply-To: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 12:05:18AM +0800, Grahame Bowland wrote: > I'm running this plan by you guys, I'm willing to implement it if > we want to do it. Great plan. In fact I think I'd prefer to skip the last step ; > Basically the idea is: > * remove all NFS mounts on mooneye > * set up things so that mail lands on mooneye ----skip this one---- > , mooneye then sends it on to morwong for delivery to mailboxes --------------------- mooneye can then run POP and IMAP servers directly. Most things will be able to access that, but we can wrap elm in a script to use "getmail" to suck mail to ~/Mail/mbox or similar. I'm not sure if there's a better way, but we can support procmail like that too. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From dayta at ucc.gu.uwa.edu.au Fri Nov 1 14:59:01 2002 From: dayta at ucc.gu.uwa.edu.au (Leighton Haynes) Date: Wed Oct 27 01:26:01 2004 Subject: [tech] Sane mail In-Reply-To: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101065901.GA250232@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 02:56:34PM +0800, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 12:05:18AM +0800, Grahame Bowland wrote: > > I'm running this plan by you guys, I'm willing to implement it if > > we want to do it. > > Great plan. In fact I think I'd prefer to skip the last step ; > > > Basically the idea is: > > * remove all NFS mounts on mooneye > > * set up things so that mail lands on mooneye > ----skip this one---- > > , mooneye then sends it on to morwong for delivery to mailboxes > --------------------- > > mooneye can then run POP and IMAP servers directly. Most things will be > able to access that, but we can wrap elm in a script to use "getmail" > to suck mail to ~/Mail/mbox or similar. I'm not sure if there's a > better way, but we can support procmail like that too. Yuck. Leighton... -- #0421 113 305 - dayta@ucc.gu.uwa.edu.au How do you expect?/I will know what to do./When all I know. Is what you tell me to. - Linkin Park/By Myself From bers at ucc.gu.uwa.edu.au Fri Nov 1 15:05:24 2002 From: bers at ucc.gu.uwa.edu.au (Nick Rohrlach) Date: Wed Oct 27 01:26:01 2004 Subject: [tech] Sane mail In-Reply-To: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> Message-ID: On Fri, 1 Nov 2002, Nick Bannon wrote: > mooneye can then run POP and IMAP servers directly. Most things will be > able to access that, but we can wrap elm in a script to use "getmail" > to suck mail to ~/Mail/mbox or similar. I'm not sure if there's a > better way, but we can support procmail like that too. IIRC, iiNet does something very similar... it's hideous. You get the same message redelivered each time if you leave it on the server etc. Please lets not. -- -_-_-_-_-_-_-_-_-_-_-_-_-_-_ Nick "bers" Rohrlach [NRR] bers@ucc.gu.uwa.edu.au _-_-_-_-_-_-_-_-_-_-_-_-_-_- From nick at ucc.gu.uwa.edu.au Fri Nov 1 15:07:22 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:02 2004 Subject: [tech] Sane mail In-Reply-To: <20021101065901.GA250232@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> <20021101065901.GA250232@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101150722.E182905@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 02:59:01PM +0800, Leighton Haynes wrote: > Yuck. Yuck the the standalone mail server? or the getmail bit? getmail's not necessary for anything that supports POP or IMAP itself. If we can convert our last elm users (John?), it's a non-issue. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From michael at ucc.gu.uwa.edu.au Fri Nov 1 15:12:44 2002 From: michael at ucc.gu.uwa.edu.au (Michael Deegan) Date: Wed Oct 27 01:26:02 2004 Subject: [tech] Sane mail In-Reply-To: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101071244.GD23205@zaphod.murdoch.edu.au> On Fri, Nov 01, 2002 at 02:56:34PM +0800, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 12:05:18AM +0800, Grahame Bowland wrote: > > I'm running this plan by you guys, I'm willing to implement it if > > we want to do it. > > Great plan. In fact I think I'd prefer to skip the last step ; > > > Basically the idea is: > > * remove all NFS mounts on mooneye > > * set up things so that mail lands on mooneye > ----skip this one---- > > , mooneye then sends it on to morwong for delivery to mailboxes > --------------------- > > mooneye can then run POP and IMAP servers directly. Most things will be > able to access that, but we can wrap elm in a script to use "getmail" > to suck mail to ~/Mail/mbox or similar. I'm not sure if there's a > better way, but we can support procmail like that too. Isn't that almost just like the step your planning to skip, only dodgier? :P I think I still prefer the idea of port-forwarding connections (SMTP, as well as IMAP/POP as required) from mooneye to morwong. People are welcome to think otherwise, of course... -MD -- _____________________________________________________________________________ Michael Deegan, Unix Support Officer, Murdoch University, Ph. +61 8 9360 6967 From mtearle at ucc.gu.uwa.edu.au Fri Nov 1 15:14:10 2002 From: mtearle at ucc.gu.uwa.edu.au (Mark Tearle) Date: Wed Oct 27 01:26:02 2004 Subject: [tech] Sane mail In-Reply-To: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> Message-ID: On Fri, 1 Nov 2002, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 12:05:18AM +0800, Grahame Bowland wrote: > > I'm running this plan by you guys, I'm willing to implement it if > > we want to do it. > > Great plan. In fact I think I'd prefer to skip the last step ; > > > Basically the idea is: > > * remove all NFS mounts on mooneye > > * set up things so that mail lands on mooneye > ----skip this one---- > > , mooneye then sends it on to morwong for delivery to mailboxes > --------------------- > > mooneye can then run POP and IMAP servers directly. Most things will be > able to access that, but we can wrap elm in a script to use "getmail" > to suck mail to ~/Mail/mbox or similar. I'm not sure if there's a > better way, but we can support procmail like that too. > > Nick. No, that's dumb. mooneye is exposed to the world and not the worlds most reliable machine in any of it's nine lives. The disk for the mail spool is on morwong, the users are on morwong, Chewbacca's a wookie, it just makes sense for mail to delivered on morwong. Yours Mark -- Mark Tearle - mark@tearle.com "Happiness is not a lifestyle, happiness is" - greeting card linux.conf.au 2003 - The Australian Linux Technical Conference http://conf.linux.org.au/ 22nd - 25th January 2003 in Perth From andrew at ucc.gu.uwa.edu.au Fri Nov 1 15:14:27 2002 From: andrew at ucc.gu.uwa.edu.au (Andrew Williams) Date: Wed Oct 27 01:26:02 2004 Subject: [tech] Sane mail In-Reply-To: <20021101150722.E182905@morwong.ucc.gu.uwa.edu.au> Message-ID: <000001c28176$566a3440$5c19a8c0@macadamia> > -----Original Message----- > From: tech-bounces@ucc.gu.uwa.edu.au > [mailto:tech-bounces@ucc.gu.uwa.edu.au] On Behalf Of Nick Bannon > getmail's not necessary for anything that supports POP or IMAP itself. > If we can convert our last elm users (John?), it's a non-issue. Hey, I use elm too! I don't want to use a nasty newfangled mail reader like Pine, I'm set in my ways... :-) (I actually spent my first year or so of email access using plain old 'mail') Andrew From dayta at ucc.gu.uwa.edu.au Fri Nov 1 15:19:09 2002 From: dayta at ucc.gu.uwa.edu.au (Leighton Haynes) Date: Wed Oct 27 01:26:02 2004 Subject: [tech] Sane mail In-Reply-To: <20021101150722.E182905@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> <20021101065901.GA250232@morwong.ucc.gu.uwa.edu.au> <20021101150722.E182905@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101071909.GB250232@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 03:07:22PM +0800, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 02:59:01PM +0800, Leighton Haynes wrote: > > Yuck. > > Yuck the the standalone mail server? or the getmail bit? > > getmail's not necessary for anything that supports POP or IMAP itself. > If we can convert our last elm users (John?), it's a non-issue. I use procmail to sort my mail, most of which gets read only occasionally, telling mutt to connect to mooneye's IMAP server is going to break that rather handily. I much prefer having mooneye forward the mail to morwong for local delivery. This should allow procmail to sort it nicely, and most importantly, requires no effort on my part to set up. Always the best solution. Leighton... -- #0421 113 305 - dayta@ucc.gu.uwa.edu.au How do you expect?/I will know what to do./When all I know. Is what you tell me to. - Linkin Park/By Myself From nick at ucc.gu.uwa.edu.au Fri Nov 1 15:25:13 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:03 2004 Subject: [tech] /services In-Reply-To: References: Message-ID: <20021101152513.G2941@morwong.ucc.gu.uwa.edu.au> On Thu, Oct 31, 2002 at 11:19:53PM +0800, James Andrewartha wrote: > On Thu, 31 Oct 2002, James Andrewartha wrote: [...] > > /home, and take 2gig out of /home to become /home/mail. Nick, I killed > > your mutt process because it was in /services on morwong. No problem. > Ok, /services is now on meito, and I've nuked the copy on morwong, and > added its space to /home. I was going to remove the 2gig homevol2 from > /home, but it's too close to 11pm. I also tried to balance home_domain, > but aborted after morwong decided to not respond to nfs requests. Great. I've commented out /home from meito:/etc/vfstab , but I can't umount it right now ; # umount /home nfs umount: unmount: RPC: Timed out (Crossmounts are bad at reboot time...) When it's off we can either create empty wheel home directories or enable the automounter. > Nick, when do you plan to repartition cybium? [...] I was aiming for tonight, but it depends on what time I'd be able to get in. Did you want to do some burning? Before doing it, it'd probably be polite to do an xcopy "copy if newer" run from cybium's D: to meito. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From nick at ucc.gu.uwa.edu.au Fri Nov 1 16:05:28 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:03 2004 Subject: [tech] Sane mail In-Reply-To: <20021101071909.GB250232@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> <20021101065901.GA250232@morwong.ucc.gu.uwa.edu.au> <20021101150722.E182905@morwong.ucc.gu.uwa.edu.au> <20021101071909.GB250232@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101160528.H2941@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 03:19:09PM +0800, Leighton Haynes wrote: > I use procmail to sort my mail, most of which gets read only > occasionally, telling mutt to connect to mooneye's IMAP server is > going to break that rather handily. Yes, I'm sure there must be a better way to handle procmail, getmail was just what sprang to mind. What do other networks do? (iiNet's a bit odd. Last I saw they even did the wrapper thing for pine running on opera, which seems strange.) > I much prefer having mooneye forward the mail to morwong for local delivery. > This should allow procmail to sort it nicely, and most importantly, requires > no effort on my part to set up. Always the best solution. On reflection, we can probably have the best of both worlds. mooneye can lose its NFS mounts and SMTP-forward anyone's mail to morwong who wants it. The rest can have it delivered then and there. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From nick at ucc.gu.uwa.edu.au Fri Nov 1 16:06:02 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:03 2004 Subject: [tech] Sane mail In-Reply-To: References: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101080602.GA251153@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 03:14:10PM +0800, Mark Tearle wrote: > No, that's dumb. mooneye is exposed to the world and not the worlds > most reliable machine in any of it's nine lives. The disk for the > mail spool is on morwong, the users are on morwong, Chewbacca's a > wookie, it just makes sense for mail to delivered on morwong. No, that's a kneejerk. mooneye is already exposed to the world and you're still dependent on it to get your mail through in any scenario. Currently, you're dependent on morwong as well. FWIW, many (most?) of the users are not on morwong (though I, personally, am). The point of the standalone mail server is to simplify the system and reduce the dependencies. If it works on mooneye it can also work on a locked down dedicated box with mirrored disks, firewalled to the hilt, and the server never has to touch an mbox format file again. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From japester at ucc.gu.uwa.edu.au Fri Nov 1 16:35:32 2002 From: japester at ucc.gu.uwa.edu.au (Jean-Paul Blaquiere) Date: Wed Oct 27 01:26:03 2004 Subject: [tech] network time Message-ID: <20021101083532.GC10507@ucc.gu.uwa.edu.au> are *all* the machines in ucc running ntpd? at least mussel is out of sync relative to morwong. ./jp -- Jean-Paul Blaquiere || Avatar of Computational japester at ucc.gu.uwa.edu.au || Thaumaturgy http://japester.ucc.asn.au || verum ipsum factum Questions are dangerous, for they have answers From dayta at ucc.gu.uwa.edu.au Fri Nov 1 16:37:05 2002 From: dayta at ucc.gu.uwa.edu.au (Leighton Haynes) Date: Wed Oct 27 01:26:03 2004 Subject: [tech] Sane mail In-Reply-To: <20021101080602.GA251153@morwong.ucc.gu.uwa.edu.au> References: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> <20021101080602.GA251153@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101083705.GC250232@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 04:06:02PM +0800, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 03:14:10PM +0800, Mark Tearle wrote: > > No, that's dumb. mooneye is exposed to the world and not the worlds > > most reliable machine in any of it's nine lives. The disk for the > > mail spool is on morwong, the users are on morwong, Chewbacca's a > > wookie, it just makes sense for mail to delivered on morwong. > > No, that's a kneejerk. mooneye is already exposed to the world and > you're still dependent on it to get your mail through in any scenario. > Currently, you're dependent on morwong as well. UCC will be dependant on morwong as a whole until NFS isn't there. Also, there is a (rather large) difference between being dependant on mooneye to short-term spool the mail before forwarding it to morwong, and being dependant on mooneye to actually _store_ the mail. Mooneye dies. Lots. > FWIW, many (most?) of the users are not on morwong (though I, > personally, am). But their files are. While some UCCans probably just pop/imap their mail off, a lot also just log in to machines and read it. > The point of the standalone mail server is to simplify the system and > reduce the dependencies. If it works on mooneye it can also work on a > locked down dedicated box with mirrored disks, firewalled to the hilt, > and the server never has to touch an mbox format file again. Except it doesn't really simplify the system much at all, and the dependancy, such as it is, is not large. And yes, we _could_ make a new dedicated box blah blah blah. But personally, I prefer the solution we can do right fucking now. Because doing a good solution now, is better than an excellent solution at some indeterminate point in the future. (And I'm not ceding that your solution is better in any case). Leighton... -- #0421 113 305 - dayta@ucc.gu.uwa.edu.au How do you expect?/I will know what to do./When all I know. Is what you tell me to. - Linkin Park/By Myself From andrew at ucc.gu.uwa.edu.au Fri Nov 1 16:41:41 2002 From: andrew at ucc.gu.uwa.edu.au (Andrew Williams) Date: Wed Oct 27 01:26:04 2004 Subject: [tech] network time In-Reply-To: <20021101083532.GC10507@ucc.gu.uwa.edu.au> Message-ID: <000101c28182$843b5e10$5c19a8c0@macadamia> > [mailto:tech-bounces@ucc.gu.uwa.edu.au] On Behalf Of > Jean-Paul Blaquiere > are *all* the machines in ucc running ntpd? > at least mussel is out of sync relative to morwong. 'ntpq' shows mussel, mermaid, morwong, mooneye and elysium no more than 10ms apart, and all are running ntpd. When did you motice a problem? Andrew From bernard at blackham.com.au Fri Nov 1 16:48:05 2002 From: bernard at blackham.com.au (Bernard Blackham) Date: Wed Oct 27 01:26:04 2004 Subject: [tech] Sane mail In-Reply-To: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101084805.GA13434@amidala> On Fri, Nov 01, 2002 at 12:05:18AM +0800, Grahame Bowland wrote: > * we don't lose our email when morwong vanishes Does email vanish as such or is it declined? Isn't there a saner way - I would've thought sendmail on mooneye would simply decline any email if it couldn't access the file it was expecting to deliver it to, in which case the sending MTA should try back in a few hours time, yes? Rgds, Bernard. -- Bernard Blackham bernard@blackham.com.au Australian Linux Technical Conference 2003: http://www.linux.conf.au/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20021101/9cd74919/attachment.pgp From john at ucc.gu.uwa.edu.au Fri Nov 1 17:14:24 2002 From: john at ucc.gu.uwa.edu.au (John West McKenna) Date: Wed Oct 27 01:26:05 2004 Subject: [tech] Sane mail In-Reply-To: <20021101150722.E182905@morwong.ucc.gu.uwa.edu.au> from Nick Bannon at "Nov 1, 2002 03:07:22 pm" Message-ID: <200211010914.RAA20110@mermaid.ucc.gu.uwa.edu.au> Nick Bannon writes: >getmail's not necessary for anything that supports POP or IMAP itself. >If we can convert our last elm users (John?), it's a non-issue. Give me something that looks and feels identical to elm (and requires no effort from me to set up) and I'll happily convert. elm is programmed into my fingers, like vi. Learning something else would require chopping my fingers off. John From japester at ucc.gu.uwa.edu.au Fri Nov 1 17:15:55 2002 From: japester at ucc.gu.uwa.edu.au (Jean-Paul Blaquiere) Date: Wed Oct 27 01:26:05 2004 Subject: [tech] network time In-Reply-To: <000101c28182$843b5e10$5c19a8c0@macadamia> References: <20021101083532.GC10507@ucc.gu.uwa.edu.au> <000101c28182$843b5e10$5c19a8c0@macadamia> Message-ID: <20021101091554.GA12702@ucc.gu.uwa.edu.au> > On Nov 01, Andrew Williams illuminated : > > are *all* the machines in ucc running ntpd? > > at least mussel is out of sync relative to morwong. > > 'ntpq' shows mussel, mermaid, morwong, mooneye and elysium no more than > 10ms apart, and all are running ntpd. When did you motice a problem? > hmmm intriging. Pretty much every time I log into mussel to read mail. which includes today, and random times in the past. the symptoms displayed are that when switching between mailboxes (in mutt) the one just left gets immediately noted as having received new mail and mutt wants to switch straight back to it. ./jp -- Jean-Paul Blaquiere || Avatar of Computational japester at ucc.gu.uwa.edu.au || Thaumaturgy http://japester.ucc.asn.au || verum ipsum factum Questions are dangerous, for they have answers From trs80 at ucc.gu.uwa.edu.au Fri Nov 1 17:37:59 2002 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Wed Oct 27 01:26:05 2004 Subject: [tech] /services In-Reply-To: <20021101152513.G2941@morwong.ucc.gu.uwa.edu.au> Message-ID: On Fri, 1 Nov 2002, Nick Bannon wrote: > On Thu, Oct 31, 2002 at 11:19:53PM +0800, James Andrewartha wrote: > > On Thu, 31 Oct 2002, James Andrewartha wrote: > > Ok, /services is now on meito, and I've nuked the copy on morwong, and > > added its space to /home. I was going to remove the 2gig homevol2 from > > /home, but it's too close to 11pm. I also tried to balance home_domain, > > but aborted after morwong decided to not respond to nfs requests. > > Great. I've commented out /home from meito:/etc/vfstab , but I can't > umount it right now ; Why did you do this? What relation does it have to /services? meito# cat vfstab #home:/home - /home nfs - yes hard,intr services:/services - /services nfs - yes hard,intr mounting services on meito from itself makes no sense. Did you confuse /home and /services? > When it's off we can either create empty wheel home directories or > enable the automounter. Automounting is bad, mmmkay? > > Nick, when do you plan to repartition cybium? > [...] > > I was aiming for tonight, but it depends on what time I'd be able to > get in. Did you want to do some burning? Before doing it, it'd probably > be polite to do an xcopy "copy if newer" run from cybium's D: to > meito. No, we stole a 2900 from UCS and are going to put some VLANs into UCC. -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Treasurer 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 grahame at ucc.gu.uwa.edu.au Fri Nov 1 19:49:47 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:05 2004 Subject: [tech] Sane mail In-Reply-To: <20021101084805.GA13434@amidala> References: <20021031160518.GB225208@morwong.ucc.gu.uwa.edu.au> <20021101084805.GA13434@amidala> Message-ID: <20021101114944.GB252703@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 04:48:05PM +0800, Bernard Blackham wrote: > On Fri, Nov 01, 2002 at 12:05:18AM +0800, Grahame Bowland wrote: > > * we don't lose our email when morwong vanishes > > Does email vanish as such or is it declined? Isn't there a saner way > - I would've thought sendmail on mooneye would simply decline any > email if it couldn't access the file it was expecting to deliver > it to, in which case the sending MTA should try back in a few > hours time, yes? When NFS isn't there it sits there trying to use it and the remote server either gives up, or we return an error code and the remote server re-queues the email. Although we have hit NFS related situations where we hard-bounce email. From grahame at ucc.gu.uwa.edu.au Fri Nov 1 19:47:58 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:05 2004 Subject: [tech] Sane mail In-Reply-To: <20021101080602.GA251153@morwong.ucc.gu.uwa.edu.au> References: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> <20021101080602.GA251153@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101114758.GA252703@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 04:06:02PM +0800, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 03:14:10PM +0800, Mark Tearle wrote: > > No, that's dumb. mooneye is exposed to the world and not the worlds > > most reliable machine in any of it's nine lives. The disk for the > > mail spool is on morwong, the users are on morwong, Chewbacca's a > > wookie, it just makes sense for mail to delivered on morwong. > > No, that's a kneejerk. mooneye is already exposed to the world and > you're still dependent on it to get your mail through in any scenario. > Currently, you're dependent on morwong as well. UWA's mail servers will quite happily queue email when there is a hard failure, the edge cases created by stale and/or unhappy NFS mounts make things difficult and broken. Delivering between SMTP servers is a known technology and hardly error prone. > FWIW, many (most?) of the users are not on morwong (though I, > personally, am). > > The point of the standalone mail server is to simplify the system and > reduce the dependencies. If it works on mooneye it can also work on a > locked down dedicated box with mirrored disks, firewalled to the hilt, > and the server never has to touch an mbox format file again. We're talking about improving the situation, not a mythical everyone- happy-dancing-in-the-wind senario. We don't have mirrored disks, we don't have a locked down dedicated box and I seriously question the ability of the UCC to configure a machine in such a state. Let's look at the failure points for my proposal: Mooneye is down: * now: asclepius queues our mail * then: asclepius queues our mail Morwong is down: * now: mooneye listens on port 25, then pauses randomly. Mail is delayed and possibly bounced with fatal errors. * then: mooneye queues the email happily Both mooneye and morwong are down: * now: asclepius queues our mail * then: asclepius queues our mail How is the proposed idea a problem? It also has nice load sharing properties, and we do hit the "ohmygod morwong is down" thing a fair bit. There is a definite difference between the utopian perfect solution and something that makes things significantly better and solves the problems you actually have. I'm not keen on the everyone on IMAP idea beacuse you sacrifice so much flexibility. I might be interested in doing it on tartarus because we're providing defined services there, but at UCC it's a good idea to let people mess about with things and do things the way they want. Cheers Grahame From nick at ucc.gu.uwa.edu.au Fri Nov 1 18:19:41 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:06 2004 Subject: [tech] /services In-Reply-To: References: <20021101152513.G2941@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021101101941.GB251153@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 05:37:59PM +0800, James Andrewartha wrote: [...] > mounting services on meito from itself makes no sense. Did you confuse > /home and /services? No, I didn't touch the services line, it must have been like that before. It should go too. (done) > > When it's off we can either create empty wheel home directories or > > enable the automounter. > > Automounting is bad, mmmkay? It's not _very_ bad. ::-) If we want it standalone, we can have extra (empty) home directories. (Fine by me) If we want meito mounting morwong:/home we need to do automounting rather than just doing that at the same time that morwong is mounting meito:/services . (Closer to how it is now, in case anyone objected) > > > Nick, when do you plan to repartition cybium? > > [...] > > > > I was aiming for tonight, but it depends on what time I'd be able to > > get in. Did you want to do some burning? Before doing it, it'd probably > > be polite to do an xcopy "copy if newer" run from cybium's D: to > > meito. > > No, we stole a 2900 from UCS and are going to put some VLANs into UCC. Fine by me, we'll do the repartitioning later. Probably best to copy stuff back overnight anyway - even on a theoretically 100Mbps network it takes a while to copy ~25GB. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From mtearle at tearle.com Fri Nov 1 19:38:08 2002 From: mtearle at tearle.com (Mark Tearle) Date: Wed Oct 27 01:26:06 2004 Subject: [tech] Sane mail In-Reply-To: <20021101080602.GA251153@morwong.ucc.gu.uwa.edu.au> Message-ID: On Fri, 1 Nov 2002, Nick Bannon wrote: > On Fri, Nov 01, 2002 at 03:14:10PM +0800, Mark Tearle wrote: > > No, that's dumb. mooneye is exposed to the world and not the worlds > > most reliable machine in any of it's nine lives. The disk for the > > mail spool is on morwong, the users are on morwong, Chewbacca's a > > wookie, it just makes sense for mail to delivered on morwong. > > No, that's a kneejerk. mooneye is already exposed to the world and > you're still dependent on it to get your mail through in any scenario. > Currently, you're dependent on morwong as well. Depends on how we order our MX's/firewall smtp to morwong. We could set up our MX's as following: 10 morwong 20 mooneye 30 asclepius (and have a firewall rule stopping asclepius by default talking to morwong (which we can quickly drop if mooneye goes boom)) > FWIW, many (most?) of the users are not on morwong (though I, > personally, am). But morwong is where the mail is to be stored, and ergo, where the MDA (mail delivery agent) should run. morwong is also where the home directories where you save your mail thus there is some sense in having the spools there too. > The point of the standalone mail server is to simplify the system and > reduce the dependencies. If it works on mooneye it can also work on a > locked down dedicated box with mirrored disks, firewalled to the hilt, > and the server never has to touch an mbox format file again. Good solution, wrong problem. It's just not a good solution to the the usage scenarios we have in UCC. Yours Mark -- Mark Tearle - mark@tearle.com "Happiness is not a lifestyle, happiness is" - greeting card linux.conf.au 2003 - The Australian Linux Technical Conference http://conf.linux.org.au/ 22nd - 25th January 2003 in Perth From david at luyer.net Sun Nov 3 23:37:59 2002 From: david at luyer.net (David Luyer) Date: Wed Oct 27 01:26:06 2004 Subject: [tech] Dispense Message-ID: <20021103153759.GA4956@luyer.net> Can whoever has the latest source tree for dispense give me an email :) I mean, I know the license doesn't let you have one, but I also know the Mb rate has changed in the last three years... I want to fix a couple of bugs (one root hole, and one free drinks problem, if the second is fixable) this week if I can, since I'm in Perth. And if I'm feeling nice I might even change the license to GPL ;-) David. -- David Luyer Phone: +61 3 9674 7525 Network Development Manager P A C I F I C Fax: +61 3 9699 8693 Pacific Internet (Australia) I N T E R N E T Mobile: +61 4 1111 BYTE http://www.pacific.net.au/ NASDAQ: PCNTF From grahame at ucc.gu.uwa.edu.au Mon Nov 4 01:47:57 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:06 2004 Subject: [tech] Cisco 2900 Message-ID: <20021103174756.GA266278@morwong.ucc.gu.uwa.edu.au> Hi guys We now have a Cisco 2900 in the middle of our network; 130.95.13.2. I've got the latest code in /tftpboot on hydra, I'll look at updating it tommorow. We'll plug morwong and the other important things into it; should be cute as we can monitor ports and do MRTG graphs. It has several VLANs configured. '2' is Clubroom, '3' is Machine Room and '4' is Guild. Eventually we will plug the guild uplink into it, which will allow us to reconfigure the network around a failed hydra remotely. It only talks ISL at the moment, I doubt even the latest code will do dot1q. So for now hydra still has all the ethernet interfaces. Perhaps UCS will have a better spare switch in the future - the accelar could also be lent, but it looked like too much pain. Also, whoever owns the 7500 in the machine room please do something with it or I'm going to make it into UCC router. It's but a PCMCIA flash card away... and I can't see any reason a Linux box would do a better job for UCC's needs. And the 7500 does v6, ISL and everything we could possibly want. Have fun Grahame From fryers at ucc.gu.uwa.edu.au Mon Nov 4 05:48:15 2002 From: fryers at ucc.gu.uwa.edu.au (Simon Fryer) Date: Wed Oct 27 01:26:06 2004 Subject: [tech] Connection Problems Message-ID: <20021103214815.GB269410@morwong.ucc.gu.uwa.edu.au> Evening all I have noticed that over the past couple of weeks I have had problems connecting to UCC. Traceroutes seem to go to parnet-uwa.parnet.edu.au and then die at villa.net.uwa.edu.au. When I can get a connection it progressed through villa.net.uwa.edu.au to ucsrouter.ucc.gu.uwa.edu.au an so on. Is there any reason why ucsrouter is intermittently visable to villa.net.uwa.edu.au? 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 zarquin at ucc.gu.uwa.edu.au Mon Nov 4 08:13:06 2002 From: zarquin at ucc.gu.uwa.edu.au (Alwyn Nixon-Lloyd) Date: Wed Oct 27 01:26:06 2004 Subject: [tech] Connection Problems In-Reply-To: <20021103214815.GB269410@morwong.ucc.gu.uwa.edu.au> Message-ID: Hey Simon, Ita actually probably due to the fact that UCC probably has had more downtime this week than in the entire year . Due to people thinking that things are cool. and avoiding other people :). but i won't go into that and let them answer for themselves. anyway. Thats what the UCC is about :) (and yes , that is also the avoiding people one too ). Cheers, Alwyn On Mon, 4 Nov 2002, Simon Fryer wrote: > Evening all > > I have noticed that over the past couple of weeks I have had problems > connecting to UCC. Traceroutes seem to go to parnet-uwa.parnet.edu.au > and then die at villa.net.uwa.edu.au. When I can get a connection it > progressed through villa.net.uwa.edu.au to ucsrouter.ucc.gu.uwa.edu.au an > so on. Is there any reason why ucsrouter is intermittently visable to > villa.net.uwa.edu.au? > > 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 grahame at ucc.gu.uwa.edu.au Mon Nov 4 11:06:40 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:07 2004 Subject: [tech] Connection Problems In-Reply-To: References: <20021103214815.GB269410@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021104030640.GA270674@morwong.ucc.gu.uwa.edu.au> On Mon, Nov 04, 2002 at 08:13:06AM +0800, Alwyn Nixon-Lloyd wrote: > Hey Simon, > Ita actually probably due to the fact that UCC probably has had more > downtime this week than in the entire year . Due to people thinking that > things are cool. and avoiding other people :). but i won't go into that > and let them answer for themselves. anyway. Thats what the UCC is about :) > > (and yes , that is also the avoiding people one too ). Loading experimental protocol stacks onto the uccrouter is just normal fun and games, after all. I'm starting to suspect there is some other hardware issue though; I'll have to prod it a bit tonight after the LCA meeting. From adrian at ucc.gu.uwa.edu.au Mon Nov 4 11:07:36 2002 From: adrian at ucc.gu.uwa.edu.au (Adrian Chadd) Date: Wed Oct 27 01:26:07 2004 Subject: [tech] Cisco 2900 In-Reply-To: <20021103174756.GA266278@morwong.ucc.gu.uwa.edu.au> References: <20021103174756.GA266278@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021104030736.GA270637@morwong.ucc.gu.uwa.edu.au> > > Also, whoever owns the 7500 in the machine room please do something with > it or I'm going to make it into UCC router. It's but a PCMCIA flash card > away... and I can't see any reason a Linux box would do a better job for > UCC's needs. And the 7500 does v6, ISL and everything we could possibly > want. Thats mine, and unless UCC has $5k to fix my CC up with, they can't have it at the moment. :) Its got a FEIP and two FEIP-ethernet cards that, IIRC, aren't exactly PA and don't exactly do vlan (but they do ISL.) As mentioned before, people can play with my 2948G/7500, but please be careful and don't use it as a fish and chips table like I've done in the past. :) Adrian From nick at ucc.gu.uwa.edu.au Mon Nov 4 15:36:28 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:07 2004 Subject: elm (was Re: [tech] Sane mail) In-Reply-To: <200211010914.RAA20110@mermaid.ucc.gu.uwa.edu.au> References: <20021101150722.E182905@morwong.ucc.gu.uwa.edu.au> <200211010914.RAA20110@mermaid.ucc.gu.uwa.edu.au> Message-ID: <20021104073628.GA264747@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 05:14:24PM +0800, John West McKenna wrote: > Give me something that looks and feels identical to elm (and requires no > effort from me to set up) and I'll happily convert. elm is programmed > into my fingers, like vi. Learning something else would require > chopping my fingers off. elm isn't dead, it's just resting. I don't really feel like MUA advocacy right now, but I was curious about the status of elm and here's the results. elm 2.4 was released in 1992, and that was that for a long time. Various people made patches for it, but they never seemed to go anywhere, while the fabled elm 2.5 just took longer and longer to appear. In particular, a guy called Michael Elkins put a lot of effort into patching elm, fixing bugs and adding features, producing ELM-ME+. Eventually he bit the bullet and wrote a new mailreader from scratch - mutt. elm 2.5 was eventually released in 1999, but no-one uses it. elm-2.4-me+ has had development by various people. On closer inspection, it seems that it even supports POP and IMAP now, but the version on mermaid and morwong (from 1999) does not. It did not make it into the Debian woody release because no-one cared enough about it to fix its bugs - a couple of people piped up in July to adopt it, but that hasn't actually happened yet (but the bugs were fixed by someone else). There are perfectly good reasons why someone might want to use a different MUA like pine/sylpheed/emacs/evolution/... , but mutt is a better elm than elm. It feels like elm, except the help menu is only one line by default (use ? for a full list of commands) and it doesn't have a built-in aliases editor (convert with elmalias, reread a file of configuration commands with "source ~/.mail_aliases" or similar) . Homepage, FAQ: http://www.mutt.org/ Comparison: http://www.math.fu-berlin.de/~guckes/mutt/elm.vs.mutt.html Mutt still sucks, of course. "All mail clients suck. This one just sucks less." - Michael Elkins, circa 1995 Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From trs80 at ucc.gu.uwa.edu.au Mon Nov 4 16:42:50 2002 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Wed Oct 27 01:26:07 2004 Subject: [tech] Connection Problems In-Reply-To: <20021103214815.GB269410@morwong.ucc.gu.uwa.edu.au> Message-ID: On Mon, 4 Nov 2002, Simon Fryer wrote: > I have noticed that over the past couple of weeks I have had problems > connecting to UCC. Traceroutes seem to go to parnet-uwa.parnet.edu.au > and then die at villa.net.uwa.edu.au. When I can get a connection it > progressed through villa.net.uwa.edu.au to ucsrouter.ucc.gu.uwa.edu.au an > so on. Is there any reason why ucsrouter is intermittently visable to > villa.net.uwa.edu.au? As others have indicated, the network hasn't been too stable. Between dodgy dumterms taking out the guild machine room to hydra halting on "Keyboard not detected, press F1 to continue" when rebooting remotely (now fixed in the BIOS) and myself occasionally taking down interfaces on morwong at 1am, I'm surprised I haven't been flamed more. On the plus side, there is now a DECserver providing serial console access to a few machines, which should help prevent some things recurring in the future. A second link into UCC wouldn't hurt either -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Treasurer 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 mustang at ucc.gu.uwa.edu.au Mon Nov 4 17:19:38 2002 From: mustang at ucc.gu.uwa.edu.au (David Manchester) Date: Wed Oct 27 01:26:08 2004 Subject: [tech] Connection Problems In-Reply-To: References: <20021103214815.GB269410@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021104091938.GC251220@morwong.ucc.gu.uwa.edu.au> On Mon, Nov 04, 2002 at 04:42:50PM +0800, James Andrewartha wrote: > On Mon, 4 Nov 2002, Simon Fryer wrote: > > > I have noticed that over the past couple of weeks I have had problems > > connecting to UCC. Traceroutes seem to go to parnet-uwa.parnet.edu.au > > and then die at villa.net.uwa.edu.au. When I can get a connection it > > progressed through villa.net.uwa.edu.au to ucsrouter.ucc.gu.uwa.edu.au an > > so on. Is there any reason why ucsrouter is intermittently visable to > > villa.net.uwa.edu.au? > > As others have indicated, the network hasn't been too stable. Between > dodgy dumterms taking out the guild machine room to hydra halting on > "Keyboard not detected, press F1 to continue" when rebooting remotely (now > fixed in the BIOS) and myself occasionally taking down interfaces on > morwong at 1am, I'm surprised I haven't been flamed more. On the plus > side, there is now a DECserver providing serial console access to a few > machines, which should help prevent some things recurring in the future. A > second link into UCC wouldn't hurt either Some details about the DECserver couldn't hurt .... D. -- " I don't get mad.... I get stabby. " - William "Fat Tony" Williams. From nick at ucc.gu.uwa.edu.au Mon Nov 4 21:22:38 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:08 2004 Subject: [tech] Sane mail In-Reply-To: <20021101114758.GA252703@morwong.ucc.gu.uwa.edu.au> References: <20021101145634.D182905@morwong.ucc.gu.uwa.edu.au> <20021101080602.GA251153@morwong.ucc.gu.uwa.edu.au> <20021101114758.GA252703@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021104132237.GC264747@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 01, 2002 at 07:47:58PM +0800, Grahame Bowland wrote: [...] Your idea is great, as I said. My suggestion is just a tweak we can do on mooneye; it makes no real difference either way and doesn't require any different infrastructure. (Some took it badly for some reason, but I have no particular desire to defend it now) > We don't have mirrored disks, we > don't have a locked down dedicated box and I seriously question the > ability of the UCC to configure a machine in such a state. [...] A box like that was set up recently - meito. Actually, we're going to have to consider doing something like that for any new server because you can't trust a hard disc as far as you can throw it these days. > I'm not keen on the everyone on IMAP idea beacuse you sacrifice so > much flexibility. I might be interested in doing it on tartarus because > we're providing defined services there, but at UCC it's a good idea > to let people mess about with things and do things the way they want. There's better solutions than everyone-on-IMAP, yeah. However, IMAP will get a fair bit more useful when it's running on the same host as the mail spool. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From proxy at zdlcomputing.com Tue Nov 5 11:41:30 2002 From: proxy at zdlcomputing.com (Davyd 'proXy' Madeley) Date: Wed Oct 27 01:26:08 2004 Subject: [tech] Flying Message-ID: <20021105114130.B22744@sg2.zdlcomputing.com> Is it just me or is flying not routing to WAIX properly? It appears to be going to hydra-2 instead of hydra-2.waix traceroute to mail.tower.net.au (203.24.100.68) (from mussel) 1 hydra.waix.ucc.gu.uwa.edu.au (172.26.42.3) 2 gugate.gu.uwa.edu.au (130.95.100.1) 3 muchacho.net.uwa.edu.au (130.95.2.128) 4 atm3-0-0.13.uwa-parnet.parnet.edu.au (203.19.110.17) 5 198.32.212.20 (198.32.212.20) 6 virtual.tower.net.au (203.24.100.68) traceroute to mail.tower.net.au (from a wireless) 1 flying-wireless.waix.ucc.gu.uwa.edu.au (172.26.42.241) 2 hydra-2.ucc.gu.uwa.edu.au (130.95.13.65) 3 hydra-2.ucc.gu.uwa.edu.au reports: Destination net unreachable Surely it should go something like 1 flying-wireless.waix.ucc.gu 2 hydra-2.waix.ucc.gu 3 gugate.gu 4 muchacho.net etc etc. Someone care to fix it? --X -- http://davyd.ucc.asn.au/ linux.conf.au Perth 2003 PGP Fingerprint 08B0 341A 0B9B 08BB 2118 C060 2EDD BB4F 5191 6CDA -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available Url : http://lists.ucc.gu.uwa.edu.au/pipermail/tech/attachments/20021105/af992c30/attachment.pgp From grahame at ucc.gu.uwa.edu.au Wed Nov 6 02:13:34 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:09 2004 Subject: [tech] Piggery Message-ID: <20021105181334.GA284265@morwong.ucc.gu.uwa.edu.au> Now has openssh-3.5p1 instead of the evil commercial ssh of hell. From trs80 at ucc.gu.uwa.edu.au Sat Nov 9 21:27:30 2002 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Wed Oct 27 01:26:09 2004 Subject: [tech] mail from mussel Message-ID: Mail sent from mussel has not been getting though for several days (since Nov 6 to be eacxt). sendmail can't seem to find asclepius: Nov 9 20:56:57 mussel sm-mta[23111]: gA9Cuvej023111: gA9Cuvek023111: return to sender: Host unknown (Name server: asclepius.uwa.edu.au.: host not found) Actually, this might be related to the libc upgrade recently (NSS was changed) ... hmm. Restarting sendmail doesn't seem to have an effect, so it seems. I just installed a new sendmail, and it seems to have fixed the problem. All the mail that didn't get sent is in /var/lib/sendmail/dead.letter - do we want to process this mail somehow? -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Treasurer 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 grahame at ucs.uwa.edu.au Thu Nov 14 17:08:53 2002 From: grahame at ucs.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:09 2004 Subject: [tech] Probable equipment free for UCC Message-ID: <200211141708.53219.grahame@ucs.uwa.edu.au> Hi guys We have a Nortel Accelar 1100-B with 16 100Mb ethernet ports. It's been taken off the asset register today as it is faulty, redundant and obsolete. We're not able to sell it as it loses packets on some of the interfaces and we were unable to reach a resolution for this. However, I suspect it's still fine to put into the UCC network. Most of the ports work and it is a reasonably fast switch-router. It'd definitely make a neat replacement for hydra: the IP routing could be done on the accelar and the Appletalk done on hydra via a VLAN trunk. Even if just run as a switch, it supports 802.1q so VLAN stuff to hydra would work. Of course it's fully managed, etc so you can get pretty MRTG graphs and the works. Is UCC interested? I have to figure out how to clobber the configuration, then I can bring it over tommorow night. Cheers Grahame -- Grahame Bowland Email: grahame@ucs.uwa.edu.au University Communications Services Phone: +61 8 9380 1175 The University of Western Australia Fax: +61 8 9380 1109 CRICOS: 00126G From nick at ucc.gu.uwa.edu.au Thu Nov 14 20:37:45 2002 From: nick at ucc.gu.uwa.edu.au (Nick Bannon) Date: Wed Oct 27 01:26:09 2004 Subject: [tech] Probable equipment free for UCC In-Reply-To: <200211141708.53219.grahame@ucs.uwa.edu.au> References: <200211141708.53219.grahame@ucs.uwa.edu.au> Message-ID: <20021114123744.GL296069@morwong.ucc.gu.uwa.edu.au> On Thu, Nov 14, 2002 at 05:08:53PM +0800, Grahame Bowland wrote: [Is UCC interested in a funky, although slightly dodgy switch-router?] Need you ask? Yeah, sounds great. Well, assuming _some_ of the ports work. ::-) > However, I suspect it's still fine to put into the UCC network. Most of the > ports work and it is a reasonably fast switch-router. It'd definitely make a > neat replacement for hydra: the IP routing could be done on the accelar and > the Appletalk done on hydra via a VLAN trunk. [...] It'd probably be a good idea to connect up hydra in parallel, at least as an experiment - we could do with a backup router. Nick. -- Nick Bannon | "I made this letter longer than usual because nick-sig@rcpt.to | I lack the time to make it shorter." - Pascal http://linux.conf.au/ | Australian Linux Technical Conference, 2003 From grahame at ucc.gu.uwa.edu.au Thu Nov 14 21:31:00 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland) Date: Wed Oct 27 01:26:09 2004 Subject: [tech] Resolved: Mooneye down 14/11/2002 Message-ID: <20021114133100.GA343752@morwong.ucc.gu.uwa.edu.au> Your ticket number for this job is: 1-494-LZ Problem summary: mooneye down. Console shows kernel panic due to null pointer dereference and death of interrupt handler. Reboot resulted in machine turning on all keyboard LEDs but no VGA out. Inspection of internals of machine showed CPU fan not running. Burnt a fresher on the CPU to check theory. Replaced CPU fan and flat MTGT unit. Problem resolved. Thanks to Thomas and the other fresher whose name I have forgotten. Regards Grahame Bowland Corporate Services The University Computer Club From mtearle at tearle.com Fri Nov 15 12:25:34 2002 From: mtearle at tearle.com (Mark Tearle) Date: Wed Oct 27 01:26:10 2004 Subject: [tech] UCC Machine Room Cleanup Message-ID: Hi all Expect downtime tonight, we're cleaning out the pigsty that is the machine room. Yours Mark -- Mark Tearle - mark@tearle.com "Happiness is not a lifestyle, happiness is" - greeting card linux.conf.au 2003 - The Australian Linux Technical Conference http://conf.linux.org.au/ 22nd - 25th January 2003 in Perth From acolyte at ucc.gu.uwa.edu.au Fri Nov 15 13:29:05 2002 From: acolyte at ucc.gu.uwa.edu.au (Andrew Bailey) Date: Wed Oct 27 01:26:10 2004 Subject: [tech] UCC Machine Room Cleanup In-Reply-To: References: Message-ID: <20021115052905.GA340476@morwong.ucc.gu.uwa.edu.au> On Fri, Nov 15, 2002 at 12:25:34PM +0800, Mark Tearle wrote: > Hi all > > Expect downtime tonight, we're cleaning out the pigsty that is the > machine room. > Remember to remove the elephants as well as the pigs :) Andrew. > Yours > Mark > -- > Mark Tearle - mark@tearle.com "Happiness is not a lifestyle, happiness is" > - greeting card > linux.conf.au 2003 - The Australian Linux Technical Conference > http://conf.linux.org.au/ 22nd - 25th January 2003 in Perth > -- "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 trs80 at ucc.gu.uwa.edu.au Sun Nov 17 01:45:32 2002 From: trs80 at ucc.gu.uwa.edu.au (James Andrewartha) Date: Wed Oct 27 01:26:10 2004 Subject: [tech] Re: [ucc] Mermaid is down (again) In-Reply-To: <20021116171017.GA6320@morwong.ucc.gu.uwa.edu.au> Message-ID: On Sun, 17 Nov 2002, David Manchester wrote: > On Sat, Nov 16, 2002 at 01:28:39PM +0800, Grahame Bowland wrote: > > Hey guys > > > > Mermaid is down, again. If anyone opens Cameron Hall could > > they ring 1175 and I'll pop over with some spare bits and > > try to fix it. > > Mermaid's been quietly happy for a long time. > Who's been trying to "fix" it lately? Nobody. It just likes to die after running for half an hour or so, whatever it's running, Linux or memtest x86. This may be due to heat damage from Thursday. Mesh, Grahame and Zarquin played with it today, but didn't have any success. Tonight I tried to swap mermaid's hdd, NIC, serial card and CPU into one of the venturis's that was lying around (which had 192meg of RAM), but couldn't get LILO to boot (it was stuck at LI), even after some tqeaking of lilo (interspersed with a few hours trying to make a working boot floppy). -- # TRS-80 trs80(a)ucc.gu.uwa.edu.au #/ "Otherwise Bub here will do \ # UCC Treasurer 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 tieryn at coman.com.au Sun Nov 17 16:02:05 2002 From: tieryn at coman.com.au (Chris Coman) Date: Wed Oct 27 01:26:10 2004 Subject: [tech] Proxy server down? (is this related to mermaid?) Message-ID: <008f01c28e0f$a1e22040$3200a8c0@krondor> http://tieryn.ucc.asn.au/map.gif Proxy Error The proxy server received an invalid response from an upstream server. The proxy server could not handle the request GET /map.gif . Reason: Could not connect to remote machine: No route to host Apache/1.3.26 Server at tieryn.ucc.asn.au Port 80 Any ideas? Coman. From adrian at ucc.gu.uwa.edu.au Sun Nov 17 16:32:11 2002 From: adrian at ucc.gu.uwa.edu.au (Adrian Chadd) Date: Wed Oct 27 01:26:10 2004 Subject: [tech] Proxy server down? (is this related to mermaid?) In-Reply-To: <008f01c28e0f$a1e22040$3200a8c0@krondor> References: <008f01c28e0f$a1e22040$3200a8c0@krondor> Message-ID: <20021117083211.GF10513@morwong.ucc.gu.uwa.edu.au> On Sun, Nov 17, 2002, Chris Coman wrote: > http://tieryn.ucc.asn.au/map.gif > > Proxy Error > The proxy server received an invalid response from an upstream server. > The proxy server could not handle the request GET /map.gif . > Reason: Could not connect to remote machine: No route to host Err, which machine? Are you trying to use the mojarra:3129 proxy? Adrian From matt at ucc.gu.uwa.edu.au Sun Nov 17 18:31:04 2002 From: matt at ucc.gu.uwa.edu.au (Matt Johnston) Date: Wed Oct 27 01:26:11 2004 Subject: [tech] Proxy server down? (is this related to mermaid?) In-Reply-To: <008f01c28e0f$a1e22040$3200a8c0@krondor> References: <008f01c28e0f$a1e22040$3200a8c0@krondor> Message-ID: <20021117103104.GE2738@morwong.ucc.gu.uwa.edu.au> This is because mermaid usually serves UCC webpages, mooneye just proxying them on to mermaid. So until mermaid is fixed, it'll be broken. If you really need something, poking around in /services might work. Matt On Sun, Nov 17, 2002 at 04:02:05PM +0800, Chris Coman wrote: > http://tieryn.ucc.asn.au/map.gif > > Proxy Error > The proxy server received an invalid response from an upstream server. > The proxy server could not handle the request GET /map.gif . > Reason: Could not connect to remote machine: No route to host > > Apache/1.3.26 Server at tieryn.ucc.asn.au Port 80 > > > Any ideas? > Coman. > > From grahame at ucc.gu.uwa.edu.au Mon Nov 18 22:05:31 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland (UCC)) Date: Wed Oct 27 01:26:11 2004 Subject: [tech] 7500 works... more Message-ID: <20021118140530.GA10349@ucc.gu.uwa.edu.au> Hi all I've upgraded the bootflash on the 7500 to rsp-boot-mz.120-13a.bin. This is good. This is sane. The bootloader can now talk to the ATA flash card disk0. This is also good. It also sees the interfaces ok. This is good. The problem: the damn thing doesn't load its config on reload. It then drops into 12.0-13a boot and the initial config prompt. Typing 'no' and then 'copy nvram:startup-config running-config' gets it running again. Go figure. As a result it doesn't boot the nice shiny new IOS on disk0:. Anyone run into this? Anyone know the solution? I can't see why it wouldn't use the config on nvram? It's bizarre. I have the right boot bootldr line in there and everything. I even tried copying startup-config to bootflash: and putting boot config bootflash:startup-config into it. The problem is definitely not the flash card, dir disk0: works fine from the bootloader, I did a format on it and copied the image over afresh. Cheers Grahame From adrian at ucc.gu.uwa.edu.au Mon Nov 18 22:08:57 2002 From: adrian at ucc.gu.uwa.edu.au (Adrian Chadd) Date: Wed Oct 27 01:26:11 2004 Subject: [tech] 7500 works... more In-Reply-To: <20021118140530.GA10349@ucc.gu.uwa.edu.au> References: <20021118140530.GA10349@ucc.gu.uwa.edu.au> Message-ID: <20021118140857.GM10513@morwong.ucc.gu.uwa.edu.au> On Mon, Nov 18, 2002, Grahame Bowland wrote: > Hi all > > I've upgraded the bootflash on the 7500 to rsp-boot-mz.120-13a.bin. This > is good. This is sane. The bootloader can now talk to the ATA flash card > disk0. This is also good. It also sees the interfaces ok. This is good. Cute! > The problem: the damn thing doesn't load its config on reload. It then drops > into 12.0-13a boot and the initial config prompt. Typing 'no' and then > 'copy nvram:startup-config running-config' gets it running again. Go > figure. Right. Have you checked out the config register? I bet I set it to ignore loading the config from nvram at startup when I played with it last - probably for password recovery. Search for "Cisco 7500 Password Recovery" on CCO or something to see what I mean. Adrian From grahame at ucc.gu.uwa.edu.au Mon Nov 18 22:21:38 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland (UCC)) Date: Wed Oct 27 01:26:11 2004 Subject: [tech] 7500 works... more In-Reply-To: <20021118140857.GM10513@morwong.ucc.gu.uwa.edu.au> References: <20021118140530.GA10349@ucc.gu.uwa.edu.au> <20021118140857.GM10513@morwong.ucc.gu.uwa.edu.au> Message-ID: <20021118142138.GB10349@ucc.gu.uwa.edu.au> On Mon, Nov 18, 2002 at 10:08:57PM +0800, Adrian Chadd wrote: > On Mon, Nov 18, 2002, Grahame Bowland wrote: > > Hi all > > > > I've upgraded the bootflash on the 7500 to rsp-boot-mz.120-13a.bin. This > > is good. This is sane. The bootloader can now talk to the ATA flash card > > disk0. This is also good. It also sees the interfaces ok. This is good. > > Cute! > > > The problem: the damn thing doesn't load its config on reload. It then drops > > into 12.0-13a boot and the initial config prompt. Typing 'no' and then > > 'copy nvram:startup-config running-config' gets it running again. Go > > figure. > > Right. Have you checked out the config register? I bet I set it to > ignore loading the config from nvram at startup when I played with it > last - probably for password recovery. > > Search for "Cisco 7500 Password Recovery" on CCO or something to see > what I mean. That was it - should have thought of that! It's happily running 12.1.9a now :-) We officially have an incredibly useful router. I'll set up authentication nicely now and let people know how to access it in person. Cheers Grahame From grahame at ucc.gu.uwa.edu.au Mon Nov 18 22:47:55 2002 From: grahame at ucc.gu.uwa.edu.au (Grahame Bowland (UCC)) Date: Wed Oct 27 01:26:12 2004 Subject: [tech] 7500 works... more In-Reply-To: <20021118142138.GB10349@ucc.gu.uwa.edu.au> References: <20021118140530.GA10349@ucc.gu.uwa.edu.au> <20021118140857.GM10513@morwong.ucc.gu.uwa.edu.au> <20021118142138.GB10349@ucc.gu.uwa.edu.au> Message-ID: <20021118144755.GA11600@ucc.gu.uwa.edu.au> > That was it - should have thought of that! > > It's happily running 12.1.9a now :-) > > We officially have an incredibly useful router. I'll set up authentication > nicely now and let people know how to access it in person. Ok guys, it works happily. It's at extra-virgin.ucc.gu.uwa.edu.au: I gave some people {mtearle,trs80,matt,grahame} access, the rest of you add yourselves to /etc/raddb/users on hydra. It's doing radius, the enable password is the same as bertoli and it's all good. That flash card can live there for a bit, as I want to try some things on there rather than on the campus 7500. That said, there is no reason this can't be stable. Adrian - how long can we reasonably expect to have this? Anyway, it's trivial to make this thing route UCC -- so very very trivial. And we can always just turn hydra back on if adrian wants the 7500 back. Cheers Grahame (And no Lathiat, it doesn't do IPv6. I haven't put that recent an IOS on there in fear it won't talk to the scary old FastEthernet interfaces.) From adrian at ucc.gu.uwa.edu.au Tue Nov 19 07:31:17 2002 From: adrian at ucc.gu.uwa.edu.au (Adrian Chadd) Date: Wed Oct 27 01:26:12 2004 Subject: [tech] 7500 works... more In-Reply-To: <20021118144755.GA11600@ucc.gu.uwa.edu.au> References: <20021118140530.GA10349@ucc.gu.uwa.edu.au> <20021118140857.GM10513@morwong.ucc.gu.uwa.edu.au> <20021118142138.GB10349@ucc.gu.uwa.edu.au> <20021118144755.GA11600@ucc.gu.uwa.edu.au> Message-ID: <20021118233117.GN10513@morwong.ucc.gu.uwa.edu.au> On Mon, Nov 18, 2002, Grahame Bowland wrote: > > We officially have an incredibly useful router. I'll set up authentication > > nicely now and let people know how to access it in person. Yay! > Ok guys, it works happily. It's at extra-virgin.ucc.gu.uwa.edu.au: I gave > some people {mtearle,trs80,matt,grahame} access, the rest of you add > yourselves to /etc/raddb/users on hydra. > > It's doing radius, the enable password is the same as bertoli and it's all > good. That flash card can live there for a bit, as I want to try some things > on there rather than on the campus 7500. :) > That said, there is no reason this can't be stable. Adrian - how long can > we reasonably expect to have this? Well, it depends if someone wants to buy it off me. > Anyway, it's trivial to make this thing route UCC -- so very very trivial. > And we can always just turn hydra back on if adrian wants the 7500 back. :) Don't treat it as a permanent fixture just yet. Once I've cleared up my credit card debt (which includes this router) I'll be in a better position to say what will happen with it. That said, feel free to muck about with it now. Again, pair it with the 2948G for some ISL loving and play with real network segmentation. :) Adrian From trent at ucc.gu.uwa.edu.au Tue Nov 26 09:15:30 2002 From: trent at ucc.gu.uwa.edu.au (Trent Lloyd) Date: Wed Oct 27 01:26:12 2004 Subject: [tech] IPv6 Half-Brokenness Message-ID: <20021126011530.GA26294@ucc.gu.uwa.edu.au> Heyas, IPv6 seems to have routes to some places, but now all Places without routes includes all of my irc servers :P Examples Places: nextgen.us.irc-desk.net kelowna.hub.irc-desk.net balhalla.uk.irc-desk.net -- Trent From trent at ucc.gu.uwa.edu.au Tue Nov 26 09:16:16 2002 From: trent at ucc.gu.uwa.edu.au (Trent Lloyd) Date: Wed Oct 27 01:26:12 2004 Subject: [tech] IPv6 Fixed? Message-ID: <20021126011616.GA71930@morwong.ucc.gu.uwa.edu.au> Seems to be working now.. few hours later.. shrug.. -- Trent From bernard at blackham.com.au Fri Nov 29 16:54:46 2002 From: bernard at blackham.com.au (Bernard Blackham) Date: Wed Oct 27 01:26:12 2004 Subject: [tech] CRO? Message-ID: <20021129085446.GA11025@amidala> 'lo, Who's currently got the UCC CRO? And can I steal it for a while? Another random but related thing - what might be the best/easiest way to convert the analog 0.7V pp from a VGA signal to 0/+5V TTL logic signal? I'm open to any ideas. :) Tya, Bernard. -- Bernard Blackham bernard at blackham dot com dot au Australian Linux Technical Conference 2003: http://www.linux.conf.au/