[00:00] <racecar56> im kind of not working on the server at the moment
[00:02] <leonel> is it possible to add a  signature to all the outgoing emails for 1 domain on a server with many virtual domains ??  this with  postfix ..
[00:07] <dsmith-work> Speaking of named..
[00:08] <fx3> dsmith-work, ?
[00:09] <dsmith-work> I just moved dns from a Debian Etch system to a new ubuntu.  And rndc freeze/thaw are acting funny.
[00:10] <dsmith-work> Probably because named is running as bind instead of root,I'm guessing.
[00:11] <fx3> that didnt make any sense to me at all :|
[00:11] <fx3> just shows how much i know about named/bind
[00:11] <dsmith-work> Hrmph.  I don;t even know what version of ubuntu this is.
[00:12] <ScottK> leonel: What kind of signature?
[00:12] <dsmith-work> Ahh. 8.10
[00:12] <Deeps> dsmith-work: bind doesn't run as root under debian, that would be silly
[00:13] <leonel> scottK the usual disclaimer note     and  this  only for one domain   and that server has many other domains with  no need for signature / disclaimer note
[00:13] <dsmith-work> Ok, that's probably an assumption I was making.
[00:13] <ScottK> leonel: OK.  I thought maybe you meant something like a DKIM signature.
[00:14] <ScottK> leonel: You can do it with restriction classes.
[00:14] <leonel> I'll check that  thanks scottK
[00:17] <dsmith-work> Ahh. I see what's going on now.  The Debian zone files are owned by bind, on Ubuntu they are owned by root.
[00:20] <dsmith-work> That would explain things.
[00:25] <fx3> ahh well, time for bed or ill start making silly mistakes
[00:26] <fx3> launching mistaked :P
[00:27] <dsmith-work> Hmm.  That's not it. Still get a
[00:27] <dsmith-work> rndc: 'freeze' failed: permission denied
[00:51] <fx3> sudo make me a sandwitch ?
[01:00] <fx3> you know, call it with sudo ?
[01:01] <dsmith-work> Ya
[01:05] <fx3> you heard of xkcd right?
[01:07] <fx3> dsmith-work, http://xkcd.com/149/
[01:07] <fx3> got to go, bye
[01:09] <dsmith-work> yes
[02:27] <kansan> is there a wayshould i install nagios from source on ubuntu?  or via package?
[02:30] <twb> When moving an existing host's root filesystem onto LVM, ever seen "LV Status: NOT available" when you get dumped into the busybox shell?
[02:33] <jmarsden> kansan: apt-cache search nagios  # will show you the available nagios-related packages :)
[02:34] <kansan> dshould i install via source
[02:34] <kansan> or via ubuntu package
[02:35] <jmarsden> Always use packages when packages exist for what you need.
[02:36] <jmarsden> kansan: See https://help.ubuntu.com/8.10/serverguide/C/package-management.html
[02:41] <Tigermuss> question - How do get x86 emulation on a IA64 installation? ;)
[02:49] <kansan> how does nagios notify me if my server goes down
[02:49] <kansan> if its running on the server?
[02:49] <kansan> anyone wanna break that down for me ? ;p
[03:06] <luckyone> hello server folk
[03:06] <luckyone> when I mount my samba share through /etc/fstab, I am unable to delete files
[03:06] <luckyone> any ideas of what I am doing wrong?
[03:06] <luckyone> I get a permission denied
[03:07] <lamont> dsmith-work: interesting... they should be the same on both distros....
[03:07] <lamont> (zone file ownership_0
[03:19] <racecar56> back
[04:35] <kansan> um, why does my server not respond to pings?  and should it (for nagios monitoring purposes)?
[05:02] <ph0dder> I'm having trouble upgrading 7.04.  I've updated source.list to point to old-versions.ubuntu.com but I'm getting things still pointing to us.archive.ubuntu.com even though it's nowhere in /etc and nowhere in the installer script that I can find:  Failed http://us.archive.ubuntu.com feisty-backports Release
[05:02] <ph0dder> for example
[05:04] <ph0dder> any ideas?
[05:05] <ph0dder> it's 7.04 server and i'm using the command line, if that matters...
[05:07] <racecar-56> hmmm
[05:08] <racecar-56> thats because feisty for some reason isnt there
[05:20] <ph0dder> can that be worked around somehow?
[05:21] <ph0dder> i know it's not there, i switched all sources to old-versions.ubuntu.com.  but it's still looking in us.archive.ubuntu.com
[05:21] <ph0dder> i even grepped the UpdateManager source looking for hard-coded references and such.
[05:21] <ScottK> ph0dder: Are you trying to update your Feisty or upgrade it to Gutsy?
[05:22] <ph0dder> upgrade
[05:22] <ph0dder> i have successfully updated from old-versions.ubuntu.com
[05:22] <ScottK> You'll want both old-versions feisty and us.ubuntu.com gutsy in your sources.list then.
[05:24] <ph0dder> ah
[05:24] <ph0dder> ok i'll try that
[05:57] <ph0dder> now it's complaining about this:
[05:57] <ph0dder> Failed to fetch http://us.archive.ubuntu.com/ubuntu/dists/feisty-backports/main/debian-installer/binary-i386/Packages.gz  404 Not Found [IP: 91.189.88.45 80]
[05:58] <ph0dder> even though us.archive is NOWHERE in my sources.list
[05:59] <ScottK> ph0dder: Did you check in /etc/apt/sources.list.d for additional lists?
[06:01] <ph0dder> yes and commented everything out
[06:01] <ph0dder> it sticks it back in there when it runs
[06:03] <ScottK> How did you install this machine originally?
[06:03] <ph0dder> i didn't.  how would i find that out?
[06:04] <ScottK> The reason I ask is that isn't anything I think a true Ubuntu installer would put in your sources.list.
[06:05] <ScottK> I would suggest go to /etc and grep -r feisty-backports * and see if anything matches
[06:07] <ph0dder> it matches only my sources.list
[06:08] <ScottK> I'd suggest comment anything about feisty backports out of your sources.list
[06:09] <ph0dder> done, then apt-get update:
[06:09] <ph0dder> Hit http://archive.canonical.com gutsy Release
[06:09] <ph0dder> Hit http://archive.canonical.com gutsy/partner Packages
[06:09] <ph0dder> Get:10 http://archive.ubuntu.com gutsy/universe Sources [1226kB]
[06:09] <ph0dder> Get:11 http://archive.ubuntu.com gutsy/multiverse Packages [158kB]
[06:12] <ph0dder> oh crap that last one was my bad, i just moved the sources.list.d directory to tmp when cleaning up files
[06:12] <ph0dder> so forget the segfault
[06:13] <ph0dder> nevermind i put it back and it still segfaults
[06:22] <ph0dder> OK!  When I comment out EVERYTHING from my sources.list file and do a apt-get update, I still get this:
[06:22] <ph0dder> root@www:/etc/apt# apt-get update
[06:22] <ph0dder> Ign http://us.archive.ubuntu.com feisty-backports Release.gpg
[06:22] <ph0dder> Ign http://us.archive.ubuntu.com feisty-backports Release
[06:22] <ph0dder> Ign http://us.archive.ubuntu.com feisty-backports/main/debian-installer Packages
[06:22] <ph0dder> Err http://us.archive.ubuntu.com feisty-backports/main/debian-installer Packages
[06:22] <ph0dder>   404 Not Found [IP: 91.189.88.31 80]
[06:23] <ph0dder> Failed to fetch http://us.archive.ubuntu.com/ubuntu/dists/feisty-backports/main/debian-installer/binary-i386/Packages.gz  404 Not Found [IP: 91.189.88.31 80]
[06:23] <ph0dder> Reading package lists... Done
[06:23] <ph0dder> E: Some index files failed to download, they have been ignored, or old ones used instead.
[06:24] <ScottK> Do you have anything in sources.list that mentions debian-installer?
[06:24] <ScottK> If so, comment it out
[06:24] <ph0dder> everything is commented out, every line begins with a #
[06:25] <ScottK> Weird.
[06:25] <ScottK> Dunno what to tell you then.
[06:31] <ph0dder> ok i deleted source.list.d/* and then it went away.  then, as a test, i added just one line to the top of my sources.list, the first line.  then i ran apt-get update, and it didn't give me any crap about us.archive.ubuntu.com.  BUT, then i ran do-release-upgrade, and it recreated that sources.list.d/prerequists-sources.list file, and filled it with this:
[06:31] <ph0dder> root@www:/etc/apt# cat sources.list.d/prerequists-sources.list
[06:31] <ph0dder> # sources.list fragment for pre-requists (one with countrymirror, one fallback)
[06:31] <ph0dder> deb http://us.archive.ubuntu.com/ubuntu feisty-backports main/debian-installer
[06:31] <ph0dder> #deb http://archive.ubuntu.com/ubuntu feisty-backports main/debian-installer
[06:31] <ph0dder> # below is just for testing
[06:31] <ph0dder> #deb http://archive.dogfood.launchpad.net/ubuntu feisty-backports main/debian-installerroot@www:/etc/apt#
[06:32] <ph0dder> then i run apt-get update
[06:32] <ph0dder> and get a different message than the first time i ran it:
[06:32] <ph0dder> root@www:/etc/apt# apt-get update
[06:32] <ph0dder> Ign http://us.archive.ubuntu.com feisty-backports Release.gpg
[06:32] <ph0dder> Get:1 http://old-releases.ubuntu.com feisty Release.gpg [191B]
[06:32] <ph0dder> Ign http://us.archive.ubuntu.com feisty-backports Release
[06:32] <ph0dder> Hit http://old-releases.ubuntu.com feisty Release
[06:32] <ph0dder> Ign http://us.archive.ubuntu.com feisty-backports/main/debian-installer Packages
[06:33] <ph0dder> Err http://us.archive.ubuntu.com feisty-backports/main/debian-installer Packages
[06:33] <ph0dder>   404 Not Found [IP: 91.189.88.45 80]
[06:33] <ph0dder> Hit http://old-releases.ubuntu.com feisty/main Packages
[06:33] <ScottK> Weird
[06:33] <ph0dder> Hit http://old-releases.ubuntu.com feisty/restricted Packages
[06:33] <ph0dder> Fetched 1B in 1s (1B/s)
[06:33] <ph0dder> Failed to fetch http://us.archive.ubuntu.com/ubuntu/dists/feisty-backports/main/debian-installer/binary-i386/Packages.gz  404 Not Found [IP: 91.189.88.45 80]
[06:33] <ph0dder> Reading package lists... Done
[06:33] <ph0dder> E: Some index files failed to download, they have been ignored, or old ones used instead.
[06:33] <ScottK> !pastebin | ph0dder
[06:33] <ph0dder> crap sorry
[06:33] <ScottK> I need to get to bed, so I'm not sure what to tell you.
[06:33] <ph0dder> ok
[06:34] <ScottK> Do you have any software RAID on this box (MD devices)?
[06:34] <ph0dder> wel thanks for trying
[06:34] <ph0dder> no
[06:34] <ScottK> What I'd do then is go back to the clean sources.list with just gutsy stuff in it and the apt-get dist-upgrade
[06:35] <ScottK> Skip the update-manager.
[06:35] <ph0dder> mmm
[06:35] <ph0dder> ok
[06:35] <ScottK> I suspect some of it's special case magic doesn't know about old releases and is getting confused.
[06:35] <ScottK> That's not the supported way to do it, but generally it works out.
[06:35] <ph0dder> ^^
[06:35] <ScottK> Make backups of course
[06:35] <ph0dder> :) ok i'll try it thanks so much
[06:38] <ph0dder> hey that looks like it's gonna work!
[06:41] <llragsll> can anyone help me with chrooting openssh
[06:41] <llragsll> :?
[06:43] <llragsll> ??
[07:00] <VK7HSE> what application is being used for http://paste.ubuntu.com/ as I would like to put something like that on my server ???
[07:11] <llragsll> Using chrootdirectory option in sshd conf file gives this error : Bad configuration option: ChrootDirectory?? any sol
[07:13] <racecar-56> gotta go
[08:27] <VK7HSE> what application is being used for http://paste.ubuntu.com/ as I would like to put something like that on my server ???
[10:10] <kraut> moin
[10:43] <EQUIV> Hi, anyone who knows how to configure squid?
[10:43] <EQUIV> I get the 401 error
[10:44] <EQUIV> And it prompts for password. I am trying to use it in reverse mode to speed up the web server
[13:07] <incorrect> how can is see what libs a binary is pulling in?
[13:08] <Deeps> ldd
[13:09] <incorrect> thanks
[13:45] <pjsturgeon> hey guys. I think i mounted a HDD wrong (mounted the actual disk not a partition) and now I cant seem to unmount it or run "fdisk -l".
[13:45] <pjsturgeon> fdisk -l gives me "Cannot open /dev/sda"
[13:46] <pjsturgeon> and apparently unmount is not a command, all the tutorials online suggest I do that. AND i rebooted to get rid of the mount (its not in my fstab) and i still cannot run fdisk -l
[13:49] <zul> what version of vnc do people recommend on intrepid?
[13:51] <soren> zul: Client?
[13:51] <soren> Anything based on gtk-vnc.
[13:51] <zul> server
[13:53] <soren> tightvnc, IIRC.
[13:53] <zul> cool thanks
[13:56] <Nicador> Does anybody have an ideea how to disable the black screen if no activity?
[13:59] <kinnaz> in console ?
[14:00] <Nicador> Yes
[14:00] <kinnaz> setterm -blank 0
[14:00] <kinnaz> setterm -powersave off
[14:00] <kinnaz> setterm -powerdown 0
[14:00] <kinnaz> that should keep terminal running
[14:00] <kinnaz> without any input
[14:01] <Nicador> Do i have to reboot or smth ?
[14:01] <VK7HSE> kinnaz: Is that a global setting or just for the console you using ???
[14:02] <VK7HSE> kinnaz: Is that a global setting or just for the console you using ???
[14:02] <VK7HSE> sorry!
[14:02] <kinnaz> just the console you are using
[14:02] <kinnaz> if you put it to .bashrc
[14:02] <kinnaz> it will be set at login
[14:02] <kinnaz> there was some file in etc what is executed by all users
[14:03] <kinnaz> so you could make it global for all users aswell
[14:03] <Nicador> And if it's no user logged in ?
[14:04] <kinnaz> quess not working
[14:04] <kinnaz> because nothing executes it
[14:04] <kinnaz> i use it for holding my screen alive
[14:04] <kinnaz> while running monitor app
[14:04] <kinnaz> to external monitor
[14:04] <kinnaz> so it wouldnot go black
[14:05] <VK7HSE> kinnaz: Thanks! *THUMBS UP*
[14:10] <Nicador> kinnaz but if i run that command on startup ?
[15:49] <fevel> hi friends
[15:49] <fevel> I have created a directory and in this directory I have put three txt files. Each file contains a name and a password on this format " john,passwd" I would like to read all the txt files on this directory and create each user on htpasswd. Can someone guide me? I was informed that I would be able to do this easily with a bash script and awk although I don't know awk that well
[15:52] <Deeps> cat /path/to/files/* | while read line; do user=`echo $line | cut -d "," -f 1`; pass=`echo $line | cut -d "," -f 2`; htpasswd -b passwordfile $user $pass; done
[15:52] <Deeps> off the top of my head
[15:52] <Deeps> you'll probably need to check the htpasswd syntax
[15:54] <evarlast> touch password file first.  htpasswd will complain if it doesn't exist.
[16:02] <fevel> how do I change eth0 to be the onboard ethernet adapter? It has detected the onboard as eth1.
[16:02] <fevel> by the way Deeps
[16:02] <fevel> I did it like this
[16:02] <fevel> or up in $( cat *.txt ) ; do USER=$( echo ${up} | cut -d, -f 1 ) ; PASS=$( echo ${up} | cut -d, -f 2 ) ; htpasswd -b ${USER} ${PASS} ;
[16:02] <fevel> *for
[16:04] <evarlast> cat *.txt | awk -F, '{print "htpasswd -b "$1" "$2}' | sh
[16:04] <evarlast> I like it that way.
[16:05] <fevel> good one everlast
[16:09] <dsmith-work> lamont: You know, I may have changed the ownership of the files on the Debian system.  I have a vague memory of something like that.
[16:09] <dsmith-work> Well it turns out that the permissions stuff is because of app armor.
[16:10] <dsmith-work> I've but my zone files in /var/lib/bind and the journal files in /var/cache/bind and now freeze/thaw doesn't complin.
[16:10] <dsmith-work> s/but/put/
[16:11] <fevel> excuse my repetition but I guess you guys missed my question. How do I change eth0 to be the onboard ethernet adapter? It has detected the onboard as eth1. I knw howto on slackware but on debian and debian-like its different
[16:12] <dsmith-work> There is a udev file I think.
[16:12] <Deeps> udev rules
[16:12] <Deeps> /etc/udev/rules.d
[16:13] <dsmith-work> fevel: Looks like /etc/udev/rules.d/70-persistent-net.rules
[16:14] <fevel> couldnt thank you giuys enough
[16:15] <fevel> Ill remember your names althrough this lifetime
[16:15] <fevel> =))))
[16:24] <ivoks> kirkland: good post
[16:24] <kirkland> ivoks: thanks for adding your comment ;-)
[16:24] <ivoks> it was an easy one :)
[16:24] <kirkland> ivoks: i'm a bit disappointed that someone would use such a venue to tear down the ubuntu server with no specific reasoning whatsoever given
[16:25] <ivoks> kirkland: i was kind of surprised, too
[16:25] <kirkland> ivoks: it's a great post for the Debian Planet
[16:25] <ivoks> kirkland: i wanted to respond, but my ISP gave me lots of problems yesterday, so i had to delay blog posting
[16:25] <kirkland> ivoks: please do, if your ISP is back up
[16:25] <ivoks> kirkland: and even there people would be: huh?
[16:26] <kirkland> ivoks: maybe, maybe not ....
[16:26] <ivoks> kirkland: i had much worse problems than not being on-line :(
[16:26] <kirkland> bummer
[16:26] <waver_> mathiaz> ping
[16:26] <genii> dsmith-work: Yeah thats the file. MAC gets cached there for ethX names and so doesn't re-use eth0   or so on
[16:26] <ivoks> kirkland: but that's a story for a beer, not public :)
[16:27] <waver_> mathiaz>  When you're back, I need to speak with you.
[16:27] <kirkland> ivoks: :-)  sure, i'll buy
[16:28] <hexmasta2> can anyone make any recommendations for a linux mail server that works well with ubuntu and supports pop3 and imap?
[16:28] <waver_> soren> ping
[16:29] <genii> hexmasta2: dovecot
[16:29] <ivoks> hexmasta2: ubuntu provides couple of mta and imap/pop solutions
[16:29] <soren> waver_: Yes?
[16:30] <waver_> soren>  you have some minutes? Can you open a privat discussion (i'm not logged)
[16:30] <soren> Eh?
[16:30] <soren> Just /msg me.
[16:31] <waver_> I can't if I'm not identified
[16:31] <soren> waver_: a) To me you can, b) just register?
[16:45] <ivoks> kirkland: have you noticed strange behaviour when you issue a reboot from screen?
[16:46] <kirkland> ivoks: hmm, strange how?
[16:46] <kirkland> ivoks: (no, to answer your question)
[16:46] <ivoks> kirkland: the 'bar' stayes there when the machine in reboot drops your ssh connection
[16:46] <kirkland> ivoks: ah, yeah, i guess i have seen that
[16:46] <kirkland> ivoks: maybe i need a clear at the bottom of the shell wrapper script
[16:47] <ivoks> well, screen gets killed, right?
[16:47] <ivoks> i don't think clearing from wrapper would solve anything :)
[16:47] <ivoks> this bug is beyond screen it self
[16:47] <ivoks> same thing happens with curses
[16:47] <kirkland> ivoks: okay, i just reproduced that problem
[16:47] <kirkland> ivoks: yeah, it might be a library problem
[16:47] <ivoks> so, right, just ignore me... i'm thinking in public :)
[16:50] <kirkland> ivoks: yeah, it is weird
[16:50] <kirkland> ivoks: its like it doesn't bother to repaint the screen
[16:50] <ivoks> right
[16:53] <Adri2000> mathiaz: can you sponsor the samba sru?
[16:55] <ivoks> bbl
[17:23] <maDChoPR> how do i see what groups a user is in?
[17:35] <mathiaz> Adri2000: it's on my todo list
[17:35] <mathiaz> waver_: Hi
[17:35] <Adri2000> mathiaz: ok
[17:37] <Scix> Whats the point of the auto.master entry in LDAP when in auto.home /home is pointed to the auto.home ou in ldap?
[17:37] <Scix> I used https://help.ubuntu.com/community/AutofsLDAP to setup my server
[17:38] <waver_> Hello mathiaz, I'm looking with cr3 for my problem.
[18:43] <phreestyle-work> hello all....All of my Ubuntu servers are giving me this error this morning and I can't figure out what the problem is.
[18:43] <phreestyle-work> * Starting PostgreSQL 8.3 database server
[18:43] <phreestyle-work>  * The PostgreSQL server failed to start. Please check the log output.
[18:43] <phreestyle-work>    ...fail!
[18:43] <phreestyle-work> but when I go to check the /var/log/postgresql logs, the last log entry is from two days ago
[18:45] <sommer> phreestyle-work: is there anything in /var/log/syslog?
[18:47] <phreestyle-work> sommer: I searched through it but it did not mention anything about postgresql
[19:01] <sommer> phreestyle-work: you might also check /var/log/daemon.log
[19:01] <phreestyle-work> sommer: it doesn't have anything about postgresql either
[19:02] <robertj> I boot my Dell R200 on LTS with a USB keyboard in there and it works fine, but if I unplug and replug it it won't find the keyboard again
[19:02] <robertj> if I however plug in a PS2 keyboard and type in lsusb the USB keyboard will immediately begin working
[19:07] <sommer> phreestyle-work: wich release are you on?
[19:07] <phreestyle-work> the LTS
[19:07] <phreestyle-work> sommer: with latest updates
[19:07] <sommer> phreestyle-work: hardy or dapper?
[19:07] <phreestyle-work> sommer: hardy
[19:08] <sommer> phreestyle-work: is there anything in the logs pertaining to apparmor by chance?
[19:09] <phreestyle-work> syslog just says it was initialized
[19:09] <phreestyle-work> sommer: and the daemon log doesn't have anything about apparmor
[19:13] <sommer> phreestyle-work: and it was working before?
[19:14] <m1r> hello
[19:14] <phreestyle-work> sommer: it was working on three different servers in different locations yesterday. Today they all failed....no changes that I know ofd
[19:17] <m1r> i am trying to share internet conection using this tutorial > http://ubuntuforums.org/showthread.php?t=91370 ; but after install dnsmasq and ipmasq i get this error: dnsmasq: failed to create listening socket: Adress already in use > failed. any tips for troubleshoting how to start dnsmasq properly ?
[19:18] <sommer> phreestyle-work: just to double check there's nothing in /var/log/postgresql/postgresql-8.3-main.log... and your doing sudo /etc/init.d/postgresql-8.3 restart ?
[19:19] <phreestyle-work> sommer: correct....and it takes nearly a minute after I issue the start command before it fails
[19:21] <phreestyle-work> at this point I'm thinking of dumping postgresql....I don't know if pg is the issue or ubuntu is, but every week something terrible happens....last week the servers would lose all data in the databases when the machines were rebooted.
[19:21] <sommer> phreestyle-work: there's got to be an error somewhere, can you tail -f /var/log/postgresql/postgresql-8.3-main.log and /var/log/syslog when you try and start?
[19:22] <phreestyle-work> sommer: alright
[19:24] <phreestyle-work> sommer: /var/log/postgresql/postgresql-8.3-main.log is empty and syslog has nothing about postgresql
[19:24] <sommer> phreestyle-work: does /var/log/syslog have any errors?
[19:24] <phreestyle-work> sommer: the last entry in syslog was from the hourly cron job and that was 10 minutes ago
[19:25] <sommer> phreestyle-work: how about /var/log/auth.log out of curiosity
[19:25] <phreestyle-work> sommer: this is what it says:
[19:25] <phreestyle-work> Feb 20 14:22:07 webserver su[6253]: Successful su for postgres by root
[19:25] <phreestyle-work> Feb 20 14:22:07 webserver su[6253]: + ??? root:postgres
[19:25] <phreestyle-work> Feb 20 14:22:07 webserver su[6253]: pam_unix(su:session): session opened for user postgres by (uid=0)
[19:25] <phreestyle-work> Feb 20 14:22:07 webserver su[6253]: pam_unix(su:session): session closed for user postgres
[19:27] <sommer> phreestyle-work: have you customized /etc/init.d/posgresql-8.3 ?
[19:27] <phreestyle-work> sommer: never touched it
[19:27] <sommer> errr /etc/init.d/postgresql-8.3
[19:27] <sommer> cause my /var/log/auth.log look very different
[19:28] <sommer> there's no su for user postgres entries
[19:28] <phreestyle-work> sommer: well, I'm also using webmin to restart postgresql, so that might be it
[19:29] <sommer> phreestyle-work: I'd say that's pretty definitely it
[19:29] <sommer> phreestyle-work: I would not recommend using webmin
[19:29] <phreestyle-work> sommer: I also did't have a reason to use webmin until I started having pg problems
[19:30] <sommer> phreestyle-work: okay, but it doesn't look like using webmin solved them :)
[19:31] <phreestyle-work> sommer: certainly not....but until eBox can get close to parity with webmin, it's not feasible to teach everyone to do all server admin stuff by hand
[19:31] <sommer> there may be more info in webmin's logs, but otherwise I'd recommend starting over with postgresql... or at least restoring the default configs
[19:32] <phreestyle-work> sommer: well, I tried reinstalling pg, but that led to a whole lot of package and dependency problems
[19:33] <phreestyle-work> sommer: don't ask me why....it just failed miserably....and I did that using apt-get myself, so I know webmin didn't do it
[19:33] <sommer> phreestyle-work: how did you install webmin?  and how did you install postgresql before?  I don't use webmin, so I'm not sure how much more I can help you
[19:33] <sommer> others may know more though
[19:34] <phreestyle-work> sommer: I installed webmin from the webmin site...they had a deb package.....I installed postgresql using apt with the standard ubuntu repos
[19:36] <sommer> phreestyle-work: okay, well looks like webmin uses a different postgresql start script so you might check the webmin logs or check the webmin bug tracker
[19:37] <phreestyle-work> sommer: well, can you help me get rid of postgresql....I'm tried of spending a third of my week always fixing it. Is there a way to get the stuff already in postgresql and move it to mysql or something?
[19:38] <sommer> phreestyle-work: no idea... you might be able to do a csv export type operation then import that into mysql
[19:38] <phreestyle-work> sommer: yea, but if I can't get pg to start, how do I do that?
[19:40] <sommer> phreestyle-work: not sure either, but if I was in your situation I'd create a new ubuntu install (with no webmin), install postgresql, copy the database over then export
[19:40] <sommer> phreestyle-work: but there's probably 10-50 steps in there I missed
[19:41] <phreestyle-work> sommer: where are the databases stored?
[19:41] <sommer> phreestyle-work: on a default install in /var/lib/postgresql/8.3/ but not sure if webmin has moved that
[19:41] <phreestyle-work> sommer: it shouldn't have....I'll check
[19:44] <phreestyle-work> sommer: ok, here's the other thing....I tried to update pg, but I get this error.... "sub-process /usr/bin/dpkg returned an error code (1).
[19:46] <phreestyle-work> sommer: then it tells me "dpkg: dependency problems prevent configuration of postgresql: postgresql depends on postgresql-8.3; however: package postgresql-8.3 is not configured yet"
[19:46] <sommer> phreestyle-work: try sudo dpkg-reconfigure postgresql-8.3 ... I guess
[19:47] <sommer> !webmin
[19:47] <phreestyle-work>  sommer: dpkg says it's broken or not fully installed
[19:49] <phreestyle-work> everything that could possibly go wrong with postgresql has gone wrong it appears....and I don't use webmin to manage postgresql, so it's hard to blame webmin for this one
[19:50] <phreestyle-work> sommer: btw, I have a bunch of folders in "/var/lib/postgresql/8.3/main"....do I backup all of them? (base, global, pg_clog, pg_multixact, etc.)
[19:51] <sommer> phreestyle-work: if webmin changes the postgresql start scripts... I'd blame it, I know that doesn't help you but...
[19:51] <sommer> phreestyle-work: yep, I'd copy the whole folder, and not permissions and such
[19:53] <phreestyle-work> sommer: could I install postgresql-8.2 and get that to work with my data?
[19:55] <sommer> phreestyle-work: I guess in theory, but since you have package dependency issues I'd recommend starting over... without webmin, no webmin at all
[19:55] <phreestyle-work> sommer: what's the best way to manage postgres without webmin?
[19:56] <phreestyle-work> sommer: well, postgres-8.2 starts up.....8.3 stlll refuses
[19:57] <sommer> phreestyle-work: the cli, pgadmin3, phppgadmin
[20:56] <markVa> hi folks, can anyone point me towards a faq or how-to on how to set up multiple virtual interfaces on a kvm VM?  I've got bridging set up fine, but I need (if possible) to have more than one interface on the VMs.
[21:02] <markVa> basically, I need more than one IP address, since apache+SSL barfs with only one IP with multiple virtual hosts.
[21:06] <domas> markVa: does it? I was able to do virtual hosts with only one IP! :)
[21:07] <ivoks> you can do virtual hosts
[21:07] <ivoks> but http protocol doesn't allow you to have multiple virtual hosts on https
[21:07] <ivoks> it does allow, but certificate will be invalid
[21:07] <ivoks> since you set up a crypted connection *before* you ask for a domain
[21:11] <ivoks> take care
[21:13] <markVa> hrm, well, I've never managed to get it to work properly with one IP....  Does this mean my original idea won't work?  when I attempt to alias on my VM, I get the ubiquitous "cannot assign the requested address" error.
[21:16] <dsmith-work> I've heard that before.  https and virtual hosts don't mix. You only get one.
[21:18] <dsmith-work> Basically, the web server parsers the http stream for the hostname and serves up different pages.  That's way too late for https.  The secure connection is already set up before then.
[21:18] <blue-frog> as ivoks said, you can have https for several vhosts and one IP but ONLY ONE certificate
[21:18] <dsmith-work> (iirc)
[21:19] <dsmith-work> Ahh, ya that makes sense.
[21:19] <markVa> dsmith-work: that was my understanding
[21:20] <Deeps> indeed, so only one domain will potentially not complain about invalid certificate
[21:21] <dsmith-work> markVa: Can't you just set up aliases in the vm's?
[21:22] <dsmith-work> ifconfig eth0:1 addr1; ifconfig eth0:2 addr2; etc
[21:23] <Deeps> aliasing is deprecated! ip addr add!
[21:23] <dsmith-work> ok
[21:23]  * Nafallo agrees with Deeps 
[21:24] <dsmith-work> my fingers and head know ifconfig.
[21:24] <markVa> hrm, maybe that's my problem - since ifconfig etho:1 <addr> doesn't work.
[21:24] <dsmith-work> never did figure out ip
[21:24] <Deeps> eth0, not etho
[21:24] <Deeps> zero not the letter o
[21:24] <markVa> mistype here, not commandline, sorry :)
[21:26] <markVa> I don't know ip addr..., will go check the man page, thanks for the tip.
[21:27] <Deeps> ip address add <new ip>/<cidr netmask> dev <device, e.g eth0>
[21:27] <Deeps> e.g.
[21:27] <Deeps> ip addr add 192.168.0.2/32 dev eth0
[21:29] <markVa> Deeps:  thanks - what's the equiv to ifconfig -a, if you don't mind the extra bit?
[21:29] <Deeps> ip addr
[21:29] <Deeps> or just ip a
[21:30] <markVa> kk, thanks much!
[21:30] <Deeps> the command is 'ip address', but commands can be shortened, there's no other param that begins with a for ip, so can just call ip a
[21:32] <dsmith-work> So in /etc/network/interfaces,  just add "iface eth0:1 static ...  iface eth0:2 static ... with appropriate lines for ..., right?
[21:32] <Deeps> iface eth0:1 inet static
[21:33] <Deeps> but yeah, for the deprecated aliasing approach
[21:33] <Deeps> if you wanted to add a sequence of ip addresses, you'd be better off with a post-up line in your eth0 declaration
[21:33] <Deeps> e.g. post-up for i in `seq 10 20`; do ip addr add 192.168.1.$i dev eth0; done
[21:33] <Deeps> to add 192.168.1.10 - 192.168.1.20 to eth0
[21:34] <Deeps> or multiple post-up lines if you have non-sequential ips
[21:35] <dsmith-work> Hmm. Ok.
[21:35] <Deeps> i've been led to believe that adding multiple ips to a single interface has the additional benefit of using less memory
[21:35] <markVa> I could add these just fine,  many thanks - I'm guessing though, as I just tried it, that these are not exposed downstream?
[21:45] <SyL> if I have a question about ubuntu-server, but it's jaunty, should I be asking it in here?
[21:45] <Deeps> markVa: they are
[21:46] <Deeps> you can bind to them like any other ip
[21:49] <markVa> Deeps: got it, yet another stick error - many thanks for the help.  I'm going to reset everything now to get everything fixed up properly.
[21:51] <KillMeNow> hello everyone...  anyone here get apparmor to work with apache2 / php / mysql ?
[21:51] <KillMeNow> using ubuntu 8.10
[21:52] <Deeps> i'd imagine it would work out of the box, assuming you kept things in default locations
[21:55] <Zerqent> KillMeNow: just install libapache2-mod-php5 php5-mysql and mysql-server and it should work I guess
[21:55] <KillMeNow> yea, i have the server up and working fine w/o apparmor...  i want to put in some security in case of any security bugs that happen
[21:56] <KillMeNow> so i've been reading an playing around with apparmor but it's hosing the websites up
[21:57] <SyL> I just installed Jaunty Jackalope and did apt-get upgrade and now the it won't boot and stalls at DMA eth1. Anybody ever see this before?
[21:58] <KillMeNow> so it just stalls at that point for how long?  do you have it set to use DHCP to obtain IP address for Eth1?
[21:59] <SyL> no, static IP. this happened after a kernel update
[21:59] <SyL> both 32-bit and 64-bit
[21:59] <KillMeNow> yea, i have had something similar happen on version 6 using a megaraid controller
[21:59] <KillMeNow> wound up having to roll the dist version backwards 1 rev
[22:00] <SyL> hrm... there is some raid card in there. that could be it
[22:00] <KillMeNow> are you using the raid card?
[22:00] <SyL> no. there is 4 drives and 3 are going to be software raid 5 and the other will be system
[22:01] <SyL> we're testing out the KVM/XEN kernel for ubuntu
[22:01] <KillMeNow> i'll leave it up to you, but if you're not using the card you might want to power the box off, pull the card and try booting the system again
[22:01] <SyL> it's a 1U server, I'm guessing it's onboard. =)
[22:01] <KillMeNow> i personally never been a big fan of software raid
[22:01] <KillMeNow> disable it in the bios
[22:02] <SyL> KillMeNow: software raid is good if your help doesn't know much about computers and they'll be the ones swapping drives at 5am... =)
[22:02] <SyL> I will disable it in the bios next...
[22:03] <KillMeNow> i suppose if it's not a mission critical box and isn't heavilly loaded software raid is a cheaper solution...  *shrug* personal preference i suppose
[22:03] <KillMeNow> let me know if disabling the raid in the BIOS worked
[22:17] <SyL> KillMeNow: hah! I went into the Bios and there is no way to disable the card. awesome!
[22:20] <KillMeNow> wunderbar!
[22:23] <KillMeNow> well, dunno what to tell ya..  you can boot the box and then select which distribution to load
[22:24] <SyL> KillMeNow: well, it boots after I install a fresh OS on it. but if I do apt-get upgrade it adds a new kernel. I reboot and it dies
[22:27] <KillMeNow> sorry about delya
[22:27] <KillMeNow> er delay
[22:27] <KillMeNow> was on phone
[22:27] <KillMeNow> well, do you know what type of raid controller is on the MB?
[22:27] <KillMeNow> course, you said it's failing at the DMA for ETH1 right?
[22:35] <flustrated> can someone help me with a ubuntu 8.04 problem with python.  I've googled and read through the ubuntu forums and can't find a way to fix my problem.  Error I'm getting is this when trying to do any apt-gets or apptitudes:  Could not find platform independent libraries <prefix> Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
[22:40] <KillMeNow> sorry flustrated...  not sure what to do about that one
[22:41] <KillMeNow> almost sounds like your user account has lost is 'path' to the python binaries
[22:42] <flustrated> I actually sudo su - so that I could get a root.  wonder if that could be the problem?
[22:58] <KillMeNow> if you're running as root?  not likely...  but as root do this:  echo $PATH
[22:58] <KillMeNow> make sure root's path has the location where the Python binaries are and such
[23:00] <flustrated> yep, they're just in /usr/bin
[23:00] <flustrated> which python /usr/bin/python
[23:02] <KillMeNow> yea, then i dunno
[23:03] <flustrated> so, I'm digging through this.. and found this.
[23:03] <flustrated> http://www.mibbit.com/pb/dCJgTg
[23:04] <flustrated> that's a pastebin
[23:07] <KillMeNow> GTK huh?
[23:08] <KillMeNow> yea, i would probably doing the same google digging as you have done
[23:08] <flustrated> okay, well I will just keep at it... I have this server at a remote site... is there a way to completely reinstall it without hands on?
[23:09] <KillMeNow> reinstall the entire OS?  probably not, but i wouldn't go that far just yet
[23:11] <flustrated> it all started with patching and the ssl binaries upgrading the ssh stuff and replacing my keys.. broke my nxserver so I uninstalled it, but now can't do hardly anything because I have a long list of apps that try to install everytime I do an apt-get install xxxxxx
[23:11] <KillMeNow> what version did you install?
[23:12] <flustrated> so after that broke actualy in the middle, after uninstalling the nxserver I upgraded to 8.04 from 7.10
[23:12] <KillMeNow> i'm seeing alot of ppl posting on google about upgrading from 7.10 to 8.04
[23:12] <KillMeNow> ahh yea
[23:12] <flustrated> so I have python 2.4 and 2.5.2 installed.. but nothing in /usr/local/bin (lots of people had an extra python installed there)
[23:13] <KillMeNow> yea
[23:13] <flustrated> my symlink goes from python to 2.5.2
[23:13] <flustrated> actually python2.5
[23:15] <KillMeNow> going some quick googling
[23:20] <KillMeNow> yea sorry flustrated...  no love for you
[23:21] <flustrated> thanks for trying to help.
[23:21] <KillMeNow> i know there was a bug reported on it for the dev's of ubuntu
[23:21] <KillMeNow> but that was back in april of 2008
[23:21] <KillMeNow> would've figured they fixed it by now
[23:22] <KillMeNow> i never went to 7.10...  i installed 8.04 and then rebuilt it using my back ups from Ubuntu 6
[23:22] <KillMeNow> well, the parts of the back ups regarding my websites and postfix stuff
[23:22] <KillMeNow> everything else i rebuilt by hand
[23:23] <flustrated> I just hate taking a visit to the data center..  I need to get the lights out stuff working on it.
[23:23] <KillMeNow> i suppose you could possibly rebuild the box if you had it on ESX or perhaps you had KVM over IP to get to the bios
[23:24] <KillMeNow> yep...  build a image box
[23:24] <flustrated> it has compaqs version of a DRAC, just don't have it setup or networked yet.
[23:24] <dazman> iLo is priceless, really.. :)
[23:25] <KillMeNow> yep, lights out has saved my hiney numerous times
[23:25] <KillMeNow> but then again, i don't live from the data center either
[23:25] <KillMeNow> that's far from
[23:25] <flustrated> we have it on all of our new windows boxes and it's awesome.
[23:25] <KillMeNow> course, now it doesn't matter cause i don't work for that company anymore
[23:26] <flustrated> i've got about 40 unix probes that have some lights out card mgmt built into it, but the vendor that supports the boxes doesn't support the lights out stuff.
[23:27] <domas> hi! how to check how much memory does apache use?! (tricky question :)
[23:27] <KillMeNow> heh, yea that's a loaded question domas
[23:28] <KillMeNow> those vendor bastages flustrated!
[23:28] <KillMeNow> how many instances of apache have you forked ?
[23:28] <domas> ~20
[23:28] <KillMeNow> how many mods are getting pulled in
[23:28] <domas> lots :)
[23:28] <domas> APC and such
[23:29] <KillMeNow> so if you do a ps -ef | grep apache how many lines does it spit out?
[23:29] <domas> how much memory is shared after forking? :)
[23:29] <KillMeNow> AFAIK, each fork uses it's own "slice" of memory and doesn't share it perse
[23:29] <domas> see, summing up RSS or VSZ doesn't really work
[23:30] <domas> KillMeNow: pages can be shared by forked kids, it is copy-on-write :)
[23:30] <domas> but ps/top doesn't account for that!
[23:30] <KillMeNow> nope it doesn't
[23:30] <domas> I know one way to answer!
[23:30] <domas> cat /proc/meminfo; killall -9 apache2; cat /proc/meminfo
[23:30] <KillMeNow> there ya go!
[23:30] <domas> now, how to do that without killing? :)
[23:31] <domas> interesting though, say, libc is shared among _all_ system processes
[23:31] <domas> so you have to remove it from every process out there
[23:31] <KillMeNow> but doing the ps -ef tells you exactly how many forks you have in memory
[23:31] <domas> does it mean anything? :)
[23:32] <KillMeNow> well, i'm not sure how to help you w/o killing all the processes
[23:32] <domas> meh
[23:33] <KillMeNow> unless you know of a way to show the memory usage for each PID
[23:33] <domas> so, my trolling goes to hell?
[23:33] <mark> you did help domas.
[23:33] <domas> I was expecting fifty folks to jump and tell that I'm idiot and should use ps and top :(
[23:33] <domas> mark: haha
[23:35] <domas> did anyone notice there's no Hoard package in Ubuntu?
[23:43] <dsmith-work> Hrm.  There are some files in proc that list exactly which pages a process has access to.
[23:45] <mark> here's this weekend's project domas
[23:46] <domas> I want to make Hoard package first
[23:46] <mark> I like the sound of that
[23:47] <dsmith-work> Maybe I'm thinking of /proc/PID/maps
[23:47] <mark> yes you are
[23:47] <domas> dsmith-work: doesn't help with anon-pages
[23:47] <domas> and some mmap'ed stuff goes to VSIZE, some to RSS, I guess
[23:47] <domas> argh, too complicated
[23:50] <domas> maps don't show physical address mappings
[23:50] <domas> just which process address space is mapped to what
[23:50] <dsmith-work> ya
[23:50] <KillMeNow> so domas, i take it you don't want to know the VSZ part of the memory usage?
[23:51] <domas> KillMeNow: nope! ;-)
[23:51] <KillMeNow> well, when you figure it out, you can educate the rest of us
[23:52] <mark> i'm sure it will be a blog entry
[23:52] <domas> I want to count physical memory pages used by family of processes
[23:52] <domas> orly!?!?
[23:52] <domas> :)
[23:52] <domas> my next one will be about hoard! :)
[23:52] <mark> you can do two at once
[23:52] <domas> I have to quote one colleague about it
[23:52] <domas> it has something about weird bald guys with sunglasses and crossed legs ;-)