[01:46] <roasted> hello friends
[01:46] <roasted> Question - using mysql as an example, when I install it, I get a password prompt. If I uninstall and reinstall, I get nothing. How do I 100% reinstall the package so I get a password prompt like the initial install?
[01:46] <sarnold> roasted: try apt-get purge mysql-whatever before installing it
[01:47] <sarnold> roasted: dpkg --remove removes most of the files, but leaves configuration files around. dpkg --purge removes the configuration files, too. apt-get purge uses the more forceful of the two methods..
[01:55] <stevePage129> for some reason my brand new VPS in which i havent installed or setup anything hands everytime apt-get install anything on UNPACKING. disk IO write times included in post (but they seem like they check out...) http://askubuntu.com/questions/315672/ubuntu-vps-hangs-constantly-when-unpacking-anything-apt-get-or-dpkg-hangs
[01:55] <stevePage129> any help would be greatly apprecaited
[01:56] <stevePage129> hangs*
[02:01] <roasted> sarnold: I did do a purge... but reinstalling and dpkg-reconfigure does nothing :/
[02:02] <sarnold> stevePage129: your diskspeed test file is way too small, it might not have even hit disk yet...
[02:03] <sarnold> stevePage129: also, the upacking and man-db operations work on thousands of small files, not one huge file -- perhaps the filesystem is mounted strictatime, and every little access dirties inodes...
[02:04] <sarnold> roasted: oh. Maybe I mis-understood which password prompt. I assumed it was something specific to mysql's configuration .. is it the sudo password prompt?
[02:04] <roasted> sarnold: it's where it asks for the root mysql password 2x.
[02:04] <sarnold> roasted: aw nuts, then I did understand :) but now I don't know.
[02:05] <roasted> sarnold: I nuked mysql thinking I'd want to use sqlite for owncloud, but then I found my mysql config I used before, so... I wanted to use that instead.
[02:05] <sarnold> hehe
[02:05] <roasted> but now when I try to fire up mysql-server it doesn't ask me for a password
[02:12] <roasted> tried apt-get remove, apt-get purge, apt-get autoremove, then reinstalled
[02:12] <roasted> no dice
[02:13] <stevePage129> sarnold: any suggestion as to a command, or a method of which i could use to, say, test the NFS which my VPS host is using, in order to test for issues and or bottle necks? im seriously having issues just getting the initial software (lamp, webmin, etc) setup for using my new VPS. the idea was to move from shared hosting to VPS to get better performance, not worse. this isnt normal for VPSs is it?
[02:13] <sarnold> roasted: does /etc/mysql/my.cnf survive the apt-get purge mysql-whatever ?
[02:13] <sarnold> roasted: note that /etc/mysql/my.cnf is owned by mysql-common, are you purging that package, too?
[02:14] <sarnold> stevePage129: it depends heavily upon a -lot- of factors. amazon's micro tier has severely penalized IO; bursts go quickly, but sustained IO is heavily penalized. installing an OS on a micro instance can take forever.
[02:16] <roasted> sarnold: trying that now
[02:17] <roasted> sarnold: my.cnf isn't around after the purge
[02:17] <roasted> I assume at this point I should expect an install to work?
[02:17] <sarnold> roasted: I'd hope so :)
[02:17] <roasted> ah bingo!
[02:17] <roasted> you rock
[02:18] <sarnold> sorry it took me ages to notice the -common package.
[02:18] <roasted> ha, all good
[02:18] <roasted> it's done :P
[02:18] <sarnold> :D
[02:19] <sarnold> stevePage129: maybe soe cheap tests like timing: for i in `seq -w 1 10000` ; do echo $i > foo_${i} ; done   ... that'll generate a huge pile of tiny files...
[02:20] <roasted> sarnold: out of curiosity, how did you catch that -common was relevant?
[02:20] <stevePage129> sarnold: i suppose this is a little specific.. but have you ever heard of cinfu.com? i originally moved from GoDaddy's shared hosting becaused they were dropping my connections all the time when transferring files over 25 mb. now i have purchased a VPS, i am having trouble simply setting up the host operating systems software (IO issues?). i really dont have to get a dedicated server just to get something that responds grac
[02:20] <stevePage129> sarnold: trying that now
[02:20] <stevePage129> well, will look into those commands some
[02:21] <sarnold> roasted: I ran 'apt-file search my.cnf" to find the full path to the file, and noticed then that the mysql-common package owned it
[02:21] <roasted> sarnold: next question would be how you knew my.cnf was relevant?
[02:21] <sarnold> roasted: I figured that mysql probably stored the root password there..
[02:21] <sarnold> stevePage129: irc has line length limits; you were cut off at "responds grac"
[02:22] <sarnold> .. which isn't a very graceful way for irc to respond. hehe. :)
[02:22] <stevePage129> sarnold: sorry about that, using some cheapskate client
[02:22] <sarnold> stevePage129: hehehe
[02:22] <stevePage129> sarnold:  really dont have to get a dedicated server just to get something that responds gracefully do i? (you dont have to answer that ^.^)
[02:24] <sarnold> stevePage129: I'd really hope vps is sufficient, it is the direction industry is headed..
[02:26] <stevePage129> sarnold: indeed. like for example, i ran sudo apt-get update/upgrade to satisfy several updates which SSH and ubuntu were complaining about. this took almost 2 hours, and i still do not know if it ever finished (machine went to sleep while i was napping connection was lost...)
[02:26] <sarnold> stevePage129: oh jeeeeeeze
[02:26] <sarnold> stevePage129: I'd say to install screen or tmux, but I'm afraid how long that would take
[02:27] <stevePage129> sarnold: will read more up on those now
[02:27] <stevePage129> host wanted to try moving me to another node, but i wanted to make sure there was nothing i could do first
[02:28] <stevePage129> if i cant get this working, im going to have to go crawling back to godaddy... lol
[02:28] <sarnold> stevePage129: take them up on that offer. it ought to be quick..
[02:38] <sarnold> stevePage129: okay, I signed back into my amazon control panel, created a new micro instance, installed 12.04.2 LTS onto it, and I'm timing an upgrade now...
[02:39] <sarnold> stevePage129: real	1m49.057s
[02:39] <stevePage129> sarnold: thank you, thank you
[02:39] <stevePage129> sarnold: ...
[02:40] <sarnold> stevePage129: that was for 71 megabytes of downloads, 48 upgraded, 3 newly installed
[02:40] <stevePage129> yeah, something isnt right
[02:40] <sarnold> stevePage129: that's an amazon US micro instance. (it went faster than I expected)
[02:40] <sarnold> stevePage129: so make sure what you're paying for can come close.
[02:40] <sarnold> time for dinner :)
[02:40] <stevePage129> sarnold: paying for a VPS with dual core 2ghz, 1 gb ram
[02:41] <stevePage129> 100 mbit connection
[02:41] <stevePage129> and no bandiwtdh monitoring :@
[02:41] <stevePage129> ty very much tho
[03:10] <centaur5> I want to setup my 1st HA cluster. Anybody have an opinion on best storage options? GlusterFS, corosync, or pacemaker?
[03:32] <roasted> hi
[04:05] <sarnold> ay
[04:27] <roasted> hi
[04:27] <roasted> I have a really, raelly dumb question
[04:27] <roasted> www-data is the "webserver user" of apache, no?
[04:27] <roasted> I'm getting an error from owncloud saying the webserver user needs write access to the apps dir within owncloud, and www-data has rwx perms but I still get the error.
[04:27] <roasted> failing to see what it's crapping itself over.
[04:29] <sarnold> roasted: how about the containing directories?
[04:29] <roasted> the dir above apps?
[04:29] <sarnold> yeah
[04:29] <roasted> rwx
[04:30] <roasted> ah wait
[04:30] <roasted> I just spotted something in their config that's still pointing to /var/www
[04:30] <roasted> I changed that
[04:30] <sarnold> aha :)
[04:31] <roasted> nice, now I have a blank white screen
[04:31] <roasted> quick, stab me
[04:31] <shauno> dr. plum, in the dining room, with the error.log
[04:33] <roasted> NO WORRIES. WE GOT THIS.
[04:33] <roasted> I guess I shouldn't have jumped the gun and also changed the sqlite3 entry to mysql
[04:33] <roasted> I'm using mysql, so, no idea why that is
[04:33] <roasted> but changing back to sqlite3 worked
[04:34] <roasted> ty shauno. picked that up in the error.log
[04:34] <roasted> I really need to not do this crap when I'm in a tired coma...
[04:35] <shauno> if it's any consolation, you've got me wondering if I ever moved mine off sqlite too
[04:35] <roasted> with owncloud?
[04:36] <shauno> yeah
[04:39] <roasted> yeah if I swithced mine to mysql it crapped itself
[04:39] <roasted> but back on sqlite3 it's fine
[04:39] <roasted> yet I'm actually using mysql AS my db....
[04:39] <roasted> maybe I'm not understanding something, but whatever. It works, so...
[05:00] <RustyShackleford> so I installed noip2 from noip's website
[05:00] <RustyShackleford> how do I set it to run at boot?
[05:00] <RustyShackleford> (ubuntu server 13.04)
[05:03] <sarnold> RustyShackleford: you've got a few options..
[05:03] <sarnold> RustyShackleford: if they provided a sysv initscript, you can drop the file into /etc/init.d/ and make the /etc/rc*.d/ symlinks
[05:03] <sarnold> RustyShackleford: or you can call a program from /etc/rc.local
[05:04] <sarnold> RustyShackleford: or you can use the @reboot cron specifier if you want to run it as a user accout..
[05:13] <RustyShackleford> the other thing that's nagging me: can I force comcast to renew my IP
[05:13] <RustyShackleford> i'd like to test and make sure this works
[05:14] <sarnold> RustyShackleford: I've had the same comcast IP for years.. I'm pretty sure I've 'release' on the router with no effect
[05:15] <RustyShackleford> interesting. I haven't really paid attention, so I can't say
[05:17] <RustyShackleford> sarnold, does it matter what I name the initscript?
[05:17] <sarnold> RustyShackleford: no
[05:18] <RustyShackleford> the readme made it seem like it was important
[05:18] <sarnold> the name matters a lot in the /etc/rc?.d/ directories
[05:18] <sarnold> Sxxwhatever and Kxxwhatever, where xx are digits..
[05:21] <RustyShackleford> sarnold, could you explain or tell me what to google?
[05:23] <sarnold> RustyShackleford: well...
[05:23] <RustyShackleford> it seems like the numbers correspond to runtime levels?
[05:24] <RustyShackleford> idk, i"m just guessing
[05:24] <sarnold> RustyShackleford: you could also write an upstart conf file. that might be a bit more work if the project supplied a sysvinit script, but it'll be much less work if they didn't..
[05:24] <sarnold> RustyShackleford: .. and upstart is significantly easier to configure than the sysv initscripts
[05:24] <RustyShackleford> an initscript is written in bash?
[05:24] <sarnold> RustyShackleford: more usually, sh
[05:25] <RustyShackleford> they did provide something which I can place in /etc/init.d/rcX.d
[05:25] <RustyShackleford> what is X?
[05:25] <RustyShackleford> or how do I find out
[05:25] <sarnold> RustyShackleford: here's some information writing an upstart configuration: http://upstart.ubuntu.com/cookbook/#concepts-and-terminology
[05:26] <RustyShackleford> from the readme: "Where the 'X' in rcX.d is the value obtained by running the
[05:26] <RustyShackleford> following command
[05:26] <RustyShackleford>         grep initdefault /etc/inittab | awk -F: '{print $2}'
[05:26] <RustyShackleford> "
[05:26] <RustyShackleford> sorry, forgot about the newlines
[05:26] <sarnold> RustyShackleford: in those cases, X -is- runlevel; back in the old days, it was 1 for single user, 3 for multiuser, 5 for multiuser with graphics, 0 for rebooting...
[05:27] <sarnold> wow, I'm forgetting the levels. acheivement unlocked! :)
[05:27] <RustyShackleford> i've never had to deal with runlevels
[05:27] <RustyShackleford> i started with ubuntu 7.10
[05:27] <sarnold> RustyShackleford: so you'd put symlinks into whichever directories you wanted the service to start and stop on -- in case you wanted it running only when graphics were up, and shut down when going down to 2 ...
[05:28] <sarnold> RustyShackleford: update-rc.d can help manage all those scripts. but I detest it's manpage. :)
[05:29] <RustyShackleford> it's a headless server
[05:29] <sarnold> RustyShackleford: ooh, check out the sysv-rc-conf package. it looks nicer than update-rc.d
[05:31] <RustyShackleford> i had to install this package manually ("make install")
[05:32] <RustyShackleford> so I'm not usre this will work
[05:32] <sarnold> RustyShackleford: try: update-rc.d -n defaults <name of the file in /etc/init.d/>
[05:33] <sarnold> RustyShackleford: if the output looks good, leave off the -n and re-run..
[05:33] <RustyShackleford> so just drop it in /etc/init.d/ and not /etc/init.d/rcX.d/
[05:34] <sarnold> RustyShackleford: right; the /etc/rcX.d/ directories are for the S and K symlinks
[05:36] <RustyShackleford> output looks good
[05:36] <RustyShackleford> restarting the server and crossing my fingers
[05:37] <sarnold> RustyShackleford: did you re-run without the -n ? :)
[05:37] <RustyShackleford> oh I saw a command on stackoverflow
[05:37] <RustyShackleford> it didn't include the -n flag
[05:37] <sarnold> aha, good
[05:38] <RustyShackleford> did I need to run with -n first?
[05:38] <sarnold> no
[05:38] <sarnold> many commands include an -n or --dry-run option to show you what they will do
[05:39] <sarnold> I'm paranoid and like to see what will happen before running it :)
[05:39] <RustyShackleford> didn't know that. it does seem useful
[05:39] <RustyShackleford> unfortunately it doesn't seem to work
[05:42] <sarnold> RustyShackleford: any error messages in the log files?
[05:43] <sarnold> RustyShackleford: what permissions are on the script file in /etc/init.d/ ?
[05:43] <RustyShackleford> yeah I think it might have been permissions
[05:44] <RustyShackleford> if I start the script manually it works
[05:44] <RustyShackleford> after setting permissions to 755
[05:46] <RustyShackleford> sarnold, yep it works
[05:46] <RustyShackleford> thank you so much
[05:46] <sarnold> RustyShackleford: woo :)
[05:47] <RustyShackleford> playing around with servers is fun
[05:47] <RustyShackleford> but a major timesink
[05:47] <sarnold> RustyShackleford: too true
[05:47] <sarnold> it'll get more fun
[05:47] <sarnold> and ... well, always a timesink :)
[05:47] <RustyShackleford> does ubuntu server upgrade from version to version well?
[05:48] <ScottK> Yes.
[05:48] <RustyShackleford> the desktop version seems less reliable. I usually just wipe and start clean
[05:48] <ScottK> I run Kubuntu for desktop, so I don't know.
[05:49] <ScottK> That's reliable enough on upgrade.
[05:49] <arooni-mobile> my dns resolutoin was messed up on ubuntu; so i edited /etc/resolv.con and put in name severs. but at top of file says: "# Dynamic resolv.conf(5) file for glibc resolver(3) generated by resolvconf(8)"  so was i not supposed to fix it that way?
[05:49] <ScottK> I've got a server that's been upgraded since gutsy.
[05:50] <sarnold> RustyShackleford: I've upgraded machines half-dozen times.. on the desktop side, worst was that unity forgot focus-follows-mouse and auto-raise settings -- but a few releases ago, it did that once every few weeks anyway
[05:50] <sarnold> RustyShackleford: I'd done machines through seven or eight years of debian, before ubuntu, without troubles..
[05:51] <RustyShackleford> i chose ubuntu because I'm more familiar with packages etc.
[05:51] <sarnold> arooni-mobile: depends. :) If you knew what you were doing, that might be the perfect solution...
[05:51] <RustyShackleford> Seems like debian is more popular for servers. #linux didn't care much for my choice of server OS
[05:51] <ScottK> IIRC, the only time I've had a server upgrade problem was once when I knew I was trying something studip.
[05:51] <sarnold> arooni-mobile: but it'll be overwritten next time dhcp brings an interface up or down, so it might not last long
[05:52] <sarnold> ScottK: hehe :)
[05:52] <ScottK> One of the main reasons to run Ubuntu over Debian is the security hardening features built into Ubuntu.
[05:52] <sarnold> arooni-mobile: so perhaps your edits shold be made to /etc/network/interfaces, or /etc/resolvconf/resolv.conf.d/head
[05:52] <ScottK> The gap narrowed significantly with the release of Wheezy, but Ubuntu is still ahead.
[05:53] <RustyShackleford> I just like that there is a Java ppa
[05:53] <RustyShackleford> etc.
[05:53] <arooni-mobile> my name resolution on my server got royally messed up
[05:54] <arooni-mobile> so i want to make theses resolutions permenent
[05:54] <RustyShackleford> certain proprietary things are hard to install on debian. I dislike "make install"
[05:54] <ScottK> arooni-mobile: /etc/network/interfaces is where I'd put it.
[05:54] <arooni-mobile> ScottK, so i jsut drop in exactly whati put in /etc/resolv.conf?
[05:54] <arooni-mobile> there
[05:54] <arooni-mobile> ?
[05:55] <ScottK> What did you put in resolv.conf?
[05:55] <sarnold> arooni-mobile: no, the format is different, it'd be best to read the interfaces(5) manpage for full details
[05:57] <ScottK> sarnold: Thanks.
[05:57] <arooni-mobile> just nameserver stuff
[05:57]  * ScottK has never messed with resolv.conf directly.
[05:59]  * ScottK can't figure out why it's even installed on a server with static IP?
[06:02] <arooni-mobile> how else would domain names get resolved
[06:02] <arooni-mobile> i mean when you need to dl updates etc
[06:06] <ScottK> I always just put the information in /etc/network/interfaces.
[08:06] <jamespage> morning folks!
[08:06] <feisar_> morning
[08:18] <caribou> people: is Memory leaks in rabbitmq-server a thing of common knowledge ?
[08:19] <caribou> I was able to fill up memory of a Raring 2Gb vm overnight with a simple message passing program
[08:19] <caribou> it seems to be even more severe on Precise
[08:49] <jamespage> Daviey, smb: iscsitarget dkms fixed up for 3.10
[08:50] <smb> jamespage, cool
[08:51] <koolhead17> hi all
[09:34] <jamespage> Daviey, did you get anywhere with go in backports yesterday?
[09:43] <Daviey> jamespage: Sort of.
[09:44] <jamespage> Daviey, fyi I'm looking at the 1.1.1 version thats currently in debian NEW
[09:45] <Daviey> jamespage: We do need to dig out specific reasons why the juju-core team need a newer golang toolchain
[09:46] <jamespage> Daviey, they don't need it - the 1.0.x series should be sufficient
[09:46] <jamespage> but thats not in precise which is the challenge
[09:46] <jamespage> 1.1.1 is preferred and has ARM support unlike 1.0.x which is probably a valid reason to support it
[09:49] <Daviey> jamespage: 2:1-5 is in precise.. Do we know why that is no good?
[09:50] <Daviey> also.. they were pretty keen to bump the epoch :/
[10:32] <jamespage> Daviey, hmm
[10:38] <jamespage> Daviey, I can't help but think that requested a MRE for a point release on a series no longer supported upstream is stupid
[10:52] <Daviey> jamespage: Hmm, can you give more details?
[10:52] <Daviey> jamespage: We need to write all this up..
[12:33] <jamespage> zul, adam_g, roaksoax_: so - zul has been working on getting the newer version of kombu into the archive
[12:33] <zul> jamespage:  so amqp got uploaded yesterday im in the middle of doing the MIR
[12:34] <jamespage> follow on question is really around rabbitmq - should we provide a CA backport for havana on 3.1.x as in saucy?
[12:34] <jamespage> specifically to support non-shared storage active/active deployments
[12:35] <jamespage> Daviey, still comtemplating
[12:35] <jamespage> contemplating rather
[12:35] <zul> jamespage:  do we need to backport erlang as well?
[12:35] <jamespage> no idea
[12:36] <zul> +1 if we dont have to backport erlang for it, -1 if we do
[12:36] <jamespage> tbh the erlang backport is probably zero cost
[12:37] <jamespage> I've done the merges for the last few cycles and its pretty much no-op
[12:38] <zul> sure merges can be different from backports though (subunit comes to mind)
[12:39] <zul> jamespage/roaksoax/Daviey: https://bugs.launchpad.net/ubuntu/+source/python-amqp/+bug/1197390
[12:39] <zul> amqp MIR
[12:43] <jamespage> zul, I think erlang in 12.04 is probably still OK
[12:43] <zul> jamespage:  +1 from me then
[12:57] <zul> jamespage:  did you +1 swift for the CA yesterday/
[12:57] <jamespage> zul, probably - lemme +1 it again  - URL?
[12:57] <zul> http://people.canonical.com/~chucks/ca/
[13:29] <jamespage> zul: +1
[13:29] <zul> jamespage:  thanks
[13:32] <Duologic> I have a Ubuntu server "precise" with Samba 3.6.3 running, is it safe to add quantal repositories to apt sources to upgrade to Samba 3.6.6? or at least, will it work?
[14:11] <sadiq_> No such process quotaon: Quota format not supported in kernel.  ? any idea ubuntu 12.04.1
[14:13] <sadiq_> Quota format not supported in kernel.  ? any idea ubuntu 12.04.1
[14:15] <genii> sadiq_: You could try loading the modules quota_v2 and quota_v1
[14:17] <sadiq_> how to load them
[14:17] <sadiq_> thanks for your reply
[14:18] <genii> sadiq_: sudo modprobe quota_v2 && sudo modprobe quota_v1   ...for instance
[14:20] <sadiq_> hi
[14:44] <ndee> hi there, I created a jail for a user and want to mount --bind a directory into his home. How can achieve that the mount will be available after a reboot?
[14:45] <feisar_> ndee: I haven't created a chroot jail before but assuming it's the same as any other mount then you need to put it in /etc/fstab
[14:45] <ndee> feisar_, ok
[14:46] <feisar_> the syntax is a little different from mounting on the command line so check out the fstab man
[14:46] <ndee> feisar_, I checked the man mount and there was the entry too, but thanks :)
[14:47] <feisar_> so you have an entry in /etc/fstab but it isn't working?
[14:48] <ndee> feisar_, I can't test it since I don't want to reboot the server ;)
[14:48] <ndee> but I added the entry and I will check the next time the server reboots, hope that will be some years from now ;)
[14:49] <feisar_> you can $ sudo mount -a to test
[14:49] <feisar_> $sudo mount -a
[14:49] <ndee> sweetness, it works. Thanks feisar_ !
[14:50] <feisar_> no probs : )
[14:50] <feisar_> there's always a way to do it on Linux without rebooting ; )
[14:50] <ndee> feisar_, true :) it's just about finding the way :)
[15:14] <psivaa> hello, i have been having an issue with most of our server smoke tests today: Unable to ping the VMs from the host
[15:15] <psivaa> dhclient: execve (/sbin/dhclient-script, ...): Permission denied is shown in syslog
[15:15] <psivaa> http://pastebin.ubuntu.com/5840606/ contains more logs.
[15:15] <psivaa> hallyn: jamespage: would you be able to shed some light pls?
[15:35] <hallyn> psivaa: can you ping the vms from the host by ip addrses?
[15:38] <psivaa> hallyn: there is no ipv4 address allocated to the concerned VMs
[15:38] <psivaa> hallyn: everything works ok until installation and the issue comes up only when the VM's gets rebooted after the install
[15:42] <hallyn> psivaa: oh, i see.  so dnsmasq is broken on the host.  what does ps -ef | grep dnsmasq show?
[15:43] <psivaa> hallyn: http://pastebin.ubuntu.com/5840697/ on the host
[15:49] <hallyn> psivaa: ls -l /sbin/dhclient-script
[15:49] <psivaa> hallyn: -rwxr-xr-x 1 root root 13005 Nov  7  2012 /sbin/dhclient-script
[15:51] <psivaa> hallyn: the file has the same permissions in the VMs as well
[15:58] <hallyn> psivaa: head -3 /sbin/dhclient-script
[16:01] <psivaa> hally:
[16:01] <psivaa> #!/bin/bash
[16:01] <psivaa> # Explicitly set the PATH to that of ENV_SUPATH in /etc/login.defs and unset
[16:01] <psivaa> # various other variables. We need to do this so /sbin/dhclient cannot abuse
[16:01] <psivaa> # the environment to escape AppArmor confinement via this script
[16:01] <psivaa> # (LP: #1045986). This can be removed once AppArmor supports environment
[16:01] <psivaa> # filtering (LP: #1045985)
[16:01] <psivaa> hallyn: ^ sorry for the typo :)
[16:03] <hallyn> psivaa: do you have auditd running on the host?
[16:03] <hallyn> psivaa: I'm wondering whether there are any apparmor denial messages hidden anywhere
[16:03] <hallyn> psivaa: please try running "strace -f -ooutout /sbin/dhclient-script" (as root) and pastebin the resulting outout file?
[16:05] <psivaa> hallyn: auditd does not seem to be running  and the above strace command returns nothing
[16:05] <hallyn> outout is empty?
[16:06] <psivaa> hallyn: yes
[16:17] <hallyn> psivaa: which host?
[16:17] <psivaa> hallyn: aldebaran
[16:31] <zul> Daviey:  can you seed python-amqp please https://bugs.launchpad.net/ubuntu/+source/python-amqp/+bug/1197390
[16:33] <hallyn> psivaa: grep 'DENIED' /var/log/kern.log ?
[16:34] <psivaa> hallyn: that too ouputs empty
[16:35] <psivaa> hallyn: i see Jul  3 16:29:56 aldebaran snmpd[3717]: error on subcontainer 'ia_addr' insert (-1) in syslog but i hear that's not too harmful
[16:39] <psivaa> hallyn: also this issue does not happen always, it's intermittent
[16:46] <psivaa> hallyn: there is output for grep 'DENIED' /var/log/kern.log in the client VM though,
[16:46] <psivaa> kernel: [    6.185870] type=1400 audit(1372862341.310:8): apparmor="DENIED" operation="file_perm" parent=509 profile="/sbin/dhclient" name="/bin/bash" pid=517 comm="dhclient" requested_mask="r" denied_mask="r" fsuid=0 ouid=0
[16:51] <hallyn> psivaa: can you ask sarnold to take a look?  it sounds to me like dhclient isn't allowed to run /bin/bash - it has special allowance for running /sbin/dhclient-script (see /etc/apparmor.d/sbin.dhclient), so i'd expect...
[16:52] <hallyn> (or jjohansen )
[16:52] <hallyn> you might just add /bin/bash rx to /etc/apparmor.d/local/sbin.dhclient and see if that fixes it...
[16:58] <sarnold> psivaa, hallyn, how interesting, my sbin.dhclient has a /etc/dhcp/dhclient-script Uxr, permission -- that script can do anything is wishes..
[16:58] <hallyn> sarnold: so does his
[16:58] <hallyn> (at least on the host)
[16:59] <hallyn> I see, I looked up the Ux semantics, caught the 'cleaned up', but missed the unconfined.  got it
[16:59] <hallyn> sarnold: but it's /sbin/dhclient-script being run, not /etc/dhcp/dhclient-script,
[16:59] <hallyn> hm, ahs same thing
[17:00] <hallyn> *has
[17:02] <jjohansen> psivaa: do we have a bug to track this?
[17:03] <psivaa> jjohansen: no i have not open one yet, i could do it in a little while
[17:47] <theazman> Hey all, quick question. On server 12.04, how do I give a domain account local logon rights?
[17:49] <psivaa> jjohansen: sarnold: hallyn: reported bug #1197484 for the above issue.
[17:49] <sarnold> psivaa: thanks
[17:49] <jjohansen> psivaa: okay, thanks
[17:50] <theazman_> If I try to add the user, it says the user already exists
[17:51] <psivaa> thank you for your time :)
[17:59] <theazman_> Anyone?
[18:00] <wxl> theazman_: ?
[18:00] <theazman_> How do I give an account local login rights to a 12.04 server
[18:01] <shollings> erm
[18:01] <shollings> have you googled that?
[18:01] <shollings> it's elementary shit, a quick google result would tell you
[18:04] <theazman> Sorry, some reason my smuxi keeps losing connection to this room. How do I give a domain account login permissions to a 12.04 server without changing the password?
[18:05] <sarnold> shollings: please don't.
[18:05] <shollings> sarnold: try and stop me.
[18:05] <shollings> :)
[18:07] <Tm_T> shollings: please be helpful when help
[18:07] <shollings> oh I am. i'm teaching him how to do things himself.  that's the most helpful thing you can do.
[18:08] <Tm_T> shollings: that's not ubuntu way
[18:08] <shollings> ubuntu doesn't use google?
[18:08] <theazman_> Anyone able to help me?
[18:08] <Tm_T> shollings: we don't tell people to google it, if you don't like to help then just don't
[18:10] <shollings> Please, feel free to help him.
[18:15] <theazman> So I think I fixed my ping so I won't keep getting kicked off, how do I give a user login access to a 12.04 server?
[18:31] <tsimpson> theazman_: unless you lock the account a user can already login, ps your connection is _not_ fixed
[18:32] <theazman_> tsimpson, yea, I noticed my connection isn't fixed, still working on that. The account is a domain account, but not a local account. I cannot su account name, when I try that it just gives me a prompt again
[18:33] <Daviey> zul: How about you seed it, and I promote it? :)
[18:34] <zul> well we dont want it on the cd do we?
[18:34] <Daviey> zul: Yes, but why do you want it in main?
[18:34] <zul> Daviey:  its a dependency for kombu
[18:34] <Daviey> zul: Dependency for kombu...
[18:35] <Daviey> Right, so - can you upload a kombu that depends on it?
[18:35] <zul> Daviey:  already did
[18:35] <Daviey> Oh
[18:35] <tsimpson> theazman_: then I guess you have to configure PAM to allow logins over the domain then, I have no idea how to do that
[18:35] <theazman> tsimpson, I'm back for now
[18:36] <Daviey> zul: ah, so yu did - an hr ago. :)
[18:36] <zul> Daviey:  https://launchpad.net/ubuntu/+source/kombu/2.5.12-0ubuntu1/+build/4767352
[18:36] <Daviey> you*
[18:36] <Daviey> zul: Not yet showing on the mismtach list.
[18:36] <zul> Daviey:  okies
[18:38] <tsimpson> theazman: there's a bunch of PAM modules for different authentication mechanisms, search for "libpam" and see what's available to start
[18:39] <Daviey> zul: done, thanks
[18:39] <zul> Daviey:  thanks
[18:42] <MAbeeTT> hi!, We have recently aquired a clone computer for being used as server (AMD FX8120 8 cores, 16G RAM, 2x1Tb HDD)
[18:44] <MAbeeTT> We want virtualize there, deployment and production tomcat+mysql+tcp-adhoc-protocol, and some future similar scenarios.
[18:44] <MAbeeTT> I would like to full virtualization and containers inside one (or more) of the VMs.
[18:45] <MAbeeTT> Is there some gui-like solution for ubuntu server 12.02 LTS?
[18:45] <MAbeeTT> I have seen openstack, juju, but seems bein some for a "farm
[18:45] <MAbeeTT> "
[18:45] <MAbeeTT> thanks.
[18:47] <sarnold> MAbeeTT: perhaps using virt-manager with kvm is about what you're after
[18:48] <MAbeeTT> could I make virtual nets also? I mean virtuals NICs "plugged"tu a virtual Switch.
[18:48] <sarnold> MAbeeTT: juju is pretty slick, you could use it with lxc containers on the host; the nice part is you could scale that out to running services on VMs with a private openstack cloud or maas setup or something, but it wouldn't easily move to just having a second computer with lxc containers in the future...
[18:49] <sarnold> MAbeeTT: linux provides bridge functionality easily, switches require other tools, last I looked at them I wasn't very impressed with what I saw. I hope they've improved..
[18:49] <MAbeeTT> ok, via bridges, I know them.
[18:52] <MAbeeTT> ok. The "impression" I have about juju is that is for huge developments. We are a small  Research group in a university. Now  we use some old computers (Pentium 4, 256/512M).
[18:53] <MAbeeTT> so, the idea is move, and generate new VMs, but as you can imagine this are not stressed services.
[19:06] <theazman_> Hey, sorry, no idea what is going on with my connection. In any case, I gave a domain account logon rights to a ubuntu server. When I try to ssh in, the session closes right after it authenticates. What could be causing this?
[19:07] <sarnold> theazman_: out of curiosity, what is a 'domain account'?
[19:08] <theazman_> sarnold, an account created on the domain, Windows AD account
[19:08] <theazman_> Sorry, should have said AD account
[19:08] <sarnold> theazman_: you can probably find more detailed reasons for the session closing in the log files, /var/log/auth.log would be my first guess..
[19:08] <theazman_> sarnold, it wasn't showing up in th elogs, I'll look again
[19:08] <sarnold> theazman_: aha :) which pam modules are you using to do the integration?
[19:09] <theazman_> sarnold, possibly krb5.so
[19:10] <theazman_> sarnold, its denying me access to the log
[19:10] <sarnold> theazman_: can you use sudo or su?
[19:10] <theazman_> I'm in as root
[19:11] <theazman_> sarnold, and it shows up blank, and when I do sudo from the admin account, says permission denied
[19:11] <sarnold> theazman_: .. and you're getting permission denied trying to read the file?
[19:11] <mnathani> theazman_: have you been rooted?
[19:11] <theazman_> mnathani, I am under root
[19:11] <theazman_> sarnold, under root, the log shows up empty
[19:11] <sarnold> theazman_: mnathani is thinking your system may have a root kit installed
[19:11] <mnathani> theazman_: I meant is there a virus / malware / rootkit installed on the box
[19:11] <theazman_> ah, I see
[19:11] <theazman_> Let me check
[19:12] <sarnold> theazman_: -crazy-. how about other log files in /var/log ?
[19:12] <mnathani> compare md5 files for common binaries like ls
[19:12] <mnathani> *files = sums
[19:13] <theazman_> sarnold, so auth.log.1 shows up
[19:13] <theazman_> sarnold, but auth.log shows blank
[19:13] <sarnold> theazman_: check filesystem free space.
[19:13] <theazman_> sarnold, the box might need to be rebooted, but that can't be done at this time
[19:13] <theazman_> sarnold, where do I check that
[19:13] <sarnold> theazman_: df
[19:13] <theazman_> mnathani, where do I see the md5 files?
[19:14] <theazman_> sarnold, the most used is 30% which is /dev/sdal
[19:15] <theazman_> sarnold, so the logs that show up, the dates stop on Jun 29
[19:16] <theazman_> sarnold, I think I did an apt-get update and upgrade on that day which might have broken it
[19:17] <theazman_> sarnold, you still there?
[19:18] <sarnold> theazman_: hrm. check the .1 log file then, perhaps it's still being used ...
[19:18] <sarnold> (I'd have expected sshd would close/reopen the log file at log rotate time..)
[19:18] <theazman_> sarnold, so the auth.log.1 stops on june 29
[19:18] <sarnold> oh right. sigh.
[19:19] <theazman_> sarnold, I'm not sure if log rotate is setup
[19:19] <theazman_> checking on that now
[19:20] <theazman_> sarnold, it is setup to rotate weekly
[19:21] <theazman_> sarnold, and syslog is empty, last log in .1 is jun 29 around the same time
[19:22] <sarnold> theazman_: I wonder if /var was full on jun 29?
[19:23] <theazman_> sarnold, how do I check that
[19:23] <theazman_> sarnold, /var doesn't have its own filesystem on here
[19:23] <sarnold> theazman_: looking through those logs from jun 29 is your best bet, but .. not a very good one.
[19:24] <sarnold> theazman_: lets try service rsyslog restart -- just a blind, restart the logging service..
[19:24] <theazman_> sarnold, I had tried that already a bit ago
[19:28] <theazman_> sarnold, that didn't fix it, I am thinking of doing a restart
[19:30] <adam_g> zul, any chance at a peak of https://code.launchpad.net/~gandelman-a/ubuntu/saucy/nova/drop_requests_cap/+merge/172709 + https://code.launchpad.net/~gandelman-a/ubuntu/saucy/nova/pyparsing_min/+merge/172639 ?
[19:32] <zul> adam_g:  done
[19:33] <adam_g> zul, thanks. i had one for kombu requirement but dropped since thats all together already. nice job
[19:33] <zul> adam_g:  thanks
[19:33] <zul> i should have one for ceilometer tomorrow
[19:33] <adam_g> zul, im going to poke dolph and see if he can cut a new keystoneclient release so cinder and ceilometer arent version capped on their requirement
[19:34] <theazman_> sarnold, any other ideas?
[19:34] <zul> adam_g:  cool
[19:34] <adam_g> zul, also, whats the plan for the neutron rename?
[19:35] <zul> adam_g:  when its done upstream i already have a branch ready for quantum quantum client not yet
[19:36] <zul> adam_g: https://code.launchpad.net/~zulcss/neutron/neutron-rename
[19:37] <adam_g> cool
[19:38] <theazman_> sarnold, I think I found the problem as to why it wasn't logging
[19:41] <theazman_> sarnold, apparently I accidently changed the owner of the log to a different user account, I changed it back but it still isn't writing to it
[19:45] <theazman_> sarnold, got the logs working, so now about getting the account to login?
[19:47] <sy_> I have a 3TB drive where I have created a raid1 set with disk missing to be added in later however mdadm is reporting 2.1TB, any ideas?
[19:48] <sy_> There is a single partition on the drive containing the full disk
[19:54] <theazman_> Anyone able to help me figure out why a user can't ssh into the server?
[19:56] <sarnold> theazman_: hey, glad to hear you've got the logging sorted out. :) are there any instructive errors in hte logs, now that they work?
[19:57] <theazman_> sarnold, not really
[19:58] <theazman_> sarnold, just session opened and then session closed at the same second
[19:58] <sarnold> theazman_: how about the user's shell? shell startup files?
[19:59] <theazman_> sarnold, adding those now
[20:00] <sarnold> theazman_: off to lunch, good luck :)
[20:00] <theazman_> sarnold, thanks for your help
[20:28] <theazman_> Anyone able to tell me why an ssh session closes as soon as I authenticate? There is nothing descriptive in the logs.
[20:29] <anepanaliptos> theazman_: how about scp?
[20:29] <anepanaliptos> maybe there is something in your bash/login script?
[20:30] <theazman_> Well, it also doesn't let me su username when I try to su username from root
[20:30] <theazman_> anepanaliptos, ^
[20:30] <anepanaliptos> (yeah it seems like it's just me and you here)
[20:31] <anepanaliptos> got another linux machine someplace?
[20:31] <anepanaliptos> see if you can scp a file into your home dir, so do..
[20:31] <theazman_> I have a bunch, its 12.04 server
[20:31] <anepanaliptos> scp file.ext username@host.ip.whatever:~
[20:32] <theazman_> Do I do that from terminal?
[20:32] <anepanaliptos> yeap
[20:32] <theazman_> What does that do?
[20:33] <theazman_> I'm also guessing I need to find a valid file?
[20:33] <anepanaliptos> it copies a file of your choice from the computer you're sitting at to the home directory of the other host
[20:33] <anepanaliptos> yeah.
[20:33] <anepanaliptos> just cd into your desktop and send something from there. do you use terminal often? did you know you can use tab for autocomplete?
[20:34] <theazman_> What account do I want to try that from, root?
[20:34] <theazman_> Yea, I know about autocomplete
[20:35] <theazman_> But I'm not by another linux machine atm
[20:35] <anepanaliptos> doesnt matter. just see if you can send a file itno that persons home dir
[20:35] <theazman_> k, sec
[20:35] <anepanaliptos> if that's the case, there is something wrong with bash
[20:36] <theazman_> Um, that didn't error
[20:36] <theazman_> let me check the dir
[20:36] <theazman_> there is nothing in the home directory of the user
[20:37] <theazman_> What makes you think it is the bash?
[20:38] <theazman_> anepanaliptos, not sure what's going on here.
[20:45] <theazman_> anepanaliptos: You still here?
[20:51] <tdn> Will software raid1 give me a performance increase in reads?
[20:58] <theazman_> Anyone able to help me figure out why when I try to ssh as users except the admin the session closes instantly after authenticating. There's nothing descriptive in the logs... Thanks
[21:00] <anepanaliptos> theazman_: sorry afk
[21:00] <anepanaliptos> hang on.
[21:00] <anepanaliptos> ok
[21:00] <anepanaliptos> so you sent it to userwithproblem@hostwithproblem:~
[21:00] <anepanaliptos> and then you checked the home dir and it wasnt ehre?
[21:01] <anepanaliptos> i think its bash because it's a "safty feature" from ssh that there is a shell or _Somthing_ running when you log in
[21:02] <anepanaliptos> if it doesnt start, or your bash exits abnoramally, that kills teh ssh session
[21:02] <raub> I know you can do that by limiting which commands you can run when you ssh
[21:03] <raub> ex: in authorized_keys
[21:03] <raub> I also think you could setup the shell you are allowed to run
[21:03] <raub> say, to /bin/fals
[21:03] <raub> false
[21:03] <theazman_> YEa, anepanaliptos, it wasn't in the home dir
[21:04] <theazman_> raub, it doesn't do local logins either
[21:04] <theazman_> anepanaliptos, ^
[21:08] <theazman_> Everyone disappear?
[21:18] <theazman_> Anyone able to help me figure out why when I try to ssh as users except the admin the session closes instantly after authenticating. There's nothing descriptive in the logs... Thanks
[21:37] <theazman_> Anyone able to help me figure out why when I try to ssh as users except the admin the session closes instantly after authenticating. There's nothing descriptive in the logs... Thanks