[01:50] <NCommander> When using MAAS on a machine with multiple NICs, how can I get both of them to show up in the webUI?
[01:50] <NCommander> I can only see eth1 and "" (which appears to be lo0)
[01:50] <sazawal>  The minimode plugin in exaile 0.3.2.2 in Ubuntu 12.10 does not remember its position. Help!
[04:55] <stiv2k> anyone here
[04:55] <stiv2k> so now that ubuntu has dropped non-PAE support, what do non-PAE people have to do?
[05:00] <SinZ> Non-PAE?
[05:19] <Gallomimia> hi guys. how's the newest release? I'm holding off on updating for a few more days...
[05:34] <stiv2k> SinZ, yeah
[05:34] <stiv2k> non PAE cpu
[05:50] <SpamapS> stiv2k: You have 4.5 more years with 12.04
[05:54] <stiv2k> SpamapS , my friend already updated to 12.10
[05:54] <stiv2k> using a non PAE cpu
[05:54] <SpamapS> stiv2k: IIRC, do-release-upgrade was supposed to stop that
[05:54] <stiv2k> SpamapS lol, apparently it didnt
[05:54] <SpamapS> stiv2k: if he did it with apt-get dist-upgrade .. not much we can do if he failed to read the release notes.
[05:55] <SpamapS> stiv2k: I suppose he didn't backup the system first either, right?
[05:55] <greppy> hrm
[05:55] <greppy> doh, wrong window
[05:58] <stiv2k> SpamapS, i cant remember which way he did it... but why would ubuntu include more than one way to upgrade?
[06:01] <SpamapS> stiv2k: there is only one supported way
[06:02] <stiv2k> that wasnt my question
[06:02] <stiv2k> if theres only one supported way, why does any other way even exist?
[06:02] <stiv2k> ya know?
[06:02] <stiv2k> why would you give the possibility of another method if you arent going to support it
[06:03] <stiv2k> or maybe giant red boldface text that says USE THE OTHER COMMAND INSTEAD when the user runs it
[06:03] <SpamapS> stiv2k: because do-release-upgrade depends on the lower level dist-upgrade method.
[06:04] <stiv2k> oh
[06:04] <stiv2k> in any case, im pretty sure i sent him this link to follow https://help.ubuntu.com/community/QuantalUpgrades
[06:04] <stiv2k> which says to run do-release-upgrade
[06:04] <SpamapS> Yeah so if thats not preventing non-PAE's from upgrading then I'd argue thats a bug
[06:05] <stiv2k> yes
[06:06] <stiv2k> i just looked at his bash history
[06:06] <stiv2k> he used do-release-upgrade
[06:06] <SpamapS> stiv2k: I just happen to be in the same building with the author of do-release-upgrade .. I'll ask him when I see him.
[06:06] <stiv2k> SpamapS, do you reckon anyone might make a ppa for non-PAE kernels/
[06:07] <stiv2k> i cant imagine it would be that hard to make
[06:07] <SpamapS> stiv2k: I'm not sure. IIRC, it was dropped because its a significantly difficult thing to support
[06:07] <stiv2k> SpamapS, what building is that? you work at ubuntu corporate office or something? :P
[06:07] <stiv2k> you mean its harder than just setting some kernel option?
[06:08] <SpamapS> stiv2k: no, Bella Center in Copenhagen. We are at a pre-UDS meeting.
[06:08] <SpamapS> stiv2k: yes, the patches that Ubuntu carries don't work with some of the non-PAE kernel bits IIRC
[06:08] <stiv2k> oh
[06:09] <stiv2k> is it possible for him to just skip the kernel updates
[06:09] <stiv2k> right now his machine is running 3.2.0-32-generic #51-Ubuntu SMP
[06:11] <SpamapS> stiv2k: yeah it should work for most things
[06:11] <stiv2k> how to skip kernel updates
[06:59] <koolhead17> philballew, hi
[07:22] <philballew> koolhead17, sorry for delay
[07:22] <philballew> got a phone call from the rents
[07:56] <Sprocks> anyone here set up their own IRC server for ubuntu?
[10:44] <radish> hi, mysql (5.5.24-0ubuntu0.12.04.1) behaves odd. If I create a new database and user, login with the user afterwards, show privileges doesn't show the database I gave privileges for. Also the user can login without providing a password allthough I did specify one
[10:58] <acidflash> hello all
[10:59] <acidflash> i checked the file /sys/block/sde/queue/scheduler and inside it was the following -> noop deadline [cfq]
[10:59] <acidflash> this means that cfg is the current scheduler?
[13:16] <Psi-Jack> What shell interpreter does upstart actually use?
[13:25] <Psi-Jack> Looks like just sh.
[13:25] <Psi-Jack> dash, or compatible. :)
[13:26] <MatBoy> guys, I cannot resolve domainnames, only their subdomains... what can be wrong here ?
[14:03] <dorftrottel_> not the other way around?
[14:39] <Aison> hello
[14:39] <Aison> i've got two dns servers and I specified them in network/interfaces with
[14:39] <Aison> dns-nameservers 2001:1620:f00:81d3::1 2001:1620:f00:81d3::2
[14:40] <Aison> why is there only the first one later in resolv.conf? nameserver 2001:1620:f00:81d3::1
[14:58] <zul> soren: ping
[15:06] <zul> hallyn: i think we should have libvirt 1.0 in raring rather than 0.10.2
[15:07] <hallyn> zul: offhand i think i agree.  well, wahtever the newest release will be around january
[15:07] <hallyn> but, i've not used anything newer than what we have in q
[15:07] <zul> hallyn:  right i though 1.0 was suppose to be out at least in november
[15:08] <hallyn> guess we should look at the timeline before next week.  feh
[15:08] <hallyn> (feh - bc i'm bearing down for some triaging :)
[15:09] <zul> hallyn: meh
[15:46] <hallyn> stgraber: hm, lucid contaienrs aren't working for me on quantal hosts
[15:47] <stgraber> mountall: Event failed
[15:47] <stgraber> mount: mount point /dev/shm is a symbolic link to nowhere
[15:47] <stgraber> mountall: mount /dev/shm [51] terminated with status 32
[15:47] <stgraber> mountall: Filesystem could not be mounted: /dev/shm
[15:47] <stgraber> hallyn: ^ ?
[15:47] <hallyn> oh right.  someone else had reported that last week
[15:48] <hallyn> gr
[15:48] <hallyn> oh.  is this bc of devtmpfs?
[15:49] <hallyn> yes it is
[15:49] <stgraber> fun
[15:50] <hallyn> i don't even know what we want to do about this.  i'll just check for (and file) a bug for now i guess.
[15:50] <hallyn> on the brigh side, since lucid containers require lxcguest from ppa, they're funky anyway
[15:51] <hallyn> stgraber: was it you who mentioned that last weekthen?
[15:51] <stgraber> hallyn: this really needs fixing before the SRU lands though
[15:52] <stgraber> hallyn: nope, wasn't me. Just heard about it now when you mentioned it (and checked against my newly rebuilt containers)
[15:52] <hallyn> stgraber: good point.
[15:52] <hallyn> stgraber: any thoughts on the right fix?
[15:52] <hallyn> maybe the right fix is for my lxcguest to mangle /dev/shm into a symlink
[15:52] <stgraber> well, /dev/shm is a symlink ;)
[15:52] <stgraber> mount: mount point /dev/shm is a symbolic link to nowhere
[15:53] <hallyn> sorry, a dir i meant :)
[17:11] <Aison> i've got two dns servers and I specified them in network/interfaces with
[17:11] <Aison> dns-nameservers 2001:1620:f00:81d3::1 2001:1620:f00:81d3::2
[17:11] <Aison> why is there only the first one later in resolv.conf? nameserver 2001:1620:f00:81d3::1
[17:11] <Aison> with ipv4 it works
[17:12] <stgraber> Aison: how many "nameserver" entries do you get in /etc/resolv.conf?
[17:12] <Aison> three, two ipv4 and the first of ipv6
[17:12] <stgraber> right, so that's why
[17:12] <stgraber> the maximum number of entries in /etc/resolv.conf is 3
[17:13] <Aison> ok, why?
[17:13] <stgraber> Aison: because someone once set MAXNS to 3 in resolv.h apparently, so the libc resolver can only understand the first 3 nameserver entries
[17:13] <stgraber> any extra one would just be ignored
[17:15] <Aison> quite funny
[18:32] <fred__> I'm trying MAAS + juju on quantal. As there is no mysql charm for quantal, I cannot deploy in using juju, should I use precise's version?
[19:12] <genii-around> Would anyone know perhaps offhand the minimum amount of free disk space needed for Lucid->Precise upgrade with do-release-upgrade? The machine is pretty tight right now: /dev/sda1 9.0G 7.6G 882M 90% /    ( although I could bindmount something like /var/cache/apt to the RAID1 which has lots of free room )
[19:16] <RoyK> genii-around: start with an apt-cache clean
[19:17] <RoyK> genii-around: it's quite impossible to determine the amount needed for a full upgrade, though
[19:17] <genii-around> RoyK: That df  is already after an apt-cache clean, actually...
[19:17] <RoyK> genii-around: that's why I always use LVM even for the root, to allow it to grow if needed
[19:18] <RoyK> perhaps check /var/log
[19:18] <RoyK> or do you have databases running under the root partition?
[19:19] <RoyK> 7,6GiB used for the root on a server is quite a lot
[19:19] <genii-around> RoyK: Yes, this machine is primarily for the surveillence so it is basically LAMP plus Zoneminder. The camera footage all goes on the RAID1 right now
[19:20] <RoyK> and you have all databases on the root partition=
[19:20] <RoyK> ?
[19:20] <genii-around> Yes
[19:21] <RoyK> sub-optimal to be diplomatic
[19:21] <RoyK> the root should be used for root stuff, not variable data
[19:22] <RoyK> databases are very variable data indeed
[19:23] <genii-around> I basically wanted to be keep the main disk size down to below 16G so I could dd it off onto a USB stick on occasion
[19:24] <genii-around> Guess I could add another drive, I have a bunch of old ones around here someplace
[19:24] <mdeslaur> SpamapS: have you ever gotten a mysql test suite failure in main.trigger-compat ?
[19:25] <fred__> genii-around: if your raid1 storage is large enough, consider moving your db on it
[19:25] <RoyK> genii-around: you don't want a database on your root partition. period.
[19:25] <RoyK> genii-around: and you don't want a database on something that's not redundant
[19:26] <genii-around> Yes, moving the /var to some subdir on the RAID1 seems to be prudent
[19:27] <RoyK> genii-around: just move /var to the raidset
[19:27] <RoyK> genii-around: next time, mirror the root as well, so that when that drive dies, the server doesn't ;)
[19:31]  * genii-around goes and tinkers
[19:36] <hallyn> not again
[19:37] <hallyn> SpamapS: /win 15
[19:38] <hallyn> well hmm
[19:38] <hallyn> nm
[19:49] <subman> I'm trying to setup an rsync cronjob using ssh with keys but I seem to be having a problem.  I've setup the keys and can log into the remote storage machine with no problem.  However I tried to duplicate the cronjob (I assume they are run as user 'root') with sudo and it still asks me for a password.  Did I need to setup the keys with the user 'root'?
[19:50] <sarnold> sudo will ask you for a password unless you configure it not to
[19:50] <sarnold> but ..
[19:50] <sarnold> you may just wish to run the command via whichever user account's crontab is appropriate?
[19:51] <subman> Yes, sudo did ask me for a password, but I am concerned with the ssh also asking me for a password.  I have already logged into that user account on the remote machine via ssh with no password required now that he keys are setup.
[19:52] <fred__> subman: you need an ssh-agent running
[19:53] <sarnold> subman: are the keys password protected? do you have an ssh-agent running to supply them with the passphrase if they are password protected?
[19:53] <subman> sarnold, No, keys have no passphrase
[19:54] <sarnold> okay
[19:54] <subman> Maybe I'm asking this question the wrong way.
[19:54] <sarnold> maybe a bit too abstract? :)
[19:54] <sarnold> which user account on which host is running which commands? :)
[19:59] <subman> Maybe I should ask this question in a forum to explain it in complete detail exactly what I'm trying to do here.
[20:01] <sarnold> subman: or maybe paste the command you've put into which crontab...
[20:01] <subman> * 0 * * * rsync -avv --delete /home kitty@192.168.2.3:/home/kitty >> /var/log/rsyncbackup.log
[20:02] <sarnold> okay :) which crontab is this in?
[20:06] <subman> I created that with crontab -e as a user on the machine that I want to backup the home directories from.
[20:08] <sarnold> subman: and does that user account have read access to the entirety of /home ?
[20:09] <subman> sarnold, Ah, I assume that the cronjobs run as root.  They run as the user that created them?
[20:09] <subman> No, this user does not have read rights to the whole /home directory.  I should change that.
[20:09] <sarnold> subman: yeah, each user has their own crontab; that's what you get with crontab -e :)
[20:09] <subman> Yeah, I'm new at this.
[20:10] <sarnold> subman: if you want to run it as root, it might be easier to add that line to the /etc/crontab file, or simplify it a little and put it into /etc/cron.daily/ or similar
[20:10] <sarnold> subman: but make sure the keys are available to root; maybe you'll need to move them to /root/.ssh/
[20:11] <sarnold> (root _does_ have a 'user crontab file', via crontab -e, but that's .. odd.)
[20:11] <subman> sarnold, I think I'm understanding it all better now.
[20:12] <sarnold> subman: cron -is- a bit cumbersome :) but it's grown to be that way by handling a lot of other needs that are hard to handle with other tools...
[20:13] <subman> I would normally use a GUI tool to do backups, but this is my first install with a non GUI server install.  Time to learn the command line way to do things!
[20:14] <sarnold> and besides, the gui will probably end up calling rsync in the end anyhow
[20:14] <sarnold> I've seen plenty of those "products" before...
[20:14] <subman> Exactly!
[20:14] <subman> I thought it about time to roll up my sleeves and dive right in.
[20:16] <subman> Now, if I change the permissions of /home to be 'read' for the user doing the rsync, will that permission carry over to any newly created files not owned by that user?
[20:17] <sarnold> subman: depending upon the site reliability, you may also like to add --partial to the rsync command line -- that will allow it to re-use the portion of a file that may already exist on the remote end. (ideally this would just be magically correct, but it isn't..)
[20:17] <sarnold> subman: no; getting read access propagated down a directory hierarchy correctly is pretty complicated
[20:18] <sarnold> subman: your choices are (a) run a command that forces all the permissions to be "correct" before running it (b) some complicated "BSD groups" behavior (awesome for when you need it, but definitely a new complication) (c) just running the rsync as root.
[20:19] <subman> I currently run the rsync as root via sudo but his won't work with the user account level cronjob.
[20:19] <subman> I assume that I must then create the cronjob as root?
[20:21] <sarnold> subman: that's what I would do
[20:21] <subman> sarnold, Great, doing that just now.  /etc/crontab.
[20:22] <subman> and move the keys....
[20:29] <subman> hmmm, keys still do not work as root
[20:31] <subman> Got it!
[20:31] <sarnold> subman: hrm. is /root/.ssh group or world readable?
[20:32] <subman> It worked.  I had to copy over both files in the .ssh directory
[20:32] <subman> Sorry, both directories under the .ssh directory of the user
[20:32] <sarnold> directories under .ssh?
[20:33] <subman> keys are held in two directories under the .ssh directory.  id_rsa and id_rsa.pub
[20:34] <subman> ls
[20:34] <subman> (oops)
[20:35] <sarnold> subman: those ought to just be files
[20:35] <subman> Yes they are, sorry for the terminology
[20:35] <subman> Too much learning today!
[20:35] <sarnold> okay ;)
[20:36] <sarnold> hehe
[20:36] <subman> So many new things, but a ton of fun here.
[20:36] <subman> sarnold, Thanks for your help here, greatly appreciated!
[20:36] <sarnold> subman: sure, I'm glad you're still having fun :)
[20:37] <subman> Oh yeah, no problem there.  I'm sure I'll make some huge errors but these are not production machines.
[20:37] <subman> Just learning machines
[20:43] <soren> zul: 'sup?
[20:44] <zul> soren: should i be bringing gloves next week :)
[20:45] <soren> zul: Of course. Just like any other UDS.
[20:46] <soren> zul: Oh, *gloves*? Uh, no. Why would you?
[20:46] <zul> just figure out stuff out
[21:29] <sazawal> The minimode plugin in exaile 0.3.2.2 cannot remember its position and moves back to top left corner. I have just upgraded to Ubuntu 12.10 and the problem started. There is something I can do in the plugin script at /usr/share/exaile/plugins/minimode. Please help
[21:31] <sarnold> hrm, seems I saw someone yesterday complaining about that. check launchpad, he or she may have filed a bug?
[21:35] <sazawal> sarnold, I guess that was me :D
[21:35] <sarnold> sazawal: oh :)
[21:36] <sazawal> Well, I should file the bug then.
[21:36] <sazawal> sarnold, do you use exaile?
[21:37] <sarnold> sazawal: no, I hadn't heard of it, the funny name stuck out though.. :)
[21:37] <sazawal> sarnold, haha, no problem. Its a media player
[21:38] <sarnold> :)
[22:16] <Aison> is it possible that isc-dhcp-server of ubuntu 12.10 is broken?