=== Lcawte is now known as Lcawte|Away [00:14] !ask [00:14] Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [00:14] !patience [00:14] Don't feel ignored and repeat your question quickly; if nobody knows your answer, nobody will answer you. While you wait, try searching https://help.ubuntu.com or http://ubuntuforums.org or http://askubuntu.com/ [00:15] !help [00:15] Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [01:19] Is there a virtualization program that I can install on the ubuntu will will act like vShpere and allow me to create virtual machines onto of ubuntu accessible remotely? [01:22] kvm [01:23] I think it even comes preinstalled [01:24] zul: added some versioning info to the test runs, should help with verification later: https://jenkins.qa.ubuntu.com/view/Openstack%20Testing/view/Openstack%20Precise/view/Overview/job/precise-openstack-essex-test/561/console [01:32] Daviey: the ipxe merge is done. [01:32] we're not merging from debian. i took the fix we needed, and (per discussions here ) we're not merging from upstream eitehr [01:32] as for libcgroup, the discussion is not whether to merge, but wehther to drop it from universe [01:33] we were chatting with jbernard about perhaps having some cgroup management utilities like libcgroup's in cgroup-bin, [01:33] but the daemons which are the core of libcgroup are not really salvagable === dendro-afk is now known as dendrobates [04:13] is there any reason my ulimit won't pick up changes in /etc/security/limits.conf? [04:14] Did you log out and back in? [04:16] yeah, i just figured that out === glebihan__ is now known as glebihan [06:39] Hello all, I have recently seen a change in cron which states that @reboot is now under some sort of limitation due to start up sequence (ubuntu-server 12.04) [06:40] does anyone know of a workaround for @reboot in cron? I really need this script to launch @reboot _after_ everything is initiated [06:53] acidflash: use rc.local? [06:53] Instead of @reboot I mean [06:54] twb: It doesnt start it at the appropriate time. [06:54] twb: rc.local boots during boot, not at the end of boot [06:55] acidflash: rc.local runs at the *end* of the sysvinit boot sequence. [06:55] When do you want @reboot to run? [06:56] if you want to run a script after "everything", because it's upstart you will need to create an upstart job that lists each thing in everything individually, by event name [06:58] twb: rc.local doesnt run at the end of sysinit, ie: it runs BEFORE security limits are assigned [06:58] twb: if I wanted a program to have greater than 1024 file descriptors, it needs to launch AFTER the security limits are assigned [06:59] @reboot did that great, all I did was add a delay of 3 minutes, [06:59] didnt work in rc.local, but maybe a delay increaase might? [07:38] security limits as in /etc/security/limits.conf ? [07:40] twb: yes [07:41] twb: one script worked fine in rc.local as replacement, other diddnt, debugging why now [07:41] the first one doesnt need the < 1024 FD [07:41] I should think that would be enforced from the get-go [07:41] second one does [07:41] Except probably root user is not subject to those limits or something [07:42] twb: when running process as root, if its in rc.local you wont get the 1M FD's [07:42] only 1024 [07:42] twb: but when you do crontab, and EVERYTHING loads, you get 1M FD's [07:42] Uh... I don't think that's limits.conf [07:43] I thought you were trying to restrict the program, not give it increased limits [07:43] limits.conf doesn't, by default, include any limits [07:44] twb: right, it doesnt, the OS does, 1024 per process [07:44] twb: by default [07:45] twb: limits.conf ALLOWS you to overcome that [07:47] Hum. [07:47] I don't suppose you can just fix your flipping program to be less stupid [07:47] 1k fds should be enough for any proc [07:48] I wish it was, we have some horrible 'enterprise' grade software that does things like connect to a bunch of grid servers, leaving 5 fds open per server on a permanent basis >.> [07:48] twb: Not when you are a server that needs to server 5k+ users, each one account for serveral TCP connections [07:49] twb: anyhow, just increasing delay in start-up script and adding to rc.local DID work === zyga is now known as zyga-afk [07:52] Adding a delay won't help you because the amount of delay will change wrt. hardware and other constraints, e.g. if DNS is slow during one bootup your 3min might not be enough, etc. [07:53] Which is the whole reason for these stupid event-based start rules [07:53] I should say: a hard-coded delay is a workaround rather than a solution [07:53] twb: yes, a work-around, which might fail is things are delayed (correct) [07:54] I had to do one in lucid to get NFS working -- because there was a cyclic dependency in upstart (stupid upstart!) I replaced one of the jobs with a simple script that ignores all deps and simply tells mountall to try to mount NFS every tenth of a second :-( [07:56] twb: what you could do, is first sleep for a bit, then check if its mounted (through PID) if not, then mount [07:56] twb: it wont rape your resources, [07:58] That's called a spinlock [07:58] pkill -USR1 mountall is not a big deal resource-wise [08:00] quick q, WITHOUT making a script, can I grep something with spaces? double and single quotes are both not working [08:00] new dates are like "June 7" [08:00] in ls -la [08:00] I need to search and sort by date [08:03] acidflash: #bash [08:03] In particular, /msg greybot ls [08:05] Perhaps you want the stat(1) command. [08:11] Daviey, do you have the URL for the server related merges to hand? want to see how we are doing.... [08:26] hi, after an update to 12.04 from 11.10 I am missing some scripts in /etc/update-motd.d/ can I just copy them from a clean 12.04 install or should I be adding them in another way? [08:27] jamespage: http://people.canonical.com/~davewalker/delta.html .. based on ubuntu server package set, and the packages we are subscriber for. [08:27] I need to make it respect the sync blacklist i think.. for the openstack stuff. [08:27] feisar: try aptitude reinstall [08:28] Actually they'll be conffiles, so that won't work [08:28] feisar: try purging and then installing the package in question [08:29] feisar: note that obviously this will lose any other site-local customizations you've made to that package's conffiles. Suggest etckeeper. [08:29] I'm assuming this has happened because you've somehow deleted the conffiles and as a result that's considered a deliberate configuration change so dpkg will preserve it across upgrades / reinstalls. [08:38] Daviey, I wanted to check on timing of the updated I need to make to the tomcat-server seed for quantal [08:38] Daviey: should I get tomcat7 into main first and then update the seed? [08:39] (only ever done one seed update...) [08:41] jamespage: either way is valid.. note that main packages are slightly more compressed than universe. [08:41] cruft from binary deb removal [08:42] morning o/ [08:52] twb: thanks, you don't know what package these confic files are linked to do you? [08:55] Ask dpkg -S / dlocate === ejat- is now known as ejat === allenap` is now known as allenap === AlanChicken is now known as alanbell === alanbell is now known as AlanBell [09:24] Am I correct to assume that Dovecot is de default imap server for ubuntu server? [09:26] It is the one the Ubuntu Server Guide documents. === Lcawte|Away is now known as Lcawte === zyga-afk is now known as zyga === railsraider_ is now known as railsraider [10:15] rbasak, do you have time to work on the apache2 merge still? just been reviewing and it needs a bit more work [10:16] jamespage: as of today, yes [10:17] rbasak, marvellous - PM [10:17] jamespage: I'd like some guidance here though. There are a multitude of ways to do the merge. I used bzr because merges reported a conflict. Is there a better way? [10:18] rbasak, sure [10:18] so I would recommend one of two ways to merge: [10:18] instructions in https://merges.ubuntu.com/a/apache2/REPORT ok? [10:19] 1) bzr branch ubuntu:apache2; cd apache2; bzr merge lp:debian/sid/apache2; quilt push -a; dch -i (and update the changelog) [10:19] or [10:19] 2) grab-merge apache2 (which uses merges.ubuntu.com) [10:19] update the changelog with the delta ... [10:20] I did more or less the first method, but the result looked very difficult to review. Not sure about the issue zul reported, I'll look at that now. [10:20] rbasak, you can still use merges.ubuntu.com even with a conflict - but you will have to resolve it either way [10:20] rbasak, the issue zul reported was a missing 'quilt push -a && bzr add .pc' [10:21] rbasak, but the changelog is also missing the last ubuntu entry as well - so it may be that the branch is not up-to-date in bzr [10:22] I just tried a bzr branch from my merge proposal and bzr bd -S works on that [10:22] ah so the changelog might be me [10:23] when I resolve a changelog conflict, I'm supposed to keep the last ubuntu entry? [10:23] rbasak, for a merge the changelog should be the combined history from Debian and Ubuntu [10:24] OK. I had been taking the debian changelog with only one new ubuntu entry at the top. I'll stop doing that then :) [10:24] rbasak, actually I'm not sure what zul did either - bzr bd -S works for me as well [10:24] I didn't like my merge proposal though. I used bzr merge and then fixed the conflict in debian/control, but I have no idea how anyone is supposed to be able to review that because it's a mess [10:25] rbasak, do you want to take another run at it? [10:25] please [10:25] Shall I use grab-merge this time? :) [10:25] it should target quantal as well BTW (:-)) [10:25] oops :-/ [10:25] rbasak, yes - if you like! [10:25] just post the debdiff to the bug report [10:25] or direct to me if you like [10:26] "It looks like this package is maintained in revision control"..."You almost certainly don't want to continue without investigating." [10:26] Is that normal? [10:28] rbasak, Vcs-Bzr: http://code.launchpad.net/ubuntu/+source/apache2 causes that [10:28] its completely superflous IMHO [10:28] thats the implicit branch for Ubuntu [10:28] OK [10:29] if I was doing the merge I would probably drop the VCS related delta - its not really required [10:30] but don't worry about that for the time being [10:31] OK will do [10:31] That's where the conflict is anyway [10:31] So debian has Vcs-Browser: http://anonscm.debian.org/gitweb/?p=pkg-apache/apache2.git [10:31] Vcs-Git: git://git.debian.org/git/pkg-apache/apache2.git [10:32] I want to change these to XSBC-Original-Vcs-Browser etc. but not add a Vcs-Bzr? [10:35] rbasak, I'd leave the VCS field from debian intact and drop the VCS-Bzr [10:35] jamespage: OK, so no XSBC-Original- as there was previously? [10:35] rbasak, spot on [10:36] well that conflict's easy to resolve then, thanks :) [10:36] rbasak, once you have something you are happy with use the merge-debuild script to generate a source package - and then you can debdiff between debian->ubuntu and ubuntu->ubuntu to check your changes [11:05] jamespage: ok, done and tested. What do I submit a debdiff against? The debian package? [11:05] rbasak, I'd attach both to the bug report [11:06] jamespage: both? Which is the other? [11:06] rbasak: debian->ubuntu debdiff and ubuntu->ubuntu debdiff [11:06] OK [11:06] make sense? [11:08] The former is on the Debian PTS page, surely? [11:08] Oh, before you upload it === dendrobates is now known as dendro-afk [11:11] jamespage: done. I'm not sure how to use the ubuntu->ubuntu debdiff for anything useful, but I did compare the old ubuntu debdiff against the new one to verify that I'm happy with it. [11:13] jamespage: so that's quantal. I'm not sure about how to SRU this though, since the fix to the apache2 package might be a bit invasive. I commented on the options in https://bugs.launchpad.net/debian/+source/apache2/+bug/988819/comments/6 [11:13] Launchpad bug 988819 in modsecurity-apache "wrong path to libxml2.so.2 in mod_security" [High,Fix released] [11:27] rbasak, I think that the backported patch is probably appropriate for the SRU IMHO [11:28] the other options feel a bit hacky [11:29] just reviewing you debdiff now [11:34] rbasak, uploaded - I made one minor tweak to the changed to say we dropped the Vcs-Bzr delta but aside from that all good [11:34] rbasak, changed->changelog [11:44] rbasak, OK if I assign you the SRU tasks for that bug? That way it appear in the right part of my SRU report :-) [11:49] hi i've sent here from xubuntu channel, the OS I use, in search for an app that changes the proxy settings systemwide. IN synaptic there is a ubuntu-system-service package I have installed for said end, but I dont see any new icon, nor alt+f2 works with that name, tips? [11:50] i've been sent* [11:50] martinphone: what kind of proxy? http? [11:50] twb, for every connection: wget, torrents, radio... [11:50] 127.0.0.1:8118 [11:51] Um, diffierent protocols have different proxis. [11:51] *proxies [11:51] this proxy worked well with ubuntu, because they have a systemwide proxy settings, something not present in xubuntu, thats why I installed that package I mentioned [11:51] If you edit something like /etc/environment, and add something like http_proxy=http://127.0.0.1:8118/, new processes should inherit that environment. [11:52] However almost all GUI apps will ignore that environment because they are stupid GUI programs [11:52] so you are saying that the only way of doing so is from the shell... [11:52] I need a fast an easy way to change systemwide proxy settings (college) [11:52] This is a server channel. We don't do GUIs as a rule. [11:52] The approach I suggested will certainly work for wget and curl and so on [11:53] I believe I have edited that file already.... [11:53] Note that you will obvioulsy need to log out and back in for a change to environment variables to take effect. [11:53] good morning [11:53] twb, im so noob I didnt know that [11:54] For GNOME apps there is dconf/gconf stuff, but AFAIK XFCE doesn't have a centralized mechanism to set or enforce proxy settings. And obviously, non-XFCE apps like firefox and thunderbird require you to set it somewhere entirely different, lke /usr/share/firefox/prefs.js [11:54] twb, can you confirm that if I do that, ALL connections from my machine will be routed through that proxy, and that if I re edit back to default and restart, proxy settings will change accordingly? [11:54] (Which isn't a conffile, so a security update will eat your changes to that file) [11:54] no problems with FF [11:54] and yes, XFCE doesnt have a centralized... [11:55] martinphone: I cannot confirm that, because it is wrong. [11:55] martinphone: only apps that honour $http_proxy will be affected [11:55] mmrmmr [11:55] Most CLI apps (inc. wget, curl, w3m) honour it; most GUI apps do not (inc. ff) [11:56] IIRC opera does, epiphany doesn't, midori usually does but some versions don't. [11:56] twb, chromium doesnt, does it? [11:56] If you need to enforce ALL traffic, you might look at a transparent proxy [11:56] jamespage: thanks! And happy to have the SRU tasks. [11:56] I cannot comment on chromium. [11:57] yes ALL traffic, its for a travel to china [11:57] martinphone: then you are better off setting up a VPN link [11:57] trip* [11:57] rbasak, are you OK on the SRU process? [11:57] jamespage: yes thanks. I've done a few before. [11:57] rbasak, marvellous [11:57] martinphone: https://help.ubuntu.com/12.04/serverguide/vpn.html [11:58] rbasak, give me a ping when you have branches proposed for merge and the SRU information documented and I'll review [11:58] will do [11:59] rbasak, just forwarded you an email re changes to SRU process as well - just in case you have not see it [11:59] ok thanks === matsubara is now known as matsubara-lunch [12:00] twb, Is VPN like torbox? [12:02] No. === zyga is now known as zyga-afk === garyposter is now known as gary_poster === dendro-afk is now known as dendrobates === zyga-afk is now known as zyga [13:32] does anyone know of a wrapper or native implementation of paraller whole file hierarchy sha1sum/md5sum utility? (i've got hundredths of thousands of files) [13:38] reisi: use xargs with the -n and -P options? [13:38] reisi: (and find) [13:41] rbasak: nice, not optimal but still great; good thing that i didn't start writing one myself :) [13:41] why not optimal? [13:50] i thought i had a small-file optimization in mind but this saturates all IO nicely even with the small files; works great, thanks [13:50] actually no, this does not work properly if i just redirect stdout to a file [13:50] there are already some splitted lines after 90k files [13:52] xargs would need to provide some kind of per-child line buffering here for this to work [13:57] apparently there's gnu's "paraller" [13:58] ... which is apparently a perl script [14:00] http://www.gnu.org/software/parallel/ [14:00] hello. Yesterday I have successfully tested adding a second mariaDB instance on ubuntu-server and wrote down my steps: second-mariadb-instance.txt ... http://paste.debian.net/173497/ [14:00] I would be glad if you could have a look and give me some feedback in case I am missing some things ... thank you in advance [14:09] reisi: GNU parallel has now been packaged and will be in 12.10. [14:09] (in universe) [14:10] rbasak: great; installation is dead simple though but it always feels a bit strange when you cannot find it with apt-get [14:22] jamespage: I've added SRU debdiffs to bug 988819 and added the SRU information [14:22] Launchpad bug 988819 in modsecurity-apache "wrong path to libxml2.so.2 in mod_security" [High,Fix released] https://launchpad.net/bugs/988819 [14:22] rbasak, marvellous - let me just clear this samba SRU I'm working on and I'll take a look [14:37] rbasak, fixes look good but the changelog entries need more information [14:38] OK. I'm never quite sure what to put in a changelog. I try and avoid writing an essay in the hope that referring to the bug will do. [14:39] rbasak, OK so the trick for SRU changelogs is that someone should be able to understand what its fixing without referring to the bug report [14:41] * rbasak pokes at the changelogs [14:48] jamespage: how about http://paste.ubuntu.com/1030468/ and http://paste.ubuntu.com/1030469/ ? [14:52] rbasak, beautiful [14:52] jamespage: can you grab from there or would you like new debdiffs? [14:52] (this is where a merge proposal would have been better, right?) [14:53] * rbasak could do that too [14:53] rbasak, pls stick debdiffs on the bug report - but yes a MP is better in this case I think [14:53] Now I'm confused? What would you like me to do - debdiffs or MP?! [14:55] rbasak, if you have them in a branch already do MP's - if not then just upload the debdiff's [14:59] jamespage: https://code.launchpad.net/~racb/ubuntu/precise/modsecurity-apache/988819/+merge/109378 and https://code.launchpad.net/~racb/ubuntu/precise/apache2/988819/+merge/109379 [15:03] rbasak, looking now [15:03] ah - the drawback of a branch - full history of apache2 :-) [15:06] zul: is there any action required for swift to resolve bug 920197? or is webob enough? [15:06] Launchpad bug 920197 in swift "[SRU] webob last stable version 1.1.1 response header bug" [Undecided,Confirmed] https://launchpad.net/bugs/920197 [15:06] webob is enough [15:11] jamespage: use shared repo's .. at least you only have to get it once :) [15:12] SpamapS, normally do that [15:12] 97M .bzr [15:12] SpamapS, congrats BTW [15:12] jamespage: thanks :) [15:13] SpamapS, did not expect to see you this week (hence why I just nicked your ceph merge....) [15:13] looks like Debian delayed apache 2.4 for wheezy btw.. definitely makes me feel better about rejecting it for 12.04 [15:13] jamespage: I'm just poking my head in.. will not be active till Monday [15:13] SpamapS, good! [15:13] jamespage: feel free to MIR libfcgi and reinstate radosgw too :) [15:14] SpamapS, yeah - I was just getting to know the package a bit better whilst I documented that blueprint in a bit more detail [15:15] The next big merge challenge for quantal is php5 .. since we have to revive suhosin and the 5.4 patch is still just barely working [15:16] lol [15:16] ceph was not to bad - had to cherry pick one upstream commit to get it to build.... [15:17] jamespage: Debian is really far behind ... [15:17] jamespage: we should grab 0.47.1 from upstream [15:17] SpamapS, yeah - I was considering doing a new upstream release [15:17] snap [15:17] ! [15:18] jamespage: they made some changes specifically to make it easier to deploy w/ juju/chef [15:18] I believe said changes landed in 0.47 [15:18] SpamapS, do you know the Debian maintainer? [15:18] anyway, time to return to the baby fueled madness [15:18] jamespage: we email from time to time, but I don't know him all that well [15:19] jamespage: upstream incorporates most of his changes and mine into their own packaging repo... makes things pretty easy to maintain. :) [15:19] anyway, out === matsubara-lunch is now known as matsubara [15:29] zul: once this set of openstack updates goes through when is the next lot planned for? [15:29] jamespage: lemme check the blueprint [15:29] * jamespage wonders whether these should be targetted to12.04.1 even though they will land well before then [15:30] week 12 [15:31] they should i think [15:36] zul, yes - please - are you OK todo that? [15:36] jamespage: yep [15:36] zul: ta muchly [15:36] mmm...beeers :) [15:37] rbasak, just uploaded both of those SRU packages - thanks for doing that. [15:38] No problem. Thanks for the guidance! [15:38] how do i get how many MB of memory a particular process is using. i'm using htop and i'm guessing the value is under the VIRT/RES columns.. but i'm not sure what either means. === ejat- is now known as ejat === fenris is now known as Guest59258 === fenris_ is now known as Guest12299 === Guest12299 is now known as ejat === dendrobates is now known as dendro-afk [16:46] Daviey: i need a +1 please: https://review.openstack.org/#/c/8344/ [16:55] hi [16:56] i'm looking for some help #xubuntu users told me to come here ^^ [16:57] !ask [16:57] Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [16:59] !patience [16:59] Don't feel ignored and repeat your question quickly; if nobody knows your answer, nobody will answer you. While you wait, try searching https://help.ubuntu.com or http://ubuntuforums.org or http://askubuntu.com/ [17:00] My Amazon server has an issue with 100% inode, >>> we are looking to supress a directory which contain thousand of files in it ... but rm-rf * doesn't respond atm [17:20] heh, i forgot linux-virtual doesn't really exist any more in quantal. was wondering why my amazon images were ru nning -generic [17:24] hello all! :) [17:49] stgraber: given the last comment in bug 947617, I think we should probably add a rule to allow a container to mount procfs to its /proc [17:49] Launchpad bug 947617 in lxc "After update, lxc does not start" [High,Fix released] https://launchpad.net/bugs/947617 [17:50] hallyn: yeah, I supposed it won't cause any harm, what we don't want is allowing procfs somewhere else than /proc [17:50] hallyn: we probably should do the same thing with sysfs then [17:51] agreed [17:51] shoudl we ask for a new bug for this? [17:51] i don't really want to pollute that bug report with another fix :) [17:52] i'll just open it [18:02] zul: around? [18:02] adam_g: yeah [18:03] zul: school me. if we're commiting changes to our -proposed and server-dev branches for these SRUs, but the packages in Ubuntu branches have had updates applied to them in the meantime, when do those merge with what we're proposing for the SRU? [18:04] do we need to sync the -proposed branches with the packaging from -updates ? [18:04] Daviey: ^ [18:05] adam_g: the lp:ubuntu branches? [18:06] zul: yes, glance and nova are a couple of examples of packages that were updated already in ubuntu. [18:06] adam_g: they should be done automatically [18:07] zul: how are conflicts handled? [18:07] adam_g: script i think [18:07] hmm [18:09] hmm [18:09] So. [18:09] if i dput something to precise-proposed right now... once accepted, it should be auot committed into that branch [18:09] once it passes validation and promoted to -updates [18:09] it will hit that bzr branch [18:10] There shouldn't ever be conflicts [18:10] so its up to us/whoever to resolve those conflicts before building the source package that gets dput'd? [18:11] adam_g: well, what conflicts are you expecting ? [18:13] Daviey: there have been security fixes applied to some of the packages that also were applied to stable branches upstream. [18:14] hallyn: I just finished validating the lxc SRU, everything looks good this time, so expecting it to land later today or (more likely) on Monday [18:14] i'm just wondering whether or not we should be actively sync'ing our CI -proposed branches with the ubuntu branches, to make things easier when its finally time to upload (and avoid the potential for error) [18:14] stgraber: cool. and that's the end of waht we have queued up now right? [18:14] (so i can re-fill on monday :) [18:14] working on the newest userns kernel right now [18:15] adam_g: i'd actually so no [18:15] adam_g: I think that Ubuntu branches currently suck big style for SRU's [18:16] adam_g: there are other issues to solve before that [18:16] hallyn: yeah, haven't looked at the fixes in quantal, might have a few more to pick for the next SRU. These apparmor fixes sound like good candidates too [18:17] jjohansen: are you around? [18:17] adam_g: i think it would have proposed branches when it gets processed by the sru team [18:17] stgraber: yep [18:17] (lunch, bbl) [18:17] jjohansen: cool. Are you aware of any remaining apparmor bug affecting lxc that'd trigger some silent rejects? [18:18] stgraber: hrmmm, maybe one in the network path if you are using denied network rules where the quiet masking isn't being applied correctly [18:19] that one got pointed out to me yesterday [18:19] stgraber: do have a bug or failure to point me at [18:19] jjohansen: I'm looking into the /usr/lib/locale/locale-archive issue I mentioned a while ago, for some reason it's still happening and I'm trying to figure out exactly what's going on there === jeremydei_ is now known as jeremydei [18:20] stgraber: hrmmm, did I miss that one or am I just not remembering it? [18:21] jjohansen: well, I'm currently trying to figure out whether it's lxc's fault or apparmor's ;) I first saw it when we were in the middle of these apparmor bugs, so just blamed it on apparmor and went to do something else [18:21] stgraber: fair enough :) [18:22] stgraber: do you want me to start poking at it as well? [18:23] jjohansen: hmm, I just finished running a first test and it looks like it's lxc's fault :) [18:23] jjohansen: will poke you if it turns out to involve apparmor in some way... [18:23] stgraber: okay, sounds good === matsubara is now known as matsubara-afk === dendro-afk is now known as dendrobates === lifeless_ is now known as lifeless === Leseb_ is now known as Leseb [20:05] Does anyone have any experience with Cloud-init and chef? [20:05] I have had good luck up until the point i try to use initial_attributes values [20:07] They jus dont ever seem to get picked up === dendrobates is now known as dendro-afk [21:29] Hi, I installed Ubuntu Server 11.10 64-bit a couple of months back, and I've been getting some help from people here to set things up [21:29] Although, I seem to have downgraded my kernel to a generic 32-bit one, and now I can't manage to install the 64-bit linux-image-server one [21:30] It's installed and all, but I can't choose it as my default kernel anywhere [21:33] And when I search apt-cache I can only find 32-bit ones [21:33] I want the x64, not the x86 ones :( [21:34] !crosspost | Plizzo (but I cannot help, sorry) [21:34] Plizzo (but I cannot help, sorry): Please don't ask the same question in multiple Ubuntu channels at the same time. Many helpers are in more than one channel and it's not fair to them or the other people seeking support. [21:35] guntbert: Sorry === pdtpatrick_ is now known as pdtpatrick === dendro-afk is now known as dendrobates === aetaric_ is now known as aetaric === pleia2_ is now known as pleia2 [23:28] hello. i'm connecting to a https server on my ubuntu server (webmin on port 10000) but i don't have an SSL certificate. instead i want to verify manually that i'm connecting to the right server by checking the sha1 fingerprint that firefox shows me against the fingerprint of my server. how do i see the fingerprint on my server? [23:35] discopatrick: does this help? http://blog.scottlowe.org/2009/03/25/next-gen-stuff-verifying-the-sha-1-fingerprint/ [23:39] CrypticSquared, possibly... that's telling me to find the ssl certificate on my machine and then check the fingerprint, but the thing is, i dont think i have an ssl certificate... but firefox is giving me two fingerprints from somewhere, sha1 and md5 [23:43] oh, hang on... i actually have a self-signed certificate [23:43] maybe webmin installs one by default, or maybe it's in ubuntu by default [23:43] ok, now i need to get it's sha1 fingerprint! [23:49] hmm, this isn't really the same as checking the fingerprint of host keys [23:50] looks like i just have to trust that this really is my server, and add a security exception to firefox