| engineer-pearl | Hi... I think something is wrong with my apache2, and I don't really know what I'm looking for ((I'm teaching myself PHP, so I deffinately need it)). | 04:36 |
|---|---|---|
| engineer-pearl | I'm getting apastebin ready | 04:36 |
| engineer-pearl | this is the output when I try to restart it: http://paste.ubuntu.com/23596809/ | 04:38 |
| jerichowasahoax | Where do systemd service files live? I need to modify one. | 05:14 |
| jerichowasahoax | I understand that Canonical is not responsible if my modifications crash my server, kill my dog, etc etc etc | 05:15 |
| jerichowasahoax | oh i found it, /lib/systemd/system | 05:16 |
| === ooooooooo is now known as zero | ||
| === zero is now known as Guest29686 | ||
| lordievader | Good morning | 08:04 |
| === iberezovskiy|off is now known as iberezovskiy | ||
| === Piper-Off is now known as Monthrect | ||
| zioproto | coreycb: hello, upgrading Liberty to Mitaka we have hit this bug: https://bugs.launchpad.net/qemu/+bug/1626972 It is already fixed in qemu xenial. Any plan to backport to trusty in the ubuntu cloud archive ? | 08:51 |
| ubottu | Launchpad bug 1626972 in qemu (Ubuntu Yakkety) "QEMU memfd_create fallback mechanism change for security drivers" [Undecided,Fix committed] | 08:51 |
| zioproto | jamespage: it looks you know about this patch | 08:52 |
| jamespage | zioproto, 1:2.5+dfsg-5ubuntu10.7~cloud0 is in mitaka-proposed in the UCA | 08:53 |
| jamespage | I think that has the fixes | 08:53 |
| zioproto | thanks I will check | 08:53 |
| zioproto | jamespage: btw, for already running VMs with the old version there is no chance to live migrate them without stopping them I guess. Is that correct ? | 08:55 |
| zioproto | sorry, I got lost on launchpad, where is the git repository for the trusty mitaka proposed packages ? | 08:56 |
| jamespage | zioproto, oh its a UCA pocket | 08:56 |
| jamespage | add-apt-repository cloud-archive:mitaka-proposed | 08:57 |
| zioproto | thanks | 08:57 |
| jamespage | zioproto, re existing VM's | 09:03 |
| jamespage | hmm | 09:03 |
| jamespage | I would suspect that a stop/start is required to pickup the new qemu binary | 09:03 |
| zioproto | Yes, that is what I also think ! | 09:04 |
| zioproto | too bad users will have downtime | 09:04 |
| jamespage | tinoco would know for suer | 09:04 |
| jamespage | or cpaelzer ^^ ? | 09:04 |
| zioproto | jamespage: I checked out the sources and looks like the patch at hw/virtio/vhost.c is not there | 09:13 |
| zioproto | I did debcheckout --git-track='*' qemu | 09:13 |
| zioproto | using deb-src http://ubuntu-cloud.archive.canonical.com/ubuntu trusty-proposed/mitaka main | 09:13 |
| zioproto | that is the correct source tree ? | 09:14 |
| zioproto | I am looking for a patch that looks like this http://paste.openstack.org/show/591758/ | 09:14 |
| jamespage | zioproto, well the fix should be in https://launchpad.net/ubuntu/+source/qemu/1:2.5+dfsg-5ubuntu10.7 -> which maps to 1:2.5+dfsg-5ubuntu10.7~cloud | 09:27 |
| jamespage | I don't know specifically where the server team put their git/bzr source | 09:27 |
| jamespage | zioproto, I'd rely on the archive to be definative here | 09:27 |
| zioproto | yes this is the launchpad page I was looking for at my first question :) | 09:31 |
| zioproto | thanks | 09:31 |
| zioproto | but this is for xenial and not for trusty | 09:32 |
| zioproto | am I wrong ? | 09:32 |
| arcimboldo_ | the corresponding bug only reports yakkey xenial and z* patches | 09:35 |
| arcimboldo_ | I made my own packages | 09:35 |
| zioproto | jamespage: arcimboldo_ is the actual person that tested the patch at University of Zurich :) | 09:43 |
| zioproto | jamespage: OK the patch is there, but is that package going to land in trusty ? | 09:44 |
| jamespage | zioproto, that pocket targets trusty - or are you actually looking for the same patch to the qemu in trusty release? | 09:45 |
| jamespage | zioproto, i.e. mitaka UCA for trusty include qemu with fixes | 09:45 |
| zioproto | no, I think we are fine. So that patch will be in mitaka UCA trusty, as soon as the package gets out of mitaka-proposed, right ? | 09:46 |
| zioproto | so we just upgraded a few days too early ? :) | 09:47 |
| zioproto | I have another question. We are running Ceph hammer and now we notice that UCA Mitaka trusty ships Ceph Jewel. If we ping Ceph to Hammer there is any known major problem ? :) | 09:53 |
| zioproto | There is anything in the MItaka nova-compute that requrires ceph jewel packages?? | 09:56 |
| cpaelzer | jamespage: sprinting and just seeing the message know | 10:04 |
| cpaelzer | jamespage: is there a quetsion left open still? | 10:04 |
| === InfoTest1 is now known as InfoTest | ||
| cpaelzer | jamespage: I#d think you need to restart to pick up, but in theory you could also migrate to a same-version sibling, retstart the original source, migrate it back there and then hacve picked up the fix for the source to be able to migrate out to the much newer target which the fix is about | 10:06 |
| cpaelzer | jamespage: complex and certainly needs to be tested on the target environment, but might be the way to go without downtime | 10:07 |
| cpaelzer | zioproto: ^^ | 10:07 |
| zioproto | cpaelzer: thanks ! But I think it is too much work :) We have like 1000 VMs in our cloud | 10:10 |
| zioproto | making this mass live migration before the upgrade sounds terrible | 10:10 |
| zioproto | and arcimboldo_ already upgraded everything to mitaka | 10:10 |
| arcimboldo_ | yes, I actualy asked in the original bug if there was a way to fix for intsances that have the bug now | 10:14 |
| === PaulW2U_ is now known as PaulW2U | ||
| jamespage | zioproto, so far as I'm aware there is no requirement for jewel | 12:35 |
| jamespage | zioproto, that said we do only test mitaka with jewel | 12:35 |
| === deadnull is now known as _deadnull | ||
| === IdleOne is now known as Guest34309 | ||
| === deadnull is now known as _deadnull | ||
| mrtAkdeniz | Howdy! | 14:38 |
| mrtAkdeniz | how can I set the default permission, I mean if a new file created on that folder, it should have same permissions | 14:38 |
| ndboost | hey folks whats the inverse of -z in a conditional | 14:58 |
| ndboost | nvm its -n | 15:03 |
| rbasak | ndboost: "man test" is pretty handy for that. | 15:03 |
| ndboost | thx | 15:03 |
| rbasak | (rather than digging through the shell documentation, etc) | 15:04 |
| ndboost | i just relied on good ol'e google lol | 15:04 |
| === nodouble` is now known as nodoubleg | ||
| xnox | jamespage, at cdo core sprint; agreed with rbasak to remove the cloud archive from the archive. | 16:07 |
| jamespage | xnox? | 16:07 |
| jamespage | not quite sure what that means | 16:08 |
| xnox | what i mean is that i want to move ubuntu archive keyring from the stand alone package into src:ubuntu-keyring and highjack the old package name | 16:08 |
| xnox | and ship the cloud archive keyring as a the key fragment in /etc/apt/trusted.gpg.d/ (similar to ppas and the ubuntu-archive keys from zesty) | 16:08 |
| xnox | without using apt-key | 16:08 |
| xnox | and without injecting those keys into /etc/apt/trusted.gpg as that is now redudnand in the new world order of zesty | 16:09 |
| xnox | so, can i highjack ubuntu-cloud-keyring into ubuntu-keyring? | 16:09 |
| xnox | or do you want to keep it as separate source? | 16:09 |
| xnox | ubuntu-keyring already has archive and cdimage keys; and i hope to add cloud-archive, cloudimg (verification keys), ddebs keys | 16:10 |
| xnox | all in one source, with multiple binary packages as needed. | 16:11 |
| xnox | jamespage, does that make sense? | 16:11 |
| * xnox feels like i should create a bileto ppa with all the diffs I am trying to land. | 16:11 | |
| xnox | on a more important note, that ubuntu-keyring has migrated to key fragments. thus it's simple dpkg managed install/remove files in /etc/apt/trusted.gpg.d (! not conf-files) | 16:14 |
| xnox | rather than have postinst scripts that depend on gpg and manipulate things. | 16:14 |
| xnox | currently in zesty we only require gpgv now, as gpg is no longer a required package. | 16:14 |
| === Monthrect is now known as Piper-Off | ||
| coreycb | beisner, jamespage: hi can you promote magnum 3.1.1-1~cloud0 to newton-updates for bug 1632743 | 16:48 |
| ubottu | bug 1632743 in Ubuntu Cloud Archive "Missing files from python-magnum 3.1.1-0~cloud0" [Undecided,New] https://launchpad.net/bugs/1632743 | 16:48 |
| === deadnull is now known as _deadnull | ||
| neubi | how can i allow a website located in ~/website/index.php on ubuntu 14.04 server | 17:07 |
| === JanC is now known as Guest38161 | ||
| === JanC_ is now known as JanC | ||
| beisner | hi coreycb - magnum released to newton-updates | 18:23 |
| === iberezovskiy is now known as iberezovskiy|off | ||
| hallyn | man neubi was in a hurry. didn't get a chance to point him/her to https://help.ubuntu.com/lts/serverguide/httpd.html | 18:38 |
| nwilson5 | was having issues with what I thought were my database, but turns out to be an issue with sorting in ubuntu 16.04 vs 14.04. | 19:30 |
| nwilson5 | sorting a file of international text (2M lines) with: `LC_COLLATE="en_US.UTF-8" sort /tmp/utest1.txt > test` | 19:30 |
| nwilson5 | takes 10 minutes on ubuntu 16.04, and 20 seconds on ubuntu 14.04 | 19:30 |
| nwilson5 | 14.04 build has eglibc 2.19. 16.04 has glibc 2.23 | 19:31 |
| sarnold | heh, the one feels comically too long, the other feels comically too short :) | 19:31 |
| nwilson5 | neither has a problem if it's sorted using collate "C" | 19:31 |
| nwilson5 | the lines are all < 40 characters each | 19:32 |
| nwilson5 | this chan seems dead compared to #ubuntu :| maybe I should ask there. | 19:35 |
| teward | nwilson5: patience is also a virtue | 19:36 |
| nwilson5 | teward, perhaps, just scrolling up it appears less active than #ubuntu. A fair assessment. | 19:36 |
| nwilson5 | not saying no one would every respond. Just perhaps I should've been asking there to begin with. | 19:37 |
| nwilson5 | ever* | 19:37 |
| tsukasadt | I know this is better for #mysql, but noone is answering in there. Does anyone know how to enabled the MySQL logs to record executed queries, not just the prepared queries? | 21:16 |
| EvilAngel | I have been avoiding asking for help with this but I can't find a real solution anywhere. For months and now every day (to the point of being nearly unusable) I get the dreaded !caled_font-cache_frozen failed error when I launch any gtk apps. Is there a solution? I've deleted nearly everything in my .cache and .config dirs and ran fc-cache -r and manually deleted all font caches. Anything I'm | 21:27 |
| EvilAngel | missing? | 21:27 |
| Walex | EvilAngel: usually that happens when there are missing glyphs, adding more fonts with wider character sets usually helps. | 22:18 |
| Walex | EvilAngel: it is a bug in font handling code, but happens only with some fonts and not others. | 22:19 |
| Walex | Walex: also try this: https://gist.github.com/alechko/b1e9436b5a2507c50224 | 22:20 |
| EvilAngel | Walex: thank you. I installed all the ttf-* pkgs | 22:23 |
| EvilAngel | I think that might help, as it's what I just read another person did to fix it. I wish this bug would bug off, cause ti's been around for ages | 22:24 |
| zigmasterhanni | Hi all, quick noob question: Can anyone point me in the right direction to correctly configure DNS servers to point my domain to my VPS? | 22:28 |
| stomplee | zigmasterhanni, what are you trying to accomplish? | 22:43 |
| stomplee | your question wasn't clear | 22:43 |
| stomplee | if you are just trying to configure client settings refer to this post: https://askubuntu.com/questions/346838/how-do-i-configure-my-dns-settings-in-ubuntu-server | 22:44 |
| stomplee | if you are trying to configure a bind server, that is a whole other ball of wax | 22:44 |
| zigmasterhanni | stomplee, thanks for the link! I think that's all I needed | 22:46 |
| stomplee | yw | 22:46 |
| === petevg is now known as petevg_afk | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!