[08:09] <seb128> xnox, hey, is bug #1721839 an upstream systemd regression you think? should it be reported there?
[08:11] <xnox> seb128, i am aware of that bug, but i currently do not have time to look into it, focusing on installer not working with static ips at the moment.
[08:12] <xnox> seb128, it smells like it, but i have not confirmed that; nor tested systemd master/v235;
[08:12] <seb128> xnox, k
[08:12] <seb128> so we can blame pitti right? ;-)
[08:13] <xnox> seb128, totes!
[08:24] <doko> rbasak: please could you walk over the list of ftbfs (main) for the last test rebuild and see what can to be fixed?  http://qa.ubuntuwire.org/ftbfs/rebuilds/test-rebuild-20170922-artful.html and http://qa.ubuntuwire.org/ftbfs/
[08:25] <doko> I mean, somebody from the server team
[08:25] <doko> jamespage: ^^^ same for the openstack team
[08:29] <Unit193> http://paste.openstack.org/show/xmFYlAoCdY1cOlu3G4Ds that's a bit cleaner network-manager.postinst in the case nplan isn't installed.
[11:45] <jamespage> doko: have a fix for the openvswitch s390x ftbfs which I've just uploaded - coreycb can you take a look at the neutron-* and swift failues ^^
[11:45] <coreycb> jamespage: sure np
[12:11] <Unit193> ogra_: BTW, finally tested '3', it works!  Not as obvious as 'text', but far better than the systemd target!
[12:11] <ogra_> oh my
[12:12] <ogra_> so the old "all runlevels are the same on debian" paradigm is actualy dead then
[12:13] <Unit193> Perhaps overall not exactly what one hopes for, but it has a nice upside. >_>
[12:31] <coreycb> doko: are the rebuilds against artful-updates or is there a PPA I need to be based on?
[14:21] <seb128> bdmurray, hey, I find that the is very little rhythmbox reports on e.u.c for artful could you check if there is a retracing problem behind or if we just don't get a lot of users/reports for it?
[14:50] <fuser> Hello. This is how my terminal looks now: https://i.imgur.com/hJqtWi6.png
[14:50] <fuser> (ubuntu 17.10)
[14:51] <fuser> The white is the terminal background color, the thickness of the border varies when I ctrl+ and ctrl-. I really doubt this was there before, I think I would have noticed.
[14:52] <fuser> Ive been on 17.10 for a long time, this is new. Might be nvidia driver update?
[15:13] <nacc> fuser: you want #ubuntu+1
[15:14] <fuser> thanks
[15:31] <smoser> is this table available in json
[15:31] <smoser>  http://autopkgtest.ubuntu.com/packages/o/open-iscsi/artful/amd64
[15:31] <smoser> or some thing ?
[15:31] <smoser> other than http scraping
[15:33] <smoser> nacc, ^ ? do you know ?
[15:33] <nacc> smoser: not that I know of
[15:34] <nacc> smoser: i *think* there might be a an archive tool that shows the same results
[15:34] <nacc> smoser: or does parse
[15:36] <bdmurray> seb128: Is it just rhythmbox or would apport not having been enabled in Artful for a while caused this?
[15:38] <bdmurray> smoser: they show up in update_excuses yaml
[15:41] <smoser> bdmurray, ?
[15:42] <smoser> i'm interested in json or something describing all the runs so that i can grab all the files and look at them, or do something other than browse
[15:43] <bdmurray> smoser: people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.yaml
[15:44] <bdmurray> smoser: that's the same content as update_excuses.html although I guess it only has failures
[15:50] <smoser> bdmurray, thanks.
[15:51] <smoser> so if i were interested in filing a feature request that i would expect would probably never get done...
[15:51] <smoser> where would i file ?
[15:53] <seb128> bdmurray, dunno, other applications are low, so maybe low number of users of desktop apps and apport was not working/being enabled recently
[15:54] <bdmurray> smoser: Oh the frontpage says SQLite database with results
[16:05] <smoser> bdmurray, front page ?
[16:06] <bdmurray> http://autopkgtest.ubuntu.com/
[16:16] <smoser> downloading 191M now :)
[16:38] <smoser> bdmurray, not that you have a real care, but i dont see in the results any artifacts or log urls.
[18:05] <jackpot51> GRUB no longer builds on artful
[18:20] <cjwatson> jackpot51: that comment would be most helpful if it came with an indication of exactly which version of the GRUB source code were involved and a pastebin of the error
[18:20] <cjwatson> you must have that information so don't make other people guess :)
[18:21] <jackpot51> Don't worry, I already have a patch to fix it. Newest version in artful doesn't build due to errors such as the one seen at the end of this log: https://launchpadlibrarian.net/340649401/buildlog_ubuntu-artful-amd64.grub2_2.02~beta3-4ubuntu6pop0_BUILDING.txt.gz
[18:21] <jackpot51> Just want to get your attention
[18:22] <jackpot51> I will make a launchpad bug with the patch attached
[18:22] <cjwatson> that's already fixed upstream too
[18:22] <cjwatson> and in Debian
[18:22] <jackpot51> Oh, then just need to wait on the sync
[18:22] <cjwatson> well it's probably a complicated merge
[18:23] <cjwatson> cyphermox mentioned this to me a couple of weeks ago so I assumed they were going to cherry-pick it ...
[18:23] <jackpot51> Is there a launchpad bug?
[18:23] <cjwatson> dunno
[18:23] <cjwatson> the upstream commit is 07662af7aed55bcec448bc2a6610de1f0cb62100
[18:24] <cjwatson> https://git.savannah.gnu.org/cgit/grub.git/commit/?id=07662af7aed55bcec448bc2a6610de1f0cb62100, more helpfully
[18:27] <seb128> slangasek, xnox, do you know if there are plan to enable systemd persistant journal for next cycle?
[18:55] <slangasek> seb128: hi, so we have the persistent journal on our "ideas" for 18.04 roadmap; we haven't committed to it, and it's my position that we shouldn't enable persistent journal by default unless we also disable syslog to disk by default
[18:56] <slangasek> seb128: which makes it a bit more controversial, and requires us to think it through carefully and discuss widely before taking a decision.  Is there a push to enable it from the desktop side, and if so what is that?
[19:01] <jbicha> Ubuntu 17.10's default log viewer is now gnome-logs instead of gnome-system-log, gnome-logs is systemd journal only
[19:09] <mwhudson> it would sure be nice if autopkgtest-virt-qemu actually worked :(
[19:48] <cyphermox> jackpot51: I sponsored grub2 for tdaitx earlier
[19:48] <cyphermox> cjwatson: ^
[19:48] <cyphermox> I was waiting for a grub2-signed, but I guess I should stop waiting and just do it now
[19:49] <tdaitx> cyphermox, sorry, I didn't realize you needed it that soon
[19:49] <tdaitx> I was working on the openjdk security updates
[19:49] <cyphermox> tdaitx: it should be uploaded with grub2, since otherwise it will stick in proposed
[19:50] <cyphermox> tdaitx: don't worry about it, I'll do it now
[19:50] <cyphermox> it's tiny
[19:50] <tdaitx> cyphermox, ack, I can stop and do it now as well, as aiming for EOD
[19:50] <tdaitx> *was
[19:50] <cyphermox> bah, don't worry about it
[19:52] <tdaitx> ok, tks cyphermox
[19:56] <juliank> oh, I already he apt SRUs are already 13 days in the queue and I have not verified them. Slow /me - time is moving to fast, aargh
[20:00] <juliank> I kind of write scripts and run them with sh -x in a docker container for validating things and then add the output to the bug as a comment :D
[20:00] <juliank> with before/after comparisons :D
[20:07] <cjwatson> cyphermox: ta
[22:37] <jackpot51> Thanks cyphermox
[23:42] <xnox> jbicha, slangasek - most of our images have systemd journal enabled.
[23:43] <xnox> slangasek, i was going to enable rsyslog precise journald plugin such that journald logs are in syslog, with matching identical timestamps, and do nanotimestamps by default.
[23:43] <sarnold> _persistent_ journal? or just journal?
[23:43] <xnox> slangasek, if we are going to break people, i'd rather do it for the good. E.g. making syslog timestamps precise will break most people's logrotate/logwatch
[23:44] <xnox> sarnold, currently we have journald set at auto -> which means it logs to /run/log, unless /var/log/journal exists. If /var/log/journal exists, it rotates early-boot logs from /run/log to /var/log/journal and then keeps logging on disk.
[23:45] <xnox> sarnold, we are discussing enabling it by default, meaning that /var/log/journal exists. It auto-rotates, compresses, and garbagge collects on both /run and /var/log such that it never uses "too much" of ram and/or disk space.
[23:45] <sarnold> aha, so it's 'yes except no one knows how to enable it' :)
[23:45] <sarnold> xnox: thanks
[23:46] <xnox> sarnold, yes, and journal thus is only available for the current boot, and after reboot past boots are gone.
[23:46] <xnox> sarnold, sucks for debugging and kernel lock ups and oopses and history or services crashing restarting.
[23:46] <xnox> sarnold, or on my systems i get something like less than than 2 weeks of logs, as my VPS does not have much RAM.
[23:46] <xnox> sarnold, many of our cloud images have /var/log/journal enabled.