=== chihchun_afk is now known as chihchun [07:27] pitti: I'm using something as ps3 media server - worked fine in vivid, works fine in wily if I boot with upstart, but if I boot with systemd - reboot/shutdown hangs for exactly 90s - what can I do to troubleshoot this now I've found what it is? [07:27] the something is serviio - uses java 8 [07:27] elfy: I assume you mean "works fine in vivid" == if you boot with upstart? [07:28] in vivid it works fine with either upstart or systemd [07:28] elfy: /usr/share/doc/systemd/README.Debian "Debugging boot/shutdown problems" [07:28] elfy: i. e. enable debug shell, switch to it while shutdown hangs, systemctl list-jobs will show you the running job(s); then you can inspect with systemctl status, journalctl, poke processes, etc. [07:28] ok - I'll have a look then - thanks :) [07:29] elfy: 90s is the standard systemd timeout for start/stop commands [07:29] pitti: ok ty === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === oSoMoN_ is now known as oSoMoN === alex-abreu|off is now known as alex-abreu === chihchun is now known as chihchun_afk [13:36] davmor2, hey, have you had a chance to check silo 32 ? [13:52] oSoMoN__, some latency here as we're sprinting [13:54] alesage, thanks! I understand there is some latency due to the sprint, I just wanted to make sure it doesn’t fall off the radar === oSoMoN__ is now known as oSoMoN [13:54] oSoMoN, it never hurts to ping :) === vrruiz is now known as rvr [14:02] oSoMoN: yeap first one today was just in a meeting [14:02] davmor2, awesome, thanks! [14:02] how is it going in Orlando btw? [14:14] oSoMoN: slowly but it is going. [15:04] pitti: seems2 things are still running, session-c2.scope and sys-subsystem-=net-device-virbr0-nic.device, no idea what either of those have got to do with this ps3 server though - and without that (and it's java tasks) it reboots as expected [15:10] elfy: the virbr0 thingy sounds relevant, something from libvirt? I think I faintly remember some similar reports during vivid, but they were pretty much completely fixed with bug 1438612 [15:10] bug 1438612 in dbus (Ubuntu) "remote file systems hang on shutdown, D-BUS stops too early" [High,Fix released] https://launchpad.net/bugs/1438612 [15:10] elfy: if you can reproduce that in a VM somehow, I'd highly appreciate a bug report [15:11] elfy: if you manually shutdown the virbr0 device in the debug shell, does shutdown resume normally then? [15:11] not sure pitti [15:11] what I'm right in the middle of is removing qemu as I don't use it much anyway [15:11] qemu? [15:12] that shouldn't affect shutdown at all; libvirt surely would [15:12] kvm or whatever - that I believe is where libvirt got installed at least [15:14] breaking things again I see elfy [15:14] balloons: there is a Y in the day :) [15:15] if you are up for it, the live q and a in 45 mins will have some QA folks in it.. I'd love for you to grill them :-) [15:17] although if you end up dual booting with qemu I might need to find a way to manually eject the disk as it won't eject the install media if it is not the first install on the vm if you try to disconnect the cd in virt manager for me but I think that is basically virt-manager doesn't expect dual boot vms [15:23] pitti: so without libvirt - it was just session-c2.scope running - seemingly no way to stop that [15:23] elfy: that's your user session; perhaps something is hung there? [15:23] anyway I'll see if I can see it in a vm install and use that [15:23] elfy: loginctl session-status c2 [15:24] elfy: that shows you the processes that are running in that session [15:24] elfy: kill them one by one to find the culprit? [15:24] pitti: if I don't go to vt9 with the debug shell - but just let it do it'sd thing - and then ctrl+al+del a bunch of times - it tells me alsa is trying to start [15:24] but - as I said going to set up a vm the same as this and hope that fails too - this rebooting in 90s is driving me nuts :p [15:25] elfy: yes, please let's debug this; you can be sure that affects more people :) [15:25] balloons: where live session- and is there an IRC channel to grill people [15:25] elfy: but I haven't gotten a report about hanging session-*.scope yet, so I can't give any tips in advance [15:26] pitti: yep,aas soon as I'm set I'll let you know and we can go from there [15:27] ubuntuonair.com, #ubuntu-on-air [15:27] happening in 30 mins :-)] [15:29] ok - just opening up the 50 page doc of grievances ... [15:29] I'll just paste that :p [15:30] elfy, perfect.. we won't be short on content [15:32] ha ha ha [15:33] balloons: 1700UTC is in ~90 minutes ... [15:34] oh nvm - got an e-mail just now :p [15:42] elfy, I was wrong.. mhall119 corrected me,,. 1700 it is [15:45] balloons: so 17:00 UTC for sure? [15:45] I have 3 nice questions lined up :) [15:46] awwwwwwwww fooey [15:46] elfy, yes 1700 for sure [15:46] wxl, if you have a question, leave it and I'll ask it [15:47] I'm sorry i jumped the gun on the time! [15:47] just excitied [15:47] thx balloons [15:47] esp. thx for including lubuntu-qa! [15:47] saves me the trouble XD [15:47] ofc! [15:48] hong kong phooey [15:49] xubuntu didn't get one :| [15:49] hating on Nick now ... [15:49] i ♥ hong kong phooey [15:49] * wxl used to have the lunch box [15:50] lol [15:51] balloons: yeah, I've updated our calendar event to match [15:52] excellent [15:54] elfy, xubuntu-devel@lists.ubuntu.com or > [15:54] ? [15:54] balloons: that is our dev list - that's where our QA stuff goes [15:58] there you go [16:40] pitti: ok - reproduced in vbox - if you really want me to jump through the hoops from qemu let me know, but at least I can debug sensibly now [16:53] are there scheduled quality team meetings? [16:56] elfy: how do you mean "jump through the hoops"? if you can reproduce it and have a recipe how to set up QEMU that it hangs, that's all I need :) [16:57] pitti: I mean actually reinstall something I don't use :p [16:57] but I'm doing that now [16:57] elfy: oh, "vbox" -- that'll do too [16:58] elfy: when I said "QEMU" I really meant "check if you can reproduce this from a clean install in a standard environment" [16:58] elfy: i. e. just open a bug, tell me what exactly you did to get the hang, and I'll see to reproduce it [16:59] pitti: what shall I report it against then? [17:00] balloons: and whomever is joining you on the Q&A: https://plus.google.com/hangouts/_/hoaevent/AP36tYc_Jg_IyTLOUIpvtL24uYefe8mWf-6639QfspZIAHUEtjyPdg?authuser=2&hl=en [17:00] pitti: I assume systemd [17:00] mhall119, ack [17:02] mhall119, we'll be a moment [17:02] ok [17:03] elfy: systemd sounds okay, I'll reassign it as appropriate (probably libvirt -- but do start with sytsemd, it collects interesting logs) [17:03] pitti: ack [17:03] I'll just finish up qemu as well now I've started :) [17:04] elfy: argh, sorry for the confusion and double work [17:04] pitti: no matter - best to prove it where I can I think :) [17:35] pitti: bug 1456730 I think I got everything [17:35] bug 1456730 in systemd (Ubuntu) "Hang after installing and running serviio" [Undecided,New] https://launchpad.net/bugs/1456730 === pgraner is now known as pgraner-food [17:47] Curious about how/if Xen gets tested (in or out of openstack) [17:51] ubuntu-qa: is it possible to automate integration testing for content transfers that require using multiple processes yet? [17:51] i created those testability peers to aid in this, but i think we still need work there [17:52] i think we could possibly test transfers between the testibility peers, but not initiated by url-dispatcher [17:53] there wouldn't be a way to test the success or failure [17:53] this is specifically around testing this branch: https://code.launchpad.net/~ken-vandine/content-hub/url-dispatcher/+merge/259039 [17:54] url-dispatcher is a fire and forget kind of thing, we won't have any way to confirm the other end succeeded in an automated way... it would launch another app [17:54] vrruiz, ^^ [18:10] nuclearbob, https://lists.ubuntu.com/archives/ubuntu-quality/2015-May/005962.html is the mailing list thread. I think a reply from you would make some folks happy :-) [18:12] balloons: anyway - so my main milestone issue is that last couple of cycles - everyone but Ubuntu has had to chase up issues that do affect Ubuntu [18:13] if - everyone diodn't bother with milestones until Final - there's just going to be more being shouted about right at the end [18:14] * wxl notes that there does not seem to be a ticket for the jenkins thing on rt [18:15] elfy, ahh. Sorry I didn't pick that up in the question. So your concern is that the issues wouldn't be noticed if flavors didn't test during milestones [18:15] in fact a search for jenkins in content only pulls up two tickets, both of which are resolved [18:16] so maybe it would be wise to report this to rt so we can track status, elfy/balloons [18:16] wxl, that's correct, there is no RT atm. [18:16] that way we don't have to ask the question about where it's at [18:19] balloons: elfy: if you guys don't have time, let me know and i'll jump in! [18:19] wxl, sure go for it :-) [18:21] oh man so excited to hear that formal testing is on the way!!! [18:21] * wxl just catching up on what i missed on ubuntu on air [18:25] balloons: yes - that's my issue in a nut shell :) [18:27] elfy, that's an excellent point. So given the fact ubuntu will continue with only a couple milestones, what can we do to mitigate issues? I think having those installer tests and the information obtained more readily consumable would allay your concerns. Would you agree? [18:29] I'd hope that would get closer to allaying them :p [18:31] sent [18:31] balloons: tbh - not sure what *we're* going to do yet milestone wise [18:31] balloons: also concerned that ALL flavours help with setting the 'pre-Ubuntu'milestones up too [18:32] Kubuntu/Xubuntu and Lubuntu did the last cycles ones [18:33] elfy, do you not think it's well represented? As I said, I thought you collectively have been doing a great job [18:33] balloons: not what I mean at all - I'll be sorely pissed off if the same people end up doing it over and over again is all :) [18:34] the other flavours need to also step up to the plate [18:34] elfy, it would seem to me however, dropping a milestone or two is still probably a good idea for most flavors. But that's just me. I think this has been reflected by teams skipping one [18:35] yea - we're likely to start at Beta again - though that's not set in stone yet [18:35] elfy, ahh.. yes, ubuntu gnome and ubuntu mate should play a role this time around. I believe / trust they will, but we can urge them towards that [18:35] ali hates irc :p [18:35] kylin as well.. :-) [18:35] yep [18:35] elfy, :-) [18:36] ok, so I think for the moment then focusing on getting ubiquity tests running again is a good way forward [18:37] yep - agreed [18:37] then, making sure other flavors provide coverage for milestones, while encouraging them to review how many they wish to have would be next [18:37] awesome, nice to be on the same page :-) [18:37] for sure :) [18:38] kenvandine, FYI we're sprinting, sure that ubuntu-qa wants to answer your q in time :) ^^ [18:38] balloons: so the next question - who was that bouncing about on your bed :p [18:39] alesage, thx, i can't think of a way we can test the url-dispatcher stuff, since it's really a fire and forget [18:39] kenvandine: what would be involved? Do you need to launch 2 apps or something at a lower level? [18:39] but vrruiz is blocking my landing :) [18:39] well i need to run url-dispatcher with a content url [18:39] then handle the app that it opens [18:40] but url-dispatcher won't tell me if the content transfer succeeded or not [18:40] it just hands it off [18:43] kenvandine: If the app that is launched is launched with testability that autopilot will be able to introspect it [18:43] elfy, you got to see Leo and Max, who are elopio and nuclearbob respectively [18:44] veebers, can url-dispatcher launch an app with testability? [18:44] balloons: I thought Leo was elopio -wasn't sure the hairy beardy was though :p [18:45] and actually it's harder than that [18:45] url-dispatcher calls content-hub-send which creates a transfer with the content-hub-service which launches messaging-app with ubuntu-app-launch [18:46] so quite a few levels there, plus dbus [18:47] we've never been able automate that stuff, and hadn't allocated the time to do all that for this sprint [18:47] elfy: thanks for your report! will check tomorrow [18:48] pitti: okey doke - thought you'd gone home by now :) [18:48] elfy: technically I never left home :) [18:48] elfy: but I was checking some important bug again [18:48] :) === pgraner-food is now known as pgraner [19:03] kenvandine: If you want to -> [blackbox] -> Apps is launched and displays received details. You should be able to. [19:04] kenvandine: Just in a meeting at the mo. but you should be able to set the testability env var for the launch of the app (at the end there) [19:14] wxl: lol - where did you drag piskie up from :) I don't see that nick much anymore === vrruiz is now known as rvr [19:36] elopio, http://people.canonical.com/~nskaggs/core-apps-test-sponsoring.html. I believe calc, terminal and docviewer are worthy of a look [20:10] elfy: [20:11] 1310 -!- elfy [elfy@ubuntu/member/forestpiskie] [20:11] 1310 -!- ircname : forestpiskie [20:11] heh - don'toften look at that :) [20:11] is that your launchpad id??/ [20:12] elfy is [20:12] everywhere but the old freenode account [20:12] and that e-mail address is almost dead now too [20:13] LP is correct [20:59] sjcazzol: http://people.canonical.com/~nskaggs/core-apps-test-sponsoring.html [21:01] evening elopio - you made my day bouncing on that bed :p [21:01] at least it looked like a bed ... [21:40] sjcazzol, hello [21:40] sjcazzol, https://code.launchpad.net/~nskaggs/autopilot/add-wm-sandbox-run/+merge/242274 [21:43] sjcazzol, elopio autopilot3-sandbox-run -X -s 400x600 [21:43] balloons, [21:43] balloons, thanks