[00:16] hmmm!!!!!! [00:16] init: job logrotate failed to stop [00:16] that is the only one [00:17] it's all starting to fit together [00:17] so... let's see what happens if i wait a minute before logging out [00:19] yeah... the delay is only 5 seconds if that logrotate is not running [00:50] aaaaaaand... fixed [00:58] urgh... this code doesn't even actually work [01:19] so... not fixed? [01:26] well the delay is fixed [01:26] no idea what side effects it would have though [01:30] the quiesce code is really silly... it prints a list of "slow" jobs... but it doesn't do it after most of the jobs have stopped [01:32] it has a routine that tries to figure out how long to wait, but it returns 5 seconds by default even if there are no jobs to wait for - that's half the problem [01:36] its pretty awesome that you've been able to crack down on this so much [01:38] Which? The lag or the indicators? :P [01:39] well, pretty much everything he's been working on :D [02:05] :O [02:06] ok the reason it always waits at least 5 seconds? [02:06] it doesn't even send the shutdown request until the end of the wait period [02:07] that's why there's a different result depending if you log out or shut down [02:51] i think the logout bug is invalid, sorry [02:52] the 5 second delay exists to catch any jobs that start after upstart told all jobs to stop [02:52] it only happens on logout because in all other situations everything is going to get killed anyway [02:53] so it's a feature not a bug...it does exactly what it's supposed to, just not what some people want? [02:53] pretty much yes. that's what it looks like anyway [02:54] i suspect ochosi found it particularly annoying cos he is logging in/out in under a minute repeatedly to test things [02:54] in that case, the logrotate job will cause an extra 5 second delay - but only if you log out less than 1 minute after you logged in [02:55] which, let's face it, is not usual user behaviour [02:56] * drc tries to remember the last time ochosi was described as normal.... [02:57] perhaps the logrotate script can be fixed to that the sleep it does can be interrupted [02:57] and with that thought, off to bed [03:01] ali1234: How much of a regression do you think will be caused with our proposed simple fix? Shall I add a comment to the bug that this may be something we end up addressing in Release Notes as a "Known Issue"? Beta 2 Freeze is tomorrow FWIW. [03:01] what fix is that? [03:02] LP Bug 1227212 where somebody was talking about commenting a line out [03:02] Launchpad bug 1227212 in upstart "Session logout takes too long" [Undecided,Confirmed] https://launchpad.net/bugs/1227212 [03:02] Though my memory may be faulty [03:03] ah that... well, that means xubuntu does not use upstart to manage the session at all [03:03] the thing is though, the 5 second delay is obviously there for a reason [03:03] Probably to let something gracefully quit [03:03] no [03:04] it's to prevent new processes starting after everything is told to quit [03:04] at least that's what it looks like [03:04] Okay [03:04] but the bottom line is either they delay is needed, in which case it's better to have it, or it's not needed and it should just be removed from upstart [03:04] That makes sense [03:05] Have we had input from Steve Langasek yet? [03:05] no [03:05] I had a root canal & major filling done today so it has been a Weird Wednesday for me [03:06] What time zone are you in? [03:06] As I know Steve Langasek is on the Pacific seaboard of the US so he may or may not still be up [03:06] er... BST? [03:07] it's 4am here [03:07] Ah [03:07] No worries [03:07] anyway, the exact cause is pinpointed in my last comment [03:07] I'm stuck in the middle in Ohio [03:07] any upstart dev should be able to confirm that the delay is really there for the reason i speculated [03:08] Okay [03:09] Well, we'll wait & see what he has to say then. [03:10] If we don't hear anything we may have to roll such into a Release Notes "Known Issue" for Beta 2. [03:10] it's not even an issue [03:10] it's only 5 seconds [03:10] i think it's expected behaviour really [03:11] :-) [03:12] We'll let it marinate then and see what comes of it [03:13] anyway, not using upstart means using something totally different with a whole other set of problems, so i wouldn't recommend that at all [03:14] Yeah, I've been watching the conflagration on the Debian side over systemd. It hasn't been pretty [03:18] bug 1190526 [03:18] bug 1190526 in upstart (Ubuntu) "init crashed with SIGABRT in nih_discard()" [Medium,Fix released] https://launchpad.net/bugs/1190526 [03:19] hmm so the whole quiesce stuff was added in a big lump by james hunt in february and nobody else has ever touched it, so you might have to ask him what it's supposed to do [03:23] * Unit193 hopes they don't go systemd. [03:26] Unit193: That would be calling down death from above on our position [03:26] ali1234: I suppose so [03:27] EXCELLENT. Steve Langasek just chimed in on the bug. [03:27] Importance shifted to High, Status is Triaged, and now assigned to James Hunt for resolution [03:28] fair enough [03:28] Steve Langasek called it as the expected shutdown procedure is *not* happening in what you've described [03:29] it still looks like it is by design, but the design is wrong :P [03:29] but who knows [03:30] The biggest example of the design itself being wrong in the US in cards was the Ford Pinto. Smack the gas tank of that thing in a collision & the car exploded into flames. Otherwise it was a drivable car. [03:30] s/cards/cars/ [03:35] * skellat wanders off [05:48] Is there anything about the Xubuntu brand that I can see from a terminal? [05:48] * astraljava has forgotten loads of that sorta stuff. [05:49] I know about lsb_release, but that only mentions the distribution, not about the flavor. [05:59] astraljava, don't think so [06:03] astraljava, is xubuntu-* installed? [06:10] echo $DESKTOP_SESSION ? :P [06:11] GDMSESSION too. [06:11] XDG_CURRENT_DESKTOP for XFCE [06:21] knome: Oh, we ignoring OEM until T? [06:25] Unit193, yeah, it doesn't work [06:32] ali1234: congrats, you managed to wake the dogs with your debugging, looks like this is actually going to get fixed! [06:33] * knome knocks the wood [06:33] or at least looked at. [06:33] ow, that hit my head! [06:33] ;) [06:34] -documentation lacks upload, then it's ready [06:34] -marketing can lag, no release-based [06:34] -qa is ready [06:34] -community can lag, no release-based [06:34] -artwork needs wallpaper and upload, then ready [06:34] -development needs uploads, then ready [06:34] ^ status [06:36] looks like we have made it again [06:36] and only a handful of postponed items [06:36] Sadly, several things getting postponed, yep. [06:36] knome: I should think so, I installed from a daily a few days ago. [06:36] ochosi: thanks :) [06:37] knome: And what are you doing up at this hour anyway?! [06:37] astraljava, ircing. [06:37] and fixing kitchen table/chairs [06:37] and stuff.. [06:37] ;) [06:37] hangin' in the office [06:37] astraljava: See above? Also, head your /etc/apt/sources.list [06:37] Unit193, the good things is that we have a lot of nice things almost prepared for 14.04 [06:38] Yep. [06:38] Unit193, gtk3 indicators back, menulibre, mugshot, ... [06:38] 4.12? ;) [06:38] knome: Think he was happy with backports, no? [06:38] maybe, but not likely [06:40] Unit193: Okay, one moment, thanks! [06:43] Unit193: Ok, thanks again! [06:43] hi astraljava [06:43] Of course, and howdy! [06:57] knome: qa would have bee ready ages ago if I'd stopped adding things to it :p [06:57] s/bee/been [06:57] ;) [06:57] Hi elfy! [06:58] shan't make that mistake again - will just use scraps of paper on my desk ... [06:58] astraljava: how's things going - good I trsut [06:58] elfy, haha, np [06:58] :) [07:53] The new tumbler version is now in saucy. [07:55] elfy: Pretty well, thanks. Busy, but good. And how's life with you? [08:09] Noskcaj: cool! [09:15] yay, our new documentation is in saucy. jjfrv8, thanks! :) [09:25] knome, Is there any chance i could get added to https://launchpad.net/~xubuntu-team so time? [09:25] "so" ? [09:25] * Noskcaj has ran out of things to do while hist pbuilder creates [09:26] you're getting there. :) [09:26] *some [09:27] yep, don't worry about that [09:27] yay, eventual success ;) [09:28] looks like you're still finding your place in the community in general, which is fine [09:28] btw, thanks with all the work on packages [09:34] it looks to me that sponsored uploads are going to be handled if they are prepared well [10:22] woo for eventual success [10:25] bluesabre, ping me when you get back. [11:13] bluesabre: what a bummer, I should have checked the control file of xubuntu-default-settings more precisely, it states that "lightdm-gtk-greeter (>= 1.0.0-0ubuntu2)" is required, so adding "Provides: lightdm-gtk-greeter,.." to lightdm-gtk-greeter-gtk2 won't change anything, because of the version requirement [11:13] a limitation of "Provides" [11:49] brainwash, ah, bummer [11:49] oi bluesabre [11:50] bluesabre, apparently, the uploads by daniel have been related to *having a patch in the bug*, so maybe you'd like to do that for the shimmer-themes package as well [11:52] ah [11:52] ochosi: need some bugs :D [12:00] bluesabre: do you think, that adding the gtk2 greeter package to the dependency list of xubuntu-default-settings might be an option to consider? so if one the gtk greeter packages is installed on the system, xubuntu-default-settings won't complain [12:01] but maybe I'm the only one who still want to use the gtk2 greeter :) [12:03] brainwash: well, the gtk2 greeter is only available in that PPA, so I might be able to ship a daily xubuntu-default-settings in that PPA as well with a patch that removes the version check [12:03] bbl [12:03] seeya knome [12:04] bluesabre, btw, remember you scheduled a meeting @ 15UTC today ;)= [12:04] knome, I gave you an action item to reschedule or cancel said meeting ;) [12:04] bluesabre: yes, good idea :) [12:05] bluesabre, no, that's fine... if people show up, feel free to have the meeting [12:05] k, if I'm around :) [12:05] btw, it looks like the beta freeze is postponed until monday, but other freezes (doc/ui) stand today [12:05] lame [12:05] heh [12:06] well, that does make sense (documentation team needs to know the UI is stable, has nothing to do with beta) [12:06] (the non-existent documentation team, i know i know!) [12:06] (for ubuntu, that is) [12:06] ok, i'm off bugging you [12:06] see you later! [12:06] -> [12:06] yeah, just gotta hope mr_pouit or micahg appears [12:07] and the non-existent freezing things so that docs are out of date ... for ubuntu [12:07] cya knome [12:07] hard to keep docs up-to-date when we are in between having indicators and not [12:08] indeed [12:08] but I wasn't talking about us :) [12:08] true [12:08] I expect at least one more api change following beta2 freeze [12:09] there's at least 4 weeks left for sabdfl to decide something just has to be released and start the argument up again :) [12:10] nah, I think he's content with what's being delivered this release [12:10] :) [12:10] the big changes come with 14.04 and unity8 [12:10] I sort of take notice of that stuff - but tend to forget it all really quickly [12:10] bluesabre: you could get in touch with someone to upload the themes-package... [12:11] ochosi: not quite sure who to ping, and I'm away from my laptop so I can't do anything if there's any additional work to do [12:20] final beta is next week isn't it? [12:21] ? [12:21] yeah, beta2 is next week [12:21] freeze is monday [12:21] beta2 release is thursday (I believe) [12:21] UIF is today [12:21] just making sure before I make one last call for testing this cycle [13:13] ochosi: not sure anymore, but you mentioned that the restart menu entry gone missing after first logout is a known issue. does a bug report already exist? [13:14] ^ lightdm gtk greeter [13:14] brainwash: https://bugs.launchpad.net/lightdm/+bug/1226509 [13:14] Ubuntu bug 1226509 in Light Display Manager "lightdm_get_can_restart returns false in Ubuntu Saucy" [Undecided,New] [13:15] ochosi: thanks [13:16] np [13:17] but it works before login [13:17] directly after boot up [13:17] yeah, it's most likely not a greeter bug [13:17] cause the greeter adds those menuitems conditionally [13:18] depending on whether lightdm says that the system can_restart, can_hibernate, can_shutdown etc [13:18] and the restart-item always used to be there, not sure what's happening with it now [13:18] it needs some debugging, but that's no fun without a working --test-mode [13:19] looks like a job for ali1234 and his awesome debugging skills :) [13:20] way too much stuff involving lightdm did break [13:22] between lightdm updates and logind, we've been busy [13:24] ochosi, --test-mode didn't work with the power options anyway :) [13:25] yeah, i think mir just really put the focus away from the desktop stuff [13:28] once its done, it will be better. The converged unity8 will mean that those involved will be working on just one project instead of maintaining two [13:28] its just going to take a bit to get there [13:29] and I think Mir is showing good promise [13:29] I do wonder, what happens to X forwarding when Mir/Wayland are complete? [13:30] iirc wayland has a solution for that [13:30] that came up at one of the talks where i heard christan skagsgard [13:31] eh, or what was his last name again? :) [13:31] kristian hogsberg i guess [13:31] høgsberg i to be exact [13:33] I know wayland has an html5 backend, dunno about the others [13:35] gkt3 has an html5 backend called broadway, i dont know how that contributes to the discussion but its something i happen to know [13:35] P: [13:35] harhar [13:35] nicely played [14:11] yay! https://bugs.launchpad.net/ubuntu/+source/shimmer-themes/+bug/1227402 [14:11] Ubuntu bug 1227402 in shimmer-themes (Ubuntu) "Please update shimmer-themes to 1.6.2" [Undecided,Fix released] [15:05] bluesabre: got the update themes - panel now looks better - but the sound indicator icon appears to have done a runner [15:06] elfy, thats odd, didn't push an update to the icon theme [15:06] ochosi ^ [15:07] bbl [15:07] elfy: could you say that in more words? [15:07] ok, obviously not :} [15:07] ochosi: icon is missing [15:07] I think is the extended translation [15:07] :D [15:08] yeah, it seems to be a race condition [15:08] it sometimes is there, sometimes isn't [15:08] after a "xfce4-panel --restart" it always is there [15:08] elfy: ^ [15:10] ochosi: yep - that got it back - hadn't quite managed to wander off yet :) [15:10] good to get that confirmed [15:10] it's the only issue i get with the gtk3 indicators at the moment [15:11] not as good as it to see the only one of those indicators I actually ever use working again :p [15:15] ochosi bluesabre - one last thing before I do go for a while - with greybird the panel is black - same with all the supplied themes - except orion and raleigh - is that expected behaviour at the moment? [15:16] yeah, those are bright themes [15:16] which is why i added a new meta-icon-theme for them (elementary-xfce-darkest) [15:16] is not greybird bright? [15:16] it's mixed [15:16] dark panel, the rest bright [15:17] e.g. ubuntu's radiance is also a bright theme [15:17] bright panel, bright rest [15:17] ok - thanks - I understand [15:18] np [15:24] i only get the missing sound icon on the first start up after a reboot so it might be some race condition in the indicator startup [15:26] same here [15:26] ochosi: https://bugs.launchpad.net/lightdm-gtk-greeter/+bug/1227735 [15:26] Ubuntu bug 1227735 in LightDM GTK+ Greeter "disable/fix password input field context menu" [Undecided,New] [15:27] brainwash: yeah, got notified via email already ;) [15:27] but thanks for the reminder [15:27] :P [15:28] when i log out the greeter says "unlock" - anyone else seen that? [15:29] ali1234: that should've been fixed with a very recent greeter update [15:33] bbl [15:39] ochosi: yep, fixed now [15:39] elfy ochosi: regarding https://bugs.launchpad.net/ubuntu/+source/shimmer-themes/+bug/1227402. I've updated shimmer-themes and restarted xfce4-panel and even though the audio icon is the panel, as I can hoover it, it's not visible [15:39] Ubuntu bug 1227402 in shimmer-themes (Ubuntu) "Please update shimmer-themes to 1.6.2" [Undecided,Fix released] [15:39] is in [15:40] slickymaster: try xfce4-panel --restart [15:41] had the same thing [15:41] elfy: already done that to no avail [15:41] no idea then atm - not really here - sorry [15:41] it is there, but it's completely black. it's just a black square [15:42] elfy: np [17:12] brainwash: No, wayland has a RDP backend. === Maple__ is now known as o3o [18:24] bluesabre: Dangit, I meant you... Wayland has a RDP backend. [18:28] ah [21:05] knome: You ever get a chance to take a peak at the -core meta mockup? [21:06] Unit193: where would that be at? [21:06] ochosi: Some pastebin a while back. :P [21:06] Unit193, i did, but i've lost the url [21:07] Ah, alright. [21:08] as i said then, it looked like a sensible base [21:09] it shouldn [21:09] 't be too hard to create a sensible base at least [21:09] Bleh, means I have to hold onto the file a little longer. :P [21:09] well, we should really have a team meeting about that [21:10] ochosi: Thinking at final beta might not be a good time, maybe brainstorm time? [21:10] yeah [21:10] Hrm, new XPL then, though, right? [21:10] but the -core meta thingy, would you want that as a package? [21:11] who knows [21:11] Just tossed in with the seed, and boom it is. [21:12] Unit193, i'm still serving the 14.04 cycle ;) [21:12] at least... [21:12] Ah, good. [21:13] +1 [21:23] slickymaster: are you using the gtk3 indicators? [21:24] ochosi, yes [21:25] ochosi, I guessing you're asking because of the audio icon, right? [21:25] yeah, just wanted to check [21:26] it's kinda known, the indicator-sound icon [21:27] * ochosi still hopes ali1234 will be able to debug it... [21:28] ochosi, thing is, it's there. when i hoover my mouse pointer over it, I get the pop-up label indicating it and if I click it opens [21:28] yeah [21:29] it's just the icon that is not displayed [21:30] ochosi, let's hope it's solved before the end of this cycle [21:32] well it's far from sure that it's going to get into 13.10 [21:34] ochosi, yeah, so it seems, but never loosing our hope ;) [21:34] indeed [21:34] If not, then maybe backports. [21:35] nice to see more people are testing this === o3o is now known as Maple__ [23:59] i can't reproduce it reliably