[00:02] skellat: Read through it all. I suspect all the above would do is s/a3// from your discussion. [00:03] skellat: I assume those of you who are about release issues are subscribed to ubuntu-release@lists, so I don't have to hunt you down? :) === Guest26134 is now known as jbicha [00:04] infinity: Bingo. I just happen to be in Eastern Standard Time and saw this. [00:04] s/who are/who care/ might make my sentence make more sense. ;) [00:04] From our discussion we were pretty much focused on Alpha 2 as target. [00:05] skellat: Check, so dropping A3 shouldn't be a big deal. [00:05] (And dropping A3 isn't really up for discussion, IMO, it was a mistake that it was there in the first place, which is fairly obvious from the tiny 2-week gap) [00:05] Shouldn't be on our end. I can't speak officially for the release team of knome and elfy but our discussion today was pretty focused on getting ready for A2. [00:07] We still haven't heard from upstream Xfce as to when 4.12 will drop so we were still gearing up for when the toolchain changes hit. [00:08] skellat: No official 4.12 release schedule yet? :/ [00:08] infinity: That's what we're investigating [00:08] Hrm, yeah, a lot of strikethrough on http://wiki.xfce.org/releng/4.12/roadmap [00:10] infinity: Yeah, the wiki says one thing but the commits to the various pieces say something else. It's been a work in progress since vUDS to find out what's happening. It has only been a little while so we still have some time to figure things out before it gets to be a messy issue. [00:11] And we've got another meeting in 2 weeks anyhow [00:12] infinity: Thank you for taking the time to chat. I gotta go spend some more time with family again. [00:12] skellat: Happy familying. :) [00:12] infinity: can we, at least, get a firm date for A1? [00:16] were any flavors even at that vUDS discussion? [00:17] anyway since currently we're tracking stable gnome instead of unstable gnome I don't think the specific dates matter too much this time for the gnome guys [00:18] except I still think late October is better than mid October for final release [00:18] jbicha: Several were, yes. :/ [00:18] jbicha: Though some seem to have forgotten since. [00:18] that way we can get the 3.10.1 version of the few 3.10 things we cherrypick [00:21] for reference, 3.10.1 is October 16 [00:21] although I'm guessing final release isn't still up for discussion at this point? [00:21] jbicha: I was not aware of such a discussion, nor would I have flagged it. Certainly from lubuntu, we go with what -release team suggest and can then decide to take part, or not. (The hint is opt-in). lubuntu were not going to do A2 as our testers asked for one per month.... [00:22] I thought Kubuntu & Ubuntu GNOME didn't participate in May's vUDS [00:22] I think dropping a 3rd alpha is pretty uncontroversial [00:31] infinity: I am on the ubuntu-release mailing list, but may I humbly suggest that you also email the ubuntu-quality list, as these are the people who actually spawn the testers :) [00:44] infinity: does https://wiki.ubuntu.com/SaucySalamander/ReleaseSchedule/phillw make any sense? [00:46] phillw: Not in light of the discussion with ScottK, no. I've got this, don't worry about it. [00:49] infinity: okies. the flavours are all out of sync, I wish you well, but at the end of the day, we need at least a date for A1 setting :) [00:49] phillw: I don't think A1 will be moving. [00:49] yet earlier it was? A decision would be handy :) [00:50] s/handy/useful [00:52] phillw: I'm waiting for Scott to get back to me, but I think we've reached a decision. [00:58] infinity: well, as the earlier idea was for https://wiki.ubuntu.com/SaucySalamander/ReleaseSchedule to be fully updated today, I'm some what puzzled.... But, Meh... these things happen. lubuntu and xubuntu were set with the schedule, and now it gets changed with no input from us? Alpha's are <> and the two flavours have done so to the schedule. [00:59] phillw: Dude. Relax. Please. [01:01] phillw: A3 is being dropped, as was discussed at vUDS. Other things we decided to move might not move as it would mess with Kubuntu a bit too much, and they weren't at the meeting to discuss it before. Life will go on. This is probably not a crisis. [01:01] phillw: Also, I don't live in the UK, there's a lot more "today" for me still. :P [01:13] infinity: Don't forget "tomorrow" [01:14] If I keep this up, infinity is going to move the RelEng meeting to Saturday just to spite me. === jbicha is now known as Guest85286 === Guest85286 is now known as jbicha_ [02:17] wgrant: ^-- That was an accept... What happened? [02:17] wgrant: I was about to assume the packages timed out of the PPA, but they seem to still be there. [02:21] * infinity tries recopying. [02:24] Ahh, and when I copy it myself, I get a reject message. [02:24] compiz 1:0.9.9~daily13.05.31~13.04-0ubuntu1 in raring (source contains expired files) [02:24] Grr. [02:25] infinity: They're not copying from a staging PPA? [02:26] No. I asked them to, but they've not yet done so. I think a second round of failed SRU attempts might convince them. :/ [02:27] This is getting tiresome. [02:27] Bad enough that I have to jump through so many hoops to review a copy in the first place, then to have it rejected because it's not actually there anymore. Argh. [02:28] We can't just un-expire SPF/BPRF when we copy them? [02:28] StevenK: We can't undelete files, no [02:29] StevenK: I was assuming the above message meant things were gone from the librarian, not just disk. [02:29] Which is a bit hard to reverse. [02:34] wgrant: Hrm, actually, a staging PPA would get me useful diffs for review too, wouldn't it? [02:35] wgrant: Cause copying would trigger a diff against the archive version? Maybe? Or is that me being overly optimistic? :) [02:35] infinity: I don't remember if copies into a PPA do that atm [02:35] I suspect not, but don't quote me on that. [02:35] Actually, why doesn't their PPA have useful diffs anyway? [02:36] The kernel PPA always gives me a pair of diffs, one against the archive, and one against the most recent version in the PPA (if that differs). [02:36] Because bug #259422 [02:36] Launchpad bug 259422 in Launchpad itself "display PPA diffs against Ubuntu" [Low,Triaged] https://launchpad.net/bugs/259422 [02:37] * infinity scratches his head at how the kernel PPA seems to DTRT. [02:38] wgrant: https://launchpad.net/~canonical-kernel-team/+archive/ppa/+packages [02:38] wgrant: See the pair of diffs for linux/raring. [02:38] wgrant: That would seem to suggest that bug is fixed... For some people. :P [02:38] infinity: that's not because it's been copied into the primary archive? [02:39] wgrant: I was fairly sure those diffs showed up before I copy to proposed. Maybe I'm mistaken. [02:39] Hard to tell right now, since everything there is in proposed. :/ [02:39] linux_3.8.0-23.34_3.8.0-25.37.diff.gz is substantially newer [02:39] Look at the LFA IDs [02:40] 141820015 vs 142361264 [02:40] Hrm. Kay. Maybe I'm mistaken, then. Oh well. Would be a great feature to have. :P [02:43] I'm now going to go be grumpy and frustrated that I wasted an hour reviewing packages I couldn't accept. [02:43] Maybe gin will help. [02:43] infinity: Gin always helps? [02:43] Heh [02:43] StevenK: Generally. [02:43] Ginerally? [02:45] Maybe if I mix gin, a shiraz, a belgian blonde, and some cheesecake, I can have all of my favourite things in one place. [02:45] It'll be awesome. [02:45] Or I'll die. [02:45] Taking bets now. [02:46] infinity: ... Mixing how? One night, or a blender? [02:46] infinity: And is Australian lamb missing from that list? [02:50] cjwatson and infinity: Thanks for offering to be flexible. I've been really busy with $work this week, so I think Riddell is a better judge of where we are on packaging 4.11 Beta 1 and which week is better for Kubuntu for having Alpha 1. === doko_ is now known as doko [10:19] infinity, hey, I read the backlog after wondering why the unity SRU vanished from the queue ... is that ok if we just republish a round without a staging ppa? [10:19] infinity, we have graphical corruption on a range of video cards that is waiting for a fix for over a month, I would like to get those fix in without having to block on infra changes from didrocks' side [10:20] infinity, it's getting a bit ridiculous that we have our default desktop known to be broken on some machines with a fix available and it's taking us over a month to land the SRU :/ === mmrazik is now known as mmrazik|lunch [10:44] like I messaged infinity, the files actually don't seem like expired this time around, yet, look at for example https://launchpad.net/~ubuntu-unity/+archive/daily-build/+sourcepub/3234004/+listing-archive-extra [10:45] it does say 'removed' at the top (only 5 days after the build), but the files are still there at the bottom and downloadable, unlike for the week older packages [10:46] so it should be looked for the next time how those files can be retrieved, but since these syncs were rejected and we can't re-establish the sync request, we need to do manual reuploads of the 13.05.31 packages anyhow [10:47] Mirv: Sort of, the files don't actually get removed immediately, but they're half way through the removal queue so it still won't let you publish them in new places [10:48] Because once they're condemned for deletion it's not possible to unqueue them. [10:50] I'd really recommend you copy them from a staging PPA [10:50] wgrant: right, even though they stay there for over two weeks after being marked for removal. [10:50] Copying a package that doesn't actually exist any more sounds like a recipe for trouble :) [10:50] 5 days is too little, SRU Team cannot be expected to do review in that time, so yes it needs to be changed [10:50] Mirv: They become candidates for total deletion a week after they're not published anywhere any more [10:51] Then it can take a few days for them to progress through to actually being removed from the librarian [10:51] I've actually the 13.05.31 release at https://launchpad.net/~unity-team/+archive/sru/+packages?field.name_filter=&field.status_filter=published&field.series_filter=raring , so we'll probably get to reupload those [10:51] This is not something that will change in Launchpad; what you're doing is unsupported, and you'll need to use a staging PPA. [10:52] Plus a staging PPA will make it easier to review :) [10:52] ok, I'll let people know. I'm all for staging PPA, it just needs to be automated somehow to make those "blessed by us" snapshots from the daily-build PPA [10:52] Well [10:53] Does it need to be automated any more than the copies into the primary archive do? [10:53] Just turns "copy from daily PPA to primary archive" into "copy from daily PPA to staging PPA, copy from staging PPA to primary archive" [10:53] yet another ppa (tm) [10:53] but I guess that's easy enough [10:54] that system has soon 5 layers on ppa though [10:54] yeah, well blessing current day's releases is just a matter of LP copy, true. so maybe nothing to do, except if wanting for some reason to bless an earlier, superseded release (LP doesn't allow copying those, even though it allows initiating the copy) [10:54] non-verificated->verified->published->copied-for-archive->archive [10:54] You can copy a superseded one [10:54] You just can't copy one that's been superseded for more than a week [10:54] wgrant: no, it gives an error at the targe PPA [10:54] wgrant: ah, right.. [10:54] I wrote the code, I know the rules :) [10:54] that explains it :) [10:55] seb128: Sure, but you keep superseding old releases by uploading new ones [10:55] Things are copyable as long as they've never not been published anywhere for more than a week at at time [10:55] StevenK, yeah, I guess I wish that ppa copies were actual copies [10:56] not symlinks to targets that might go away [10:56] seb128: PPA copies are, unreviewed queue copies aren't [10:57] Not ideal, but not easy to fix without significant model reworks, and easy to work around. [10:58] right, that seems fair enough, some easy having for didrocks next week I guess === mmrazik|lunch is now known as mmrazik === peterm-ubuntu is now known as Peter-lunch [12:20] infinity, if date for A1 is changing why wasn't there a mail-out discussion on the ubuntu-release mail list. [12:20] ?? [12:21] ScottK, from discussions I had with Riddell yesterday, Kubuntu is going to be participating. [12:27] infinity, after looking at Britney update_excuses.html I still don't get why linux-mako/linux-manta are stuck in -proposed. [12:28] You must read update_excuses.html and update_output.txt together - they're two phases of checks [12:29] Though, OK, not an issue here [12:29] That looks like it's the same NBS-in-proposed problem as yesterday. I'll confirm and sort it out [12:30] cjwatson, ok. in this case update_output.txt didn't have anything in it that made sense (which is why I asked I guess) [12:30] Yeah, sorry, that was a knee-jerk :) [12:30] linux-{mako,manta} binaries are still in NEW, if nothing else [12:31] * cjwatson processes [12:31] thanks [12:33] That ought to be enough, I believe [12:33] cool [12:38] stgraber, from discussions with highvoltage, Edubuntu is not participating. [12:39] * skaet was working through some of the day-7 tasks yesterday, and planed a mail out to ubuntu-release later today, to figure out opt-in. [12:40] Will hold off on email until I understand a bit more on the date churn I've seen from discussion between ScottK, infinity in the backscroll [12:42] I think we need to decide if this is day -7 or not first. [12:42] * ScottK is waiting for Riddell to appear and voice an opinion. [12:43] hello [12:43] I think we should do it [12:43] Riddell: When? [12:43] Are we OK with moving it a week later? [12:45] yeah that's fine [12:45] infinity: ^^^ There's your answer. [12:46] skaet: ^^^ I guess we're at -14, not -7. [12:46] since we're not yet done with beta 1 and upstream doesn't seem to give us any packaging time now we can do our alpha then get back to upstream's beta === Peter-lunch is now known as Peter === Peter is now known as Guest61024 [14:01] ScottK, Riddell, infinity - ack. I'll go in and change the schedule then later. === Sweeshar1 is now known as Sweetshark === mmrazik is now known as mmrazik|afk [15:15] Riddell, ScottK: Lovely, I'll push that change out in a second, then. [15:39] we think we're done editing the schedule now :-p [15:39] ? [15:45] balloons: Yep. My bad for not getting full concensus and editing it earlier. Should be all finalized now. [15:46] no worries infinity I was just thinking I was crazy and had imagined the conversation at vUDs [15:47] balloons: You might still be crazy, but you didn't imagine the meeting. :) [15:47] :-) [15:49] uh in the past there have been problems having UI Freeze the same day as Docs Freeze; how can the Docs Team document things that haven't landed [15:51] jbicha: Hrm. I suspect there may be overload as to what constitutes a "docstring". [15:52] maybe the Unity (and related projects) won't land significant UI changes the night of UI freeze (or a week or two later) any more :) [15:53] A man can dream. Anyhow, freezes are always a bit flexible if one can show things like "this documentation is clearly out of whack with reality". [15:54] And, I've always taken doc string freeze to be an internal string freeze for software docs, and shipped things like the slideshow. External docs should be living, breathing things, and the idea of freezing them at all is probably silly. [15:55] In that sense, it makes sense for a UI and string freeze to be bundled in one, since strings in UIs are the very things we don't want changing before translators get their hands on them. But I can see the inverse interpretation too, that it's only about docs for software, not docs from software. [15:56] jbicha: Anyhow, of all the changes we made, that one's likely the most flexible, and I don't think anyone's keen on arbitrarily hamstringing documentation one way or the other. And that's something we can discuss right up to release, nearly. === slangase` is now known as slangasek [16:19] "how can the Docs Team document things that haven't landed" <-- The same way the always do when stuff lands late. ;-) [16:19] ScottK: Clairvoyance? [16:20] Dunno, I'm not on the docs team. [16:20] Probably the most important skill to list on a tech writer job req, yet I never see it listed. [16:21] I am loving all the back slapping from the Unity team on how complete and stable the 100 scopes stuff is in Saucy when the only reason it didn't go into Raring (and be late/immature/the usual stuff) is we told them no. [16:22] ScottK: Aww, let them have their fifteen minutes. Everyone likes to brag about doing something right, even if it's done right because someone made them. ;) [16:23] Yeah, this is a much better result than them seething with resentment [16:23] Damn my eyes. I read that as "teething with resentment" on the first pass. [16:23] That's why I'm venting a bit here instead of on planet.ubuntu.com. [16:24] I'm pretty sure children actually do that. [16:24] Oh yes they do [16:24] Also fire and the sword [16:25] * ScottK defers to cjwatson's more recent experience. === rtg_ is now known as rtg-afk === seb128_ is now known as seb128 [17:24] infinity: I was thinking about turning on apport's crash reporting to Launchpad now, additionally I think it could happen earlier in the cycle than in the past due to quality improvements. [17:28] bdmurray: I have almost zero opinion on the topic, I'd recommend running it past pitti.