=== jbicha is now known as Guest91535 | ||
babyface_ | anybody knows why there is no new precise build since July 15 ? http://cdimage.ubuntu.com/precise/daily-live/current/ | 09:23 |
---|---|---|
babyface_ | the daily iso test is blocked for this | 09:24 |
babyface_ | no new desktop precise build and no new server precise build | 09:28 |
cjwatson | it's still running | 09:30 |
cjwatson | not sure why it's late today | 09:30 |
cjwatson | looks like it's almost there though | 09:30 |
babyface_ | cjwatson, ok, could you help to check what's wrong ? we can not stand on this, coz it makes the test result ugly in Jenkins. thanks! | 09:33 |
cjwatson | babyface_: just wait | 09:33 |
babyface_ | cjwatson, ok | 09:33 |
cjwatson | since it's making progress right now there isn't really any point in me investigating anything | 09:34 |
cjwatson | and we already know what the underlying problem is, namely that there's only one useful arm livefs builder at the moment | 09:34 |
cjwatson | so I don't need to redo that investigation | 09:34 |
gema | cjwatson: is there a way to make both builds independent from each other? | 09:36 |
cjwatson | gema: get more arm livefs builders | 09:36 |
cjwatson | otherwise, no | 09:36 |
gema | cjwatson: what does that mean, buying more HW? | 09:36 |
cjwatson | this is all already in progress | 09:36 |
gema | cjwatson: do you know when we can expect it to be in place? | 09:37 |
gema | cjwatson: (to avoid asking you the same questions again and again) | 09:37 |
cjwatson | sorry, no, ogra_ might, but AFAIK it is at least partially in IS' hands | 09:38 |
cjwatson | anyway, your precise images are there now | 09:38 |
gema | cjwatson: thanks | 09:38 |
gema | cjwatson: will follow up with ogra_ | 09:39 |
* ogra_ reads backlog | 09:39 | |
ogra_ | ah, yeah, already in progress, i dont know any exact status though | 09:39 |
gema | ogra_: do you have an RT ticket or such? | 09:40 |
ogra_ | it was discussed at the last release team meeting and several people are on it atm, should be solved "soon" :) | 09:40 |
ogra_ | (several people ... including your boss) | 09:40 |
ogra_ | gema, i dont know if there is an actual RT (if so, i dont know the number), but there is definitely work going on, i'll ping you as soon as something changes | 09:41 |
gema | ogra_: thanks a lot ! | 09:43 |
cjwatson | I couldn't find a ticket for it | 09:45 |
gema | cjwatson, ogra_ I will check with pgraner tomorrow, he'll surely know the status of it | 09:46 |
cjwatson | I think the (possibly informal) search keyword would be "mandabox" | 09:46 |
ogra_ | heh, yeah | 09:46 |
ScottK | cjwatson: I taught sru-accept to also spit out the invocation for 'queue' in addition to providing the link to the LP page. | 13:21 |
cjwatson | ScottK: Cool, thanks. I was considering having it poll for a minute or two attempting the accept itself. | 13:21 |
cjwatson | But that may not be worth it since the fact that it can't be automatically accepted is a filed LP bug. | 13:21 |
ScottK | Personally I like the intervening step of thinking about do I really want to accept that into -updats. | 13:22 |
ScottK | updates | 13:23 |
ScottK | Since I'm just about to make something available to ~every user of that release, it bears an extra moment of thought. | 13:23 |
ScottK | I just didn't like having to think through the ./queue syntax each time. | 13:23 |
cjwatson | sru-accept could have an extra confirmation prompt, if you like | 13:26 |
cjwatson | I don't think the extra moment of thought needs to be quite as cumbersome. | 13:27 |
ScottK | That makes sense. | 14:14 |
ogra_ | gema, RT 54490 (in case pete didnt tell you yet) | 15:25 |
gema | ogra_: thanks, he hasn't told me yet | 15:26 |
cjwatson | skaet: I'm going to be out this evening at the time when the experimental freeze is due to start; do you think you could deal with asking webops on irc.c.c/#launchpad-ops to set quantal's status to "Pre-release Freeze" at the appropriate time? | 15:36 |
ScottK | bdmurray: Who's in charge of the code behind the pending-sru.html page? | 15:47 |
ScottK | It looks like there's a bug because the precise SRU for visualvm shows no bugs, but there are two in the changelog. | 15:48 |
ScottK | https://launchpad.net/ubuntu/+source/visualvm/1.3.2-0ubuntu2.1 | 15:48 |
ScottK | Both bugs are verification done, but I guess no one noticed because they don't show up. | 15:49 |
stgraber | ScottK: I don't think it really has an owner though it's been TIL by cjwatson and bdmurray (sru-report in lp:ubuntu-archive-tools) | 15:51 |
ScottK | stgraber: Thanks. | 15:51 |
skaet | cjwatson, can do. | 15:53 |
bdmurray | ScottK: I happen to be working on it now so will have a peek | 15:54 |
ScottK | bdmurray: Great. I'll hold off accepting the SRU so you have a test case. | 15:54 |
stgraber | xkeyboard-config is good to go (I just marked verification-done the last bug), so if someone is doing accepts, would be great to have it included (as some of the fixes have been hitting a lot of users, mostly french users) | 15:56 |
cjwatson | So, I'm taking a day off tomorrow, but we have the test freeze; could people keep an eye on unapproved uploads to quantal-release and accept them as quickly as possible, per the discussion on ubuntu-devel@? | 16:13 |
cjwatson | The command line to use is typically 'queue -s quantal -Q unapproved -e accept <package name(s)>' | 16:13 |
cjwatson | If it times out or otherwise misbehaves, please let me know | 16:14 |
cjwatson | If it completely explodes, SMS me | 16:14 |
cjwatson | (To the point where development is impeded, I mean) | 16:14 |
skaet | cjwatson, do we want to just hold off for a day then? | 16:15 |
* skaet thinking most of the problems are going to be in first 24 hours | 16:15 | |
cjwatson | Nah, I'm just as happy for more of the attempts to involve people who aren't me | 16:15 |
cjwatson | Since I've already used the new client myself quite a bit | 16:15 |
cjwatson | If for some bizarre reason there are things that are requesting-user-dependent, I'd like to know | 16:16 |
cjwatson | For example I'd certainly like at least one attempt at accepting an upload to main to be from somebody who's in ubuntu-archive but not in ubuntu-core-dev | 16:16 |
bdmurray | ScottK: for some reason the changes file has no Launchpad-Bugs-Fixed | 16:23 |
bdmurray | https://launchpadlibrarian.net/106865775/visualvm_1.3.2-0ubuntu2.1_source.changes | 16:23 |
bdmurray | I maybe shouldn't have accepted it then | 16:24 |
Laney | tumbleweed: ^^^^^ | 16:25 |
seb128 | hum | 16:25 |
Laney | borked DEB_VENDOR | 16:25 |
seb128 | do we have a standard way to deal with regression that entered <release>-updates,security? | 16:26 |
seb128 | we have an xorg segfault bug which is likely a regression from the recent SRU which got included in the recent -security update as well | 16:26 |
ScottK | Then it'll have to have a -security update and that'll get propogated to -udpates. | 16:30 |
ScottK | I'd talk to the security team. | 16:30 |
seb128 | sbeattie, ^ | 16:32 |
sbeattie | seb128: what bug number? | 16:32 |
seb128 | sbeattie, bug #1021517 | 16:33 |
ubot2 | Launchpad bug 1021517 in xorg-server "Xorg-server crashes reproducible with GIMP usage" [High,Confirmed] https://launchpad.net/bugs/1021517 | 16:33 |
seb128 | sbeattie, trying to get the #ubuntu-x guys on it but mlankhorst is not feeling well, cnd said he will have a look | 16:33 |
seb128 | sbeattie, it's likely a fallout from -0ubuntu10.3 | 16:34 |
ScottK | bdmurray: Thanks. I think that makes it not the scripts fault. | 16:38 |
* ScottK releases is then. | 16:39 | |
ScottK | is/it | 16:40 |
sbeattie | seb128: are they likely just to revert the -0ubuntu10.3 patch at this point or strictly trying to debug and fix? | 16:40 |
seb128 | sbeattie, I'm trying to get a sense of that | 16:41 |
stgraber | skaet: hmm, I guess I'll have to start to worry about oversizedness of the 12.04.1 images as it's not going to be as trivial to fix when stuff need to be fixed in -proposed/-updates ... | 17:30 |
skaet | stgraber, ack. | 17:31 |
cjwatson | stgraber: I think there's still an outstanding not-yet-implemented in livecd-rootfs/live-build whereby it doesn't have the thing we used to do in livecd-rootfs to make sure there's only one ABI version of linux-headers in the image | 17:41 |
cjwatson | probably doesn't help | 17:41 |
stgraber | cjwatson: hmm, indeed. kernel headers should compress pretty well, but that's certainly still not helping | 17:51 |
skaet | Daviey, where should we be picking up arm server image from? not seeing it: http://cdimage.ubuntu.com/ubuntu-server/daily/20120716/ | 17:51 |
skaet | and last preinstalled version was 0627 | 17:51 |
ogra_ | skaet, preinstalled fails atm (for unresearched reasons), we were actually waiting for the squashfs switch on x86 to then move on with it | 17:53 |
skaet | ogra_, what's the outlook on the squashfs switch? and where will the images be going? | 17:54 |
skaet | ie. shouldn't they be picked up with the dailies now, rather than under pre-installed? | 17:54 |
ogra_ | like for the switch on desktop from /daily-preinstalled to /daily-live the images will move from server/daily-preinstalled to server/daily | 17:55 |
ogra_ | arm is not enabled yet for normal daily server builds we're waiting for the switch here | 17:56 |
skaet | ok, had the right mental model at least ;) who are you waiting on for the switch? | 17:56 |
ogra_ | some announcement on -release or a ping from either colin or Daviey that it happened on x86 | 17:57 |
skaet | ogra_, who's been doing the work for the switch? | 18:01 |
* skaet figures this needs sorting before we get into next week. | 18:01 | |
ogra_ | skaet, Daviey and cjwatson worked on the squashfs stuff i think | 18:02 |
ogra_ | not sure where it stands | 18:02 |
* ogra_ has to go now ... | 18:03 | |
infinity | cjwatson: Oh, bah, we didn't port that hack over? Should be trivial, I guess. | 18:09 |
skaet | infinity, you on the squashfs stuff? | 18:35 |
skaet | or know what the plan should be here? | 18:36 |
bdmurray | slangasek, infinity, whoever: we'd talked about sorting the pending sru report by days in proposed should it go from 1 -> 1000 or 1000 -> 1 | 18:59 |
slangasek | bdmurray: I think oldest first makes most sesne | 18:59 |
slangasek | sense | 18:59 |
* ScottK was going to say 7 -> 1000, 1 - 6. | 19:00 | |
ScottK | That could just be me being contrary though. | 19:00 |
infinity | bdmurray: Oldest first seems sanest to me too. | 19:05 |
infinity | bdmurray: Ultimately, direction doesn't matter as much as the sorting, though, we should all be good enough at detecting patterns to figure out where "old" and "new" are, if there's a pattern at all. :P | 19:05 |
infinity | skaet: squashfs stuff? Oh, you mean for server? No, I've been staying out of that one. | 19:06 |
infinity | skaet: But I could look into the multiple kernels issue on precise, since that's a bit of a nasty regression. | 19:07 |
skaet | infinity, thanks. Please do. Will wait for cjwatson or Daviey to shed some light then... | 19:08 |
stgraber | cjwatson: apparently the parsing of the image name in post-qa for precise doesn't quite work. I "think" the following diff will fix it: http://paste.ubuntu.com/1095401/ | 19:18 |
stgraber | oh and I still need to change the value of $image to strip the release from the path (or tweak the expressions to match /$dist/ too) | 19:19 |
slangasek | Daviey: is there a blueprint which tracks the work you're doing on server squashfs? | 19:21 |
Daviey | slangasek: there is not. | 19:55 |
slangasek | Daviey: ok | 19:57 |
skaet | Daviey, since there are dependencies on the work emerging, can it be added to one of the existing blueprints, so we can minimize surprises please. | 19:58 |
Daviey | skaet: TBH, i don't /think/ there is anything left.. cjwatson had the baton last, so i'd need to check if there is anything outstanding. | 20:00 |
Daviey | ie, i wouln't know what WI's to raise at this stage. | 20:00 |
skaet | Daviey, well, it does seem to be the blocker on Ubuntu Server images for ARM at this point, so getting this sorted needs to happen. | 20:02 |
Daviey | skaet: I'll discuss with cjwatson was is left.. when he next shows. | 20:05 |
skaet | thanks Daviey | 20:05 |
skaet | https://wiki.ubuntu.com/PrecisePangolin/ReleaseManifest/12.04.1 has been posted, which should contain what is possible target for 12.04.1. | 20:36 |
* skaet needs to confirm with the signoff contacts that there will be 12.04.1 participation | 20:37 | |
infinity | skaet: I think netboot armhf+highbank should probably end up in that list too, but I suppose there's no point until the right bits are actually landed to make it happen. | 20:45 |
skaet | infinity, yes. | 20:46 |
=== yofel_ is now known as yofel | ||
tumbleweed | Laney: what are you ^^^ing me for? | 21:16 |
Laney | you sponsored the upload in question | 21:17 |
infinity | Laney: Wait, the close header is only added if vendor is ubuntu? That seems silly. | 21:23 |
infinity | Laney: Unless DAK actually explodes on the unknown field (which seems unlikely), what's the harm in having it in the .changes? | 21:24 |
infinity | (And DAK obviously doesn't explode on it, because http://packages.qa.debian.org/i/initramfs-tools/news/20120709T173229Z.html was accepted just fine) | 21:25 |
Riddell | could someone promote libkgapi to main? bug 1023954 | 21:30 |
ubot2 | Launchpad bug 1023954 in libkgapi "[MIR] Please promote libkgapi to main" [Undecided,Fix committed] https://launchpad.net/bugs/1023954 | 21:30 |
infinity | Riddell: Sure. | 21:32 |
infinity | Riddell: Wait. Why isn't kdepim-runtime in universe? | 21:32 |
Laney | infinity: Dunno. I don't make the rules. Ask buxy. | 21:32 |
ScottK | infinity: python-kde depends. | 21:33 |
ScottK | (which needs to be in Main for a variety of joint Ubunt/Kubuntu infrastructure reasons) | 21:33 |
ScottK | Riddell: I think you can promote it anyway if the overrides script. | 21:33 |
infinity | ScottK: Ahh, that makes some sense. | 21:33 |
Riddell | ScottK: where's that? | 21:34 |
infinity | change-override in ubuntu-archive-tools | 21:34 |
infinity | But already done. | 21:34 |
ScottK | As infinity says. | 21:34 |
Riddell | ah hah, I needed a bzr update to get it | 21:34 |
Riddell | thanks infinity | 21:35 |
bdmurray | why would the udev diff for precise be pending? | 21:39 |
infinity | * Rename mahjongg to gnome-mahjongg, following upstream. Provide a | 21:40 |
infinity | transitional package for upgraders. | 21:40 |
infinity | HAHAHAHAHAHAHA. | 21:40 |
infinity | Didn't the inverse just happen last cycle? | 21:40 |
* infinity head -> desk. | 21:40 | |
infinity | Laney: Seriously, didn't we just mangle a bunch of seeds for that rename in the other direction, or am I on crack? | 21:42 |
Laney | I'm not aware. | 21:42 |
* Laney checks le changelog | 21:42 | |
infinity | ubuntu.oneiric/desktop: * (gnome-mahjongg) [!powerpc] | 21:42 |
infinity | ubuntu.precise/desktop: * (mahjongg) [!powerpc] | 21:42 |
Laney | haha | 21:43 |
infinity | Etc. | 21:43 |
infinity | Double-U Tee Eff. | 21:43 |
Laney | http://git.gnome.org/browse/gnome-games/commit/?id=3ff289e78836186b3d6cf91cefe3a6e59e3dc518 | 21:43 |
Laney | go abuse Robert. | 21:43 |
cjwatson | skaet: I'm working on it still on and off in between emergencies. | 21:48 |
cjwatson | (server squashfs) | 21:48 |
cjwatson | stgraber: post-qa> looks fair | 21:49 |
skaet | cjwatson, it appears to be the blocker reason why we haven't had ARM server images for a couple of weeks now - is there a workaround? | 21:49 |
cjwatson | skaet: No. | 21:49 |
skaet | so we can get the testing infrastructure working again>? | 21:49 |
cjwatson | Not my fault somebody jumped the gun. :-P | 21:50 |
skaet | cjwatson, ahhh, understood. | 21:50 |
cjwatson | I've made no firm promises about dates at any point, as far as I know :-) | 21:50 |
* infinity wasn't aware this gun had been jumped. | 21:50 | |
infinity | Maybe I should be happy that I've been unaware. | 21:50 |
cjwatson | If people are suddenly depending on that work ... | 21:50 |
skaet | unfortunately, so it appears. :P | 21:51 |
cjwatson | skaet: No freeze? | 21:54 |
skaet | cjwatson, doing now, was otp | 21:56 |
* skaet appears to have hung her G+ hangout session. :P | 22:02 | |
bdmurray | skaet: what is the url for that? | 22:02 |
skaet | https://plus.google.com/hangouts/_/609d0fe5116fe21262bef201b8864cd22c519539?authuser=0&hl=en-US# | 22:03 |
skaet | bdmurray, let me know if that works for you ^ | 22:04 |
skaet | Slangasek, ScottK ^ can you join? | 22:05 |
skaet | slangasek, ^ | 22:05 |
slangasek | this remains to be determined | 22:05 |
skaet | RAOF, ^ | 22:05 |
slangasek | signs point to yes | 22:05 |
skaet | :) | 22:05 |
RAOF | skaet: Um, I don't see any url? | 22:11 |
skaet | https://plus.google.com/hangouts/_/609d0fe5116fe21262bef201b8864cd22c519539?authuser=0&hl=en-US# | 22:12 |
skaet | ROAF, ^ | 22:12 |
cjwatson | Well, 'queue -s quantal -Q unapproved accept' worked | 22:14 |
cjwatson | Hmm, one thing that occurs to me is that I bet P-a-s won't work properly | 22:15 |
cjwatson | Since the appservers don't have a copy of it | 22:16 |
cjwatson | I may have to move forward the work to get that into the DB | 22:16 |
cjwatson | skaet: ^- want to try that one? | 22:16 |
slangasek | skaet: ok, so the hangout exploded my laptop | 22:17 |
skaet | cjwatson, as soon as I get out of the meeting. will do if noone beats me to it. | 22:17 |
slangasek | after two reboots and manually killing the GoogleTalk plugin that took over my desktop, I'm trying again | 22:18 |
skaet | slangasek, ack. :( | 22:18 |
infinity | cjwatson: I thought wgrant (or you?) had done some work to make sure P-a-s was respected everywhere. | 22:20 |
cjwatson | infinity: No. I've made the code changes locally, but wgrant pointed out that in order for it to actually work I was going to have to arrange for appservers to know about the contents of P-a-s, and the preferred way to do that is to write a job to import it into the database. | 22:25 |
cjwatson | Then one of the script servers will need to have a bzr checkout but that's it. | 22:26 |
cjwatson | Fortunately the side-effects of P-a-s not being respected are generally along the lines of "annoying" rather than "OMG". | 22:27 |
cjwatson | (As we know because they already aren't respected for copy archives, manual accepts from the queue, direct copies (i.e. most PackageCopyJobs), and a corner case when initialising new series.) | 22:28 |
cjwatson | infinity: bug 564759 | 22:28 |
ubot2 | Launchpad bug 564759 in launchpad "P-a-s ignored when accepting non-sync uploads from queue" [Low,Triaged] https://launchpad.net/bugs/564759 | 22:28 |
infinity | cjwatson: Ahh, check. | 22:32 |
skaet | ./queue -s quantal -Q unapproved -e accept gnome-system-log ^ seems to have worked but it may have been someone else? | 22:52 |
skaet | cjwatson, ^ | 22:52 |
infinity | I think I got there before you did. | 22:53 |
infinity | Accepting gnome-system-log/3.5.4-0ubuntu1 | 22:54 |
skaet | infinity, I'm spotting kde-runtime in the queue, let me see what happens when I try that one. | 22:54 |
* skaet has permissions like the bot. :/ | 22:55 | |
cjwatson | Well, for upload. Queue admin should be independent. | 22:56 |
skaet | hmm... got the accepted message on my terminal but not seeing it showing up on the IRC channel. | 22:57 |
cjwatson | It takes a minute or two. | 22:57 |
skaet | and there we go | 22:58 |
cjwatson | Good. That's the only specific test I wanted to do; I suggest people just shoot on sight for a while when they see unapproved entries. | 22:58 |
cjwatson | I'll have a look through -changes at some point and see (a) if there's anything weird there (b) if there've been any accepts with a decent number of bugs. | 22:59 |
skaet | cjwatson. ok. Let us know if we can turn this off before Thursday. :) | 23:00 |
cjwatson | The P-a-s thing above is a bit annoying but not really much of a regression given that it was the case when people accepted stuff through the web interface. | 23:00 |
cjwatson | (This is why there were weird discrepancies before where sometimes P-a-s seemed to work during freezes and sometimes it didn't; it depended on whether the archive admin was using the queue script or the web UI.) | 23:01 |
cjwatson | skaet: Yeah, hopefully. I assume if I decide I have enough data I can just go ask ops for that myself? | 23:02 |
cjwatson | It's pot luck what kinds of uploads people do. :-) | 23:02 |
infinity | cjwatson: You can flip the bit back yourself. | 23:03 |
infinity | cjwatson: (And you could have flipped it on too) | 23:03 |
cjwatson | Er, I can? | 23:03 |
cjwatson | When did that change? | 23:03 |
infinity | cjwatson: Oh, I lied. I thought you had joined ~launchpad recently. | 23:05 |
infinity | cjwatson: (Which, for reasons I don't grok, has that permission) | 23:05 |
cjwatson | That was ~canonical-launchpad-committers. | 23:06 |
infinity | Ahh. Yeah, I didn't read the mail closely. | 23:06 |
infinity | cjwatson: In that case, you can ask ops, or me. :P | 23:07 |
cjwatson | DistroSeries.status requires launchpad.Moderate. ModerateDistroSeries is ModerateByRegistryExpertsOrAdmins, which is in_admin or in_registry_experts. ~registry contains ~launchpad. | 23:07 |
cjwatson | s/^/Editing / | 23:08 |
cjwatson | ~registry confers all kinds of weird god bits. | 23:08 |
infinity | It does, yeah. | 23:08 |
infinity | Like a slightly gimped duckie. | 23:08 |
cjwatson | The justification for status being massively restricted is nonsense, mind | 23:09 |
ScottK | skaet: As promised, I was away at the meeting time. | 23:11 |
infinity | cjwatson: Wasn't there some waffling about the world exploding if one made a series supported by accident, and then dropped back to devel? | 23:11 |
cjwatson | Sounds like handwaving to me ... | 23:11 |
infinity | cjwatson: Though, I suspect that if that was ever true, it really shouldn't be anymore. | 23:11 |
cjwatson | At one point there was a bit of badness if you have multiple development series, but dogfood's in that state and it works OK. | 23:12 |
skaet | ScottK, I'm going to schedule something for a morning later this week for those who couldn't make the afternoon slot. Prob on Thursday. Any times to avoid? | 23:12 |
cjwatson | I suppose setting precise to development would be kind of bad, but I think the distribution owner can be trusted not to be that stupid. | 23:13 |
ScottK | skaet: After 9AM Eastern any day that's not tomorrow should work. | 23:13 |
cjwatson | It could potentially cause the indices to be republished, but it wouldn't be massive mirror churn, just annoying. | 23:13 |
skaet | ScottK, thanks. | 23:13 |
ScottK | So what 'plugins' do I need to install to make a hangout work in Chromium? | 23:16 |
skaet | ScottK, not sure, am using it from Firefox myself. Let me know if you want to do a test at some point before then. | 23:20 |
ScottK | OK. Will do. | 23:20 |
ScottK | So far it's insisting I install some binary blob from Google. | 23:34 |
RAOF | Yeah, I think hangouts require that blob. | 23:35 |
ScottK | Sigh. | 23:38 |
ScottK | Not installing that on the production laptop. I guess I need to get the netbook fired up. | 23:39 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!