[00:02] one hour left :) [00:04] ^^ eucalyptus... and with plenty of time to spare, eh robbiew ? :) [00:05] heh...yeah "plenty" [00:05] :P [00:05] * robbiew is more interested in seeing if an ubuntu-font package makes it [00:05] robbiew: I hope you weren't expecting everything to be built by midnight ? :) [00:05] * Daviey goes home for that day.. thanks :) [00:06] robbiew, I thought sabdfl mentioned on a bug that it "might be worth waiting"? [00:26] ack...but things change ;) [00:26] Have fun.... o/ [07:14] morning [07:17] good morning! [07:24] I'll accept the font [07:25] which means CD candidates won't be built for a couple of hours [07:53] Riddell, do you want me to prepare the tracker for RC? [07:53] ara: what does that involve? [07:54] Riddell, I mean, just to clean the old milestone, create the new one, change the message. So you can start posting RC images whenever they are ready [07:54] ara: yes please [07:56] Riddell, OK, done. Tracker is ready :-) [08:03] hi all. i've just found https://bugs.launchpad.net/ubuntu/+source/mountall/+bug/649591 [08:03] Launchpad bug 649591 in mountall (Ubuntu) "mountall spins eating cpu when 'nobootwait' option exists in fstab (affects: 1) (heat: 8)" [Critical,Confirmed] [08:04] which is somewhat critical. [08:06] hmm, gtk failed on all but i386, lovely [08:08] cjwatson: able to look at that mountall bug? [08:10] cjwatson, i can get you (or anyone) access to an instance that demonstrates it, but i have to go to bed. === jjohansen is now known as jj-afk [08:42] Could I trouble whomever is up to look at bug 649597 ? ubiquity-slideshow-ubuntu 27 is sitting in the queue, and if possible and there's still time left, I'd like to squeeze it into the RC. [08:42] Launchpad bug 649597 in ubiquity-slideshow-ubuntu (Ubuntu) "Everything exception request for ubiquity-slideshow-ubuntu 27 (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/649597 [08:42] assuming we haven't begun spinning CDs [08:42] and there it is :) [08:46] :) [08:48] ev: accepted [08:48] I wonder if I should be running the publisher manually [08:48] sadly, I don't remember how [08:48] Riddell: you rock! [08:53] Riddell: uh, hang on, I just woke up and was dealing with kids [08:54] smoser's investigations certainly have a smoking gun at the end of them. I'll get you something as soon as I'm vaguely compos mentis [09:09] * Riddell wonders where the release notes draft wiki page should be [09:11] MaverickMeerkat/TechnicalOverview? [09:14] yes, thanks [09:17] my current suspicion is http://paste.ubuntu.com/501999/ but I REALLY want to run this through something that isn't my brain for testing [09:17] * cjwatson tries to set up some kind of test harness [09:23] ok, initial test program hangs, good [09:38] mountall fix uploading [09:42] ogra: https://bugs.launchpad.net/bugs/649327 [09:42] Launchpad bug 649327 in ubuntu-cdimage "OMAP3 preinstalled image lack partition table, bootloaders (affects: 1) (heat: 6)" [Undecided,New] [09:42] that one yours? [09:45] cjwatson, your fix was removal of the i-- ? [09:45] as suggested in the pastebin ? [09:45] yes [09:45] i've got a test program here that passes when that fix applied. but fails otherwise [09:45] (just some bit of verification) [09:46] it was moving the next option back, but then stepping back when it shouldn't, which took it to the preceding comma; that then infinite-looped because it was repeatedly doing strncmp(...,...,0) which always succeeds [09:46] so I guarded against j being 0 as well, since that happens if e.g. you have a double comma in fstab options [09:46] full fix was http://paste.ubuntu.com/502011/ [09:46] smoser: yes, I have a test program here too [09:47] i can't seem to make it fial with double comma [09:47] but i'll trust you. [09:47] thank you cjwatson [09:47] previous code failed here with 'defaults,,comment=cloudconfig' [09:48] the function in question, BTW, is removing mountall's private options before passing them on to mount [09:49] i'm gonna sleep some. a job is polling archive for mountall 2.19 and then starting new RC ec2 builds. [09:49] my code must be buggy then, that string doesn't hang here. http://paste.ubuntu.com/502012/ [09:49] cjwatson, yep, fixed already [09:49] cjwatson, we just didnt get any image builds due to arch all/any skew [09:50] cjwatson, was caused by u-boot-linaro-omap3-begale not being in main [09:50] * ogra waits for the archive to settle down to trigger a new build [09:52] smoser: oh, right, it goes round the loop again and does another memmove and then unsticks itself [09:52] I don't know whether that's better or worse. it has the effect of sanitising the double comma [09:55] I think I prefer being more literal, which is what my 2.19 upload will do [09:57] so mountall 2.19 will get into archive sometime soon ish ? its in the unapproved queue now. [09:59] awaiting revew [09:59] review [10:00] since Riddell asked me to look at it, I assume he won't spin images without it [10:00] accepted [10:01] ta [10:10] some work to do to get the universe even remotely installable I see. [10:10] ... and by the universe I mean main ... [10:18] cjwatson: what problems do you see? [10:19] loads of FTBFS, mostly transient [10:20] I'm working through them but it will take a couple of hours to clear everything I'd say, more on powerpc [10:20] http://people.canonical.com/~ubuntu-archive/testing/maverick_probs.html is giant [10:20] mostly that period when libglib2.0-bin was uninstallable completely killed us [10:22] yes, powerpc is not happy. glib and gtk are a pain but I think sorted now for main arches [10:23] not quite [10:23] libgnome, evolution-data-server still to build, libgnomeui can't even start yet [10:23] sorry I mean for kubuntu bits which is where I'm starting for the sake of simplicity [10:23] ah, fair enough [10:24] if you want to supervise that I can let you know when the other stuff is ready [10:24] armel is still having trouble with gtkish things [10:24] not quite as far behind as powerpc is [10:27] I must say, qa.ubuntuwire.com/ftbfs is gorgeous these days [10:35] * Riddell runs publisher by hand to get mountall in [10:39] hmm, that was suspiciously fast [10:42] but no mountall, how do I know when it's able to be published? seems fine on launchpad https://edge.launchpad.net/ubuntu/+source/mountall/2.19/+build/1978060 [10:42] <\sh> hmm..regarding my available updates from just now, ubuntu-desktop wants to be removed, including totem*, evince, ubuntu-private-fonts [10:42] Riddell: want me to do it? [10:43] \sh: we know about this on non-i386 architectures, please don't ask about it for a few hour [10:43] s [10:43] cjwatson: well am I doing something wrong? [10:43] <\sh> cjwatson: k [10:43] Riddell: I don't know, you didn't send the log to the usual file [10:43] running /srv/launchpad.net/codelines/current/cronscripts/publishing/cron.publish-ftpmaster [10:43] did you set LPCONFIG? [10:43] ftpmaster [10:44] and did you use the redirections from the crontab? [10:44] no [10:44] why not? [10:44] they're really useful [10:44] means that other people can see what's going on [10:44] ok I'll mind and do that [10:45] * cjwatson pokes around for mountall [10:45] can you put the publisher log somewhere? [10:46] It looks to be at least partly published. [10:46] indeed [10:46] we don't seem to have Packages files for it though [10:47] I am having massive packet loss to the UK at the moment, so it will take me a moment to poke around... [10:47] I wonder if process-accepted.py ran, but not publish-distro.py. [10:47] That almost matches the timing. [10:49] mountall 2.19/{amd64,armel,i386} is in DONE, which is consistent with that theory [10:49] definitely not on disk though [10:49] What's the publication status? [10:49] where do I see that? [10:49] https://launchpad.net/ubuntu/maverick/i386/mountall [10:50] Pending [10:50] Is the last entry Published or Pending? [10:50] And can my ISP break their international routing for the third time this evening? [10:50] 2.18 Published, 2.19 Pending [10:50] Hm. So it sounds like publish-distro.py didn't get anywhere. Logs would be nice. [10:54] Riddell: not sure what happened, but at this point it's probably most economical to turn the cron job back on and wait for it, maybe? [10:54] second run http://people.canonical.com/~jriddell/tmp/publisher [10:54] cjwatson: ok [10:55] Err. [10:55] you don't have LPCONFIG set properly. [10:55] mkdir: cannot create directory `/srv/launchpad.net/ppa': Permission denied [10:55] That's a little suspicious. [10:56] LPCONFIG=ftpmaster-publish [10:56] LPCONFIG=ftpmaster is wrong for this [10:56] be sure to copy things carefully from the crontab [10:56] Be glad there were no custom uploads... [10:57] will this have harmed any PPA publications? [10:57] Shouldn't have. That's a separate cron job. [10:57] * wgrant looks closer. [11:00] It could have gone a little badly if there were d-i or dist-upgrader uploads. But otherwise it broke before it started doing anything on disk. [11:01] I wonder why that config is like that, though. It doesn't make much sense. [11:06] it changed a few months back (maybe as long ago as a year?) [11:06] used to share a config with ftpmaster scripts [11:07] Yeah, the old copy of the configs I have doesn't have ftpmaster-publish. [11:17] Riddell, do we have an ETA of the first images? === doko_ is now known as doko [11:24] ara: kubuntu live are building now [11:24] Riddell, OK, thanks. And ubuntu? [11:24] after that? or are we waiting for something else? [11:27] ara: ubuntu desktop still has a bunch of problems [11:27] Riddell, OK, thanks for the update [11:29] http://people.canonical.com/~ubuntu-archive/testing/maverick_probs.html <- about that far away from working [11:42] looks like we're good now on amd64 except vim and those language packs [11:43] gtk finished building on armel which should fix most of the GNOME issues [11:44] those were an arch all any mismatch on gtk [11:44] yay [11:47] didrocks: it's your ISOs you're delaying :) [11:47] seb128: yes, I'm tracking it [11:48] it's not quite that simple as there's libgnome on top etc. etc. [11:48] Riddell: ok, getting there it in a couple of minutes :) [11:55] doko: OOo not for RC I presume? [11:55] Riddell: yes, when else? [11:55] doko: after RC [11:56] since we're already making images for RC [11:56] And OOo takes *forever* on some architectures... [11:56] not longer than gtk/gnome [11:56] Fair. [11:57] well they got in some hours before the deadline, this is too late whatever timezone we're counting [11:58] telepathy-gabble is dep-wait, is there going to be a new telepathy-glib? (I hope not for RC though) [12:01] seb128: ^^ [12:03] cjwatson, no new telepathy-glib for RC [12:03] i couldn't get the new version to build yesterday (and the current version also won't build with the new vala either) [12:05] so the package needs to be removed from the archive? [12:06] ... uh? [12:06] that's unlikely to help matters, there's an old telepathy-gabble which is still good [12:06] if we can get a new telepathy-glib after RC then it should work out OK [12:06] that's how much sympathy I have for obscure packages that don't build :) [12:07] Task: ubuntu-desktop, edubuntu-desktop, ubuntu-netbook [12:07] must be some new Kubuntu definition of "obscure" ;-) [12:07] i was thinking the same ;) [12:07] there's quite a bit of stuff in gnome that uses it [12:08] joking aside it's important that it's resolved one way or another for release [12:08] ok, i'll have to find someone who knows more about vala, so they can figure out why vapigen fails to create the bindings now [12:14] Riddell: new unity in the pending queue [12:18] do I have to do anything to make EC2 images? [12:18] 09:49 i'm gonna sleep some. a job is polling archive for mountall 2.19 and then starting new RC ec2 builds. [12:18] suggests not [12:19] so I have to look out for them appearing somewhere to add to the ISO tracker? [12:20] didrocks1: no immediate sign of it === ara_ is now known as ara === didrocks1 is now known as didrocks [12:21] Riddell: ok, just keep me in touch :) [12:22] ah, there she is [12:23] unity is written in vala? [12:24] didrocks: you don't want to keep it as a separate patch, just direct in the .diff.gz? [12:24] Riddell: yeah, unity is in vala and upstream is using bzr [12:24] ok, accepted [12:24] Riddell: so bzr merge is the easiest way to pick patchs in the packaging branch :) [12:24] thanks Riddell! [12:32] Riddell, cjwatson, chrisccoulson: ups sorry about that, telepathy-gabble is the jabber provider for the default im client [12:32] so yeah we need to fix that after rc [12:33] hey guys, have you experience in Maverick that after using a USB key it is not longer recognized? [12:34] it happens twice (with two differnet usb keys) in the last 10 days [13:04] hmm, the most recent kernel upload is more recent than the most recent debian-installer upload. I think I should upload debian-installer to sync up. Riddell? [13:04] (should have spotted this last night, sorry) [13:04] mm, fooey [13:04] but yeah, go ahead [13:05] how do I remove the alternates from the iso tracker? [13:07] Riddell: check them in the main display and press "Disable selection" [13:08] oh aye === rgreening_ is now known as ghost === ghost is now known as rgreening [14:27] ah, marjo, I'm to ask you something [14:28] Notify Marjo Mercado to begin ReleaseValidationProcess [14:28] Notify Marjo Mercado and ask for re-certification on test hardware [14:28] Riddell, what test case(s) should Kubuntu Mobile i386 have? [14:29] ara: just live session [14:29] Riddell, OK, sorry, adding it know [14:29] ara: you did do another upgrade test on the mini9 with dpkg from -updates without any changes in the time, right? [14:30] ara: its odd, I get similar results on my hdd based system, its a bit scary how slow it is :/ [14:30] Did we build xubuntu RC images for testing? [14:30] mvo, yes, I did [14:30] charlie-tca: not yet, we will shortly [14:30] Thank you [14:31] mvo, one was 10.04.1 directly to Maverick, the other was, 10.04.1 + dpkg from updates -> Maverick [14:31] mvo, both similar times [14:32] ara: thanks! [14:34] Riddell, fader: ack on Notify Marjo Mercado to begin ReleaseValidationProcess [14:34] Notify Marjo Mercado and ask for re-certification on test hardware [15:07] Riddell: any ETA for ubuntu-server RC candidates ? [15:07] just need to check d-i is in place [15:14] building ubuntu-server.. [15:17] Riddell: hi! could you perhaps add edubuntu i386/amd64 to http://iso.qa.ubuntu.com/ ? [15:19] highvoltage: mm, it's not built [15:20] highvoltage: given how broken gnome was when those were built I wouldn't trust them [15:22] Riddell: ah, I'm installing todays image and it seems just fine. perhaps the gnome brokenness is just not all that visible [15:26] highvoltage: looks broken to me http://people.canonical.com/~ubuntu-archive/livefs-build-logs/maverick/edubuntu-dvd/20100928/livecd-20100928-amd64.out [15:28] ttx: ubuntu-server is up [15:28] Riddell: thanks [15:28] * Riddell wonders who thought purple was a good idea for unvisited links colour in the new ubuntu cdimage index page [15:29] I'm fine having them rebuilt, though live session was fine and the installed system looked ok too ... [15:29] stgraber: it's weird, according to those logs the system you installed earlier should've looked broken, especially if it didn't have edubuntu-artwork or breath-icon-theme installed, and yet it looked fine in your image [15:31] ok, found the issue ;) [15:31] if the live fs doesn't built it uses an older one [15:32] yeah, I just noticed that. I thought it used to fail completely before and wouldn't appear on cdimage [15:32] heh, so much for my recent testing then :) [15:41] Riddell: I just pushed a change from smoser to the cloud images seed, to restore installation of tasksel in them [15:41] Riddell: I guess we need to wait for tat to be taken into account before generating the cloud images [15:42] Riddell: http://bazaar.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/ubuntu.maverick/revision/1755 [15:42] ttx: I guess so but I'm afraid I've no idea how cloud images are made [15:42] oh, I know how they are made [15:43] but I suspect we need a germinate run in between [15:43] which might require a manual push [15:44] you're probably right [15:45] would anyone like to own up to /srv/cdimage.ubuntu.com/debian-cd-2 on antimony and what the point of it is? [15:46] cjwatson: oh that'll be me, I messed up with the original [15:46] ttx: no point in the manual push until after the next cron.germinate run [15:46] I committed to /srv/cdimage.ubuntu.com/debian-cd [15:46] rather than /srv/cdimage.ubuntu.com/bzr/debian-cd and now they've diverged [15:47] so do I need to fix anything up? [15:47] cjwatson: when that would be ? [15:47] cjwatson: work out how to get /srv/cdimage.ubuntu.com/debian-cd merging happily from /srv/cdimage.ubuntu.com/bzr/debian-cd [15:47] ttx: an hour from now or so [15:48] Riddell: what's wrong with bzr pull --overwrite, if you just want to throw away the inappropriate commits to /srv/cdimage.ubuntu.com/debian-cd? [15:48] that sounds promising [15:48] cjwatson: perfect [15:48] it should never be a merge [15:48] thanks [15:48] cjwatson: let me try [15:49] ok. i will come back in ~ 1 hour and ask for manual push. [15:50] that did the trick, thanks [16:22] charlie-tca: you're up! [16:22] Thanks! [16:37] hi, please could somebody approve new packages / syncs from universe? see bug #642344 [16:37] Launchpad bug 642344 in libspring-2.5-java (Ubuntu Maverick) (and 1 other project) "libspring-2.5-java needs an initial manual build (affects: 1) (heat: 398)" [High,Confirmed] https://launchpad.net/bugs/642344 [16:43] doko: yes, those should be fine. could you please file a separate sync request for libhibernate3-java using requestsync? that way I can use our automatic tools to process that one and it's very much less work [16:43] actually, hm [16:43] maybe I can just open a task and nobble sync-helper a bit [16:43] never mind [17:02] cjwatson, skaet, Riddell: what sort of uploads will you review between the rc and release? [17:03] is it worth uploading GNOME 2.32 tarball where we have 2.31.92 when changes are only translations and version update to a stable number? [17:03] tarballs [17:03] there is a small stack of 2.31.92 to 2.32 updates which have only translations updates [17:04] the benefit would be to have the translations imported for the next langpack export (though that will probably for -updates later on) and have stable version number which is a minor detail [17:05] I'm just trying to figure if we should bother uploading those this week or target SRU updates later on [17:05] translations updates to 2.32 should be fine [17:05] seb128: I'd be fine with that, it helps upstream with debugging to have a well kent version number [17:05] ok, we will queue those then, thanks [17:06] i have a "sort of upload" too, gwibber 2.32.0, no change from 2.31.95-0ubuntu2, incorporates the patch included in 2.31.95-0ubuntu2 and the version number bump [17:07] I'm not sure I'd bother with things that are purely version number bumps [17:07] I don't know what sort of buildd contention we're going to have [17:07] i would prefer using the stable version... for bug triaging, etc [17:07] useful [17:08] but certainly nothing urgent [17:08] understood, and you can certainly upload it, but acceptance will be conditional on how much other stuff is going on [17:09] yeah, it is uploaded already [17:09] thx cjwatson [17:25] ok. i will come back in ~ 1 hour and ask for manual push. [17:26] this is me returning and asking for a manual push of germinate data to fix bug 649833 [17:26] Launchpad bug 649833 in cloud-init (Ubuntu) "uec images motd suggests tasksel, but tasksel not installed (affects: 1) (heat: 8)" [Medium,Fix committed] https://launchpad.net/bugs/649833 [17:27] * Riddell wonders how to do a manual push of germinate data [17:27] * smoser wonders if he's asking for the right thing [17:28] Riddell: carefully? [17:28] cjwatson: can you enlighten us? [17:30] I'd rather do it ... :-) [17:30] it's pretty delicate [17:31] and I always have to look up the details on the fly [17:32] must submit an LP change at some point to make it easier [17:32] actually, sod it. the easiest way is simply to accept some leaf package that doesn't matter for the CD builds. I will find a suitable one and do that. [17:33] I don't like running bits of the publisher manually when I don't have to === jj-afk is now known as jjohansen [18:19] highvoltage, stgraber: edubuntu syncing now [18:28] can someone reject my python-apt upload please? [18:30] mvo, done [18:30] thanks seb128 [18:31] Hi all. I own a proposal to merge desktopcouch 0.6.9-0ubuntu1 to Maverick. This is a new minor source release to fix two problems: 1) raw couchdb-accessing objects we helped create for clients didn't know how to reconnect when couchdb crashes. 2) Ubuntu One's credentials now use the official Ubuntu SSO OAuth creds, and the old two-legged hardcoded-consumer will no longer work for new accounts. [18:32] Riddell: thank you [18:33] CardinalFang: can it wait until after RC at this point? We've already started building candidate images [18:35] cjwatson, I don't wish to cause a respin, so it can wait. If we find a reason to respin, please do include it the, though. [18:38] "then" [18:39] Riddell: ^- [18:41] got it [18:55] cjwatson: updated bug #642344, working around "main" [18:55] Launchpad bug 642344 in libspring-2.5-java (Ubuntu Maverick) (and 3 other projects) "libspring-2.5-java needs an initial manual build (affects: 1) (heat: 10)" [High,Confirmed] https://launchpad.net/bugs/642344 [18:55] ttx: ^^^ [18:59] doko: I'm ok with it... that's better than not having spring at all, I guess [19:00] more to push onto JamesPage's java housekeeping natty spec [19:01] ok, thanks [19:11] Riddell: still syncing? I see http://cdimage.ubuntu.com/edubuntu/dvd/20100928.1/ is still quite empty [19:17] Riddell: ah, it's there now :) [19:19] cjwatson, so how will i knw when that push has been done (for task updates) [19:29] highvoltage: these DVDs aren't small [19:48] smoser: apt-get update; apt-cache show tasksel | grep ^Task: [19:48] (it should be done) [19:48] thanks [19:53] Hey! Just so i understand the process now.. Should debdiff's be on bug reports for ack by the release team... or bake in the queue for review? [19:54] seb128, cjwatson: http://people.canonical.com/~ubuntu-archive/NBS/python-gnome2-desktop [19:54] but even stuff in main depends on it ... [19:54] stuff in main being one source [19:55] we can rebuild it after rc [19:58] ok, for python-ubuntuone-client ubuntuone-client-gnome python-gnome2-desktop awn-applets-python-extras this is an alternative dependency [19:58] bzr-gtk labyrinth need to be fixed [19:58] Daviey: uploading should be ok [19:59] seb128: could the desktop team take care of? [19:59] Riddell, great, thanks [19:59] doko, it's in universe, could we let that to motus? [19:59] dinner bbl [19:59] seb128: nbs is not motu stuff, unfortunately [20:13] doko: I filed a bug about labyrinth ... oh, you got there? [20:14] https://bugs.edge.launchpad.net/ubuntu/+source/labyrinth/+bug/648469 [20:14] yeah, somewhat faster then the desktop team ... [20:14] Launchpad bug 648469 in labyrinth (Ubuntu Maverick) (and 1 other project) "depends on NBS python-gnome2-desktop (affects: 1) (heat: 6)" [High,New] [20:14] accepted, can you close that bug please? [20:15] doko: actually, contrary to what you said above, the MOTU representative in the release team has been mentioning NBS issues for some weeks now [20:15] the other packages listed in http://people.canonical.com/~ubuntu-archive/NBS/python-gnome2-desktop are false positives [20:16] I checked through them all the other day [20:16] well, except for bzr-gtk but that's only a recommends [20:16] still ought to be fixed of course [20:16] ahh, ok [20:16] release team> I meant "release meeting" above [20:17] anyway, both uploaded, can be removed when these are ain the archive [20:20] cjwatson: can I remove packages, if they are only recommended? [20:21] it's probably better to fix it [20:22] bzr-gtk isn't on CDs I think, let me check [20:22] python-gnomeprint can go, if python-gnome2-desktop is gone [20:22] no, all universe [20:23] universe doesn't mean not on CDs [20:23] (since there are non-Canonical-supported CDs) [20:23] anyway, accepted [20:37] cjwatson, the alternate installation takes very very long in kvm, is that expected? [20:39] it hasn't seemed unreasonable to me [20:39] ARM images take almost as long to build as DVDs [20:39] how long are we talking about? [20:42] doko, yeah, somewhat faster then the desktop team ... [20:42] doko, not a very constructive comment [20:43] doko, and no, we don't have the ressources to investigate all the universe ftbfs [20:43] come on, was you reply constructive? [20:43] how was it not constructive? [20:44] "no we don't have ressources to investigate universe issues we still have main ones to work on" [20:47] cjwatson, it started one hour ago, and it is on its 65% [20:48] that sounds plausible enough given the ext4 nonsense [20:48] nobody's come up with a decent fix for that yet [20:48] cjwatson, OK, thanks [20:51] cjwatson: we'll probably need another seed fix for bug 650566 [20:51] Launchpad bug 650566 in foomatic-db-engine (Ubuntu Maverick) (and 1 other project) "print-server task is not installable on Maverick RC (affects: 1) (heat: 6)" [High,Confirmed] https://launchpad.net/bugs/650566 [20:51] commit it and I'll find a universe package to push through the queue at the appropriate time [20:52] * ttx is confused by cjwatson's black magic, but doesn't really want to understand [20:53] it's not actually all that complicated [20:54] germinate is run separately from the main publisher, after it [20:54] but the output of germinate is fed back into the publisher, so there's some hysteresis going on [20:54] in practice it always stabilises, but it takes an extra publisher run to do so sometimes [20:55] the publisher only runs for any given suite if it thinks it has something to do [20:55] unfortunately, nothing tells it that it has something to do if the germinate output changes [20:55] one can work around this by stopping the automatic publisher runs and running it with a command-line option to publish a suite anyway [20:56] but it's easier to simply push a random package through so that it has something real to do [20:56] thus, we often like to keep a couple of reasonable but uncritical universe/multiverse packages in the queue during frozen periods, for this purpose [20:56] cjwatson: makes sense, thanks for the explanation :) [20:57] will commit I soon as I get serverteam peer review on it [20:58] your proposed change looks fine to me [20:59] foomatic-filters is probably what's actually wanted and foomatic-db-compressed-ppds depends on that already [21:00] yes, I think this seed could use some refactoring :) [21:00] will commit then [21:00] interestingly enough, this was caught early on by the automated ISO testing framework, Hudson-based, that JamesPage and mathiaz have been working on [21:05] cjwatson: done [21:05] http://bazaar.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/ubuntu.maverick/revision/1756 [21:25] hi - I've just upload a fix for eucalyptus to maverick [21:25] and ttx fixed a problem in the seed which lead to cups not working in the installer [21:26] The later will require a -server iso respin [21:26] and I'd like to have the new eucalyptus included in it [21:26] so - could the eucalyptus package be accepted in maverick? [21:27] * ttx eods [21:27] Riddell: http://people.canonical.com/~ubuntu-archive/component-mismatches.txt [21:27] do you want to seed these? [21:29] doko: please leave those alone [21:29] doko: I temporarily unseeded them to work around uninstallability [21:29]