[08:40] <seb128> infinity, hey
[08:41] <seb128> infinity, maybe you can help me ;-)
[08:41] <seb128> infinity, https://launchpad.net/ubuntu/+source/signon-ui ... do you know how Colin got 0.15daily13.06.12-0ubuntu1 in saucy release?
[08:41] <seb128> infinity, http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html complains about  "out of date on powerpc: signon-ui (from 0.14-0ubuntu2)"
[08:42] <seb128> infinity, I could delete the binary but the fact that it's still there suggest Colin forced it another way for the previous upload?
[08:43] <infinity> Lemme look.
[08:43] <seb128> thanks
[08:44] <infinity> Looks like Colin forced it.  Naughty.  I'll see about cleaning up a bit more pleasantly.
[08:44] <seb128> infinity, thanks
[08:44] <seb128> infinity, good luck, I think the reason Colin forced it was because it was non trivial to clean up nicely
[08:45] <Laney> he said he wanted to see what it made uninstallable
[08:45] <infinity> Yeah, it looks like it has no rdeps (anymore) on powerpc.
[08:46] <infinity> But I'll double-check this harder before I go breaking anything.
[08:46] <Laney> I thought seb128 said he deleted the ppc binaries yesterday
[08:47] <infinity> Should migrate in the next run.
[08:47] <seb128> Laney, signon-plugin-oauth2 and signond I removed
[08:47] <seb128> Laney, not signon-ui itself
[08:47] <Laney> ah
[08:47] <seb128> Laney, sorry if there was a misunderstanding
[08:47] <Laney> np
[08:47] <seb128> infinity, Laney: thanks
[08:47] <infinity> seb128: Erm, deleting the signon PPC binaries isn't clever, unless you also make it stop building on PPC, cause they'll just come back.
[08:49] <seb128> infinity, I think I just misunderstood what Laney said when he listed the rdepends/recommends, I though I needed to delete those 2 to have signon-ui move to release
[08:50] <infinity> seb128: Yeah, it's all cleaned up now.
[08:50] <infinity> seb128: But please get the signon source package to stop being arch:any (or make it fail early in debian/rules on !x86 !arm)
[08:51] <infinity> seb128: I'd almost prefer the failure thing, as it's easier to track if we ever get a Qt5 that supports more than 2.5 arches, but meh.
[08:51] <seb128> infinity, right, I'm going to talk to Ken about that
[08:51] <infinity> (Or make it needlessly build-dep on qt5-declarative, which makes it happily just dep-wait forever)
[08:52] <infinity> All the debian/control arch hacks we've been doing will be awful to unwind (or add new arches to). :/
[08:52] <infinity> Build-deps on the missing qt5 bits would have been much more elegant.
[08:53] <infinity> And I don't think it's too hideous to build-dep on your runtime deps for this purpose.
[09:19] <infinity> seb128: Aaaand, fixed a little harder.  Should go on the next cycle. :P
[09:19] <seb128> infinity, thanks!
[12:50] <xnox> infinity: debian started boost1.54 transition, but i'd want to hold that transition off until next cycle. And deffo out of main.
[13:15] <ScottK> If someone will kick ^^^ out of New, that'll get it and two other perl modules to migrate.
[13:18] <stgraber> ScottK: I'll take a look
[13:33] <stgraber> does anyone know what was used to generate https://wiki.ubuntu.com/PrecisePangolin/ReleaseNotes/ChangeSummary/12.04.2 ?
[13:56] <stgraber> slangasek: I'm done setting up https://wiki.ubuntu.com/PrecisePangolin/ReleaseNotes for 12.04.3 (moved all the existing ones to -12.04.2 and added all the needed links, updated includes, ...)
[13:56] <stgraber> I just need to ping IS to get https://wiki.ubuntu.com/PrecisePangolin/ReleaseNotes/UbuntuDesktop changed as it's currently immutable...
[13:57] <stgraber> the common wiki page is: https://wiki.ubuntu.com/PrecisePangolin/ReleaseNotes/CommonInfrastructure
[14:01] <stgraber> and updated the section about the backport kernel to cover the new 3.8 and lts-raring enablement stack
[14:05]  * xnox ponders if www.ubuntu.com is also ready for 12.04.3, as it started to encode point-release number on download pages et. al. pinged Peter about it.
[14:09] <stgraber> announcement draft: http://pad.ubuntu.com/ubuntu-12-04-3-announce
[14:10] <stgraber> flavors: please make sure the links on the wiki and on the announcement are correct, if not, let me know or just edit in place
[14:11] <stgraber> Riddell, ScottK, highvoltage, superm1, zequence: ^ (wiki pages for your release notes, download links and announcement URLs)
[14:12] <highvoltage> merci
[14:12] <smartboyhw> stgraber, working on the Ubuntu Studio ones...
[14:12] <Riddell> gotit
[14:14]  * stgraber adds the usual upgrade products
[14:14] <xnox> stgraber: does wubi need a bump .2 -> .3, upload and sign?
[14:15] <stgraber> xnox: that's a good question
[14:15]  * smartboyhw deletes the Ubuntu Studio upgrade 12.04.3 ones
[14:16] <stgraber> ah yeah, I should remove all those that weren't LTS before 12.04
[14:17] <stgraber> ah, actually they can pretty much all go as 10.04 desktop is no longer supported
[14:17] <stgraber> so only ubuntu server needs lts-to-lts upgrade testing
[14:18]  * stgraber cleans up
[14:18] <stgraber> xnox: so I'm not sure, it looks like we've been building wubi images lately, so maybe we should update wubi. I'd prefer to wait for slangasek to show up (any minute now) though.
[14:19] <stgraber> xnox: either way, I suspect we'll need a respin, either to get a new wubi to work with .3 or to get rid of wubi
[14:19] <slangasek> I asked ev to look at wubi, I wasn't sure if the point release instructions were still current
[14:19] <slangasek> xnox: do you know the answer?
[14:19] <slangasek> do we need a wubi respin?
[14:19] <slangasek> ... and if so, can you please take care of it (asap)?
[14:20] <xnox> slangasek: as in, wubi will try to download 12.04.2 metalink & .iso, when executed stand-alone, which I believe will error-out ones we release 12.04.3.
[14:20] <xnox> slangasek: so last time around metalinks got updated.
[14:22] <superm1> stgraber: all set
[14:23] <stgraber> slangasek: oh, and do we care about the world being oversized?
[14:24] <slangasek> stgraber: no
[14:24] <stgraber> ok
[14:24] <slangasek> xnox: what does updating metalinks mean?  is that a wubi rebuild, or what?
[14:29] <smartboyhw> stgraber, I have changed the link for Ubuntu Studio release announcement
[14:31] <xnox> slangasek: yeah, sed -i "s/12.04.2/12.04.3/g" * in the wubi/precise branch. Rebuild, scp to ~ubuntu-archive, sign by is, symlink update to "release"
[14:31] <slangasek> xnox: can you take care of building?
[14:41] <xnox> slangasek: ok after the meeting
[14:41] <slangasek> xnox: thanks
[14:52] <plars> slangasek: are we missing anything? Are there going to be netboot images?
[14:55] <stgraber> plars: http://archive.ubuntu.com/ubuntu/dists/precise-updates/main/installer-amd64/20101020ubuntu136.13/images/raring-netboot/
[14:55] <stgraber> plars: http://archive.ubuntu.com/ubuntu/dists/precise-updates/main/installer-i386/20101020ubuntu136.13/images/raring-netboot/
[14:55] <plars> stgraber: I hadn't seen them show up on the iso tracker
[14:57] <stgraber> plars: I just added them on it now, but the links will be wrong (should point to those I listed above or you won't get the enablement kernels)
[14:57] <plars> stgraber: ok, thanks
[16:50] <knome> infinity, looks like i can't grab an uploader soon enough for the docs SRU. i'm sorry for bothering you - i think you agree it's definitely too late now.
[16:51] <infinity> knome: Probably too late, yeah, but don't let that stop you from getting it uploaded. :)
[16:51] <infinity> knome: You might want to just ask to have it added to your packageset so you can upload yourself.
[16:52] <knome> infinity, "my" packageset? :)
[16:52] <knome> i personally do not understand enough about packaging.
[16:53] <knome> or at least, i won't confess i do... :)
[16:56] <infinity> knome: Oh, heh.  For some reason, I thought you had PPU rights to the xubuntu packageset.  Clearly not.  I blame my having been awake for only 50m this morning.
[17:02] <knome> nah, i don't, and that's not my goal. we need more people who have, though... and no problem :)
[17:02]  * knome pours infinity some coffee
[17:52] <sil2100> Core-dev needed!
[17:52] <sil2100> http://10.97.0.1:8080/view/cu2d/view/Head/view/Platform/job/cu2d-platform-head-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_qtubuntu_0.52+13.10.20130821-0ubuntu1.diff <- could any core dev ACK this packaging change for me?
[17:53] <sil2100> ogra_: ? ^ ;)
[17:56] <sil2100> infinity: ^ ;)
[17:57] <ScottK> sil2100: That's not a release team issue.  Please use #ubuntu-devel.
[17:58] <sil2100> ScottK: k
[20:51] <bdmurray> slangasek: what do you think about linking to an error report (to watch for it not occuring in the new version) in an SRU bug instead of a test case?
[20:53] <infinity> bdmurray: If it's so hard to reproduce that errors is about the only way to know if you've fixed the bug, I guess that's fine.
[20:53] <slangasek> bdmurray: I think that's a thing I've done before :)
[20:53] <infinity> bdmurray: But a testcase is always better, if one can be cooked up.
[20:54] <bdmurray> infinity: I tried a bit.
[20:54] <infinity> (Well, assuming the testcase is a valid reduction of the bug, and not some fabrication that tests the wrong thing :P)
[20:54] <bdmurray> and the patch is one line
[20:54] <bdmurray> -                except (IOError, zlib.error) as e:
[20:54] <bdmurray> +                except (IOError, EOFError, zlib.error) as e:
[20:55]  * infinity nods.
[20:55] <infinity> This whole "use errors to track progressions" thing will fail if the people who want -proposed to be NotAutomatic win.
[20:56] <infinity> Cause then no one will install the proposed versions except, maybe, the one guy who reported the LP bug.
[20:57] <bdmurray> infinity: which people?
[20:57] <infinity> You know.  "people".
[20:57] <infinity> It's come up a few times. :P
[21:01] <slangasek> people like me :)
[21:01] <slangasek> but yeah, we can't have it both ways on this one
[21:02] <slangasek> OTOH, NotAutomatic+ errors + phased-updates, maybe for those cases we just make sure it builds and installs and then throw it straight to -updates
[21:04] <infinity> (Those cases should be very, very rare, I'm not sure we want to model the archive usage after the corner case)
[21:04] <infinity> Unless you're suggesting that's how all SRUs should eventually go, which could perhaps be a tenable position when phased-updates proves its mettle.
[21:05] <pkern> canaries ftw :)
[21:14] <slangasek> infinity: I'm saying that -proposed should be noautomatic in its own right, and if one of the consequences is that we don't have enough data in errors to know if certain bugs are fixed in -proposed, then we have enough safeguards in place that we shouldn't be afraid to push it out to -updates
[23:25] <xnox> "autopkgtest for ubiquity 2.15.14: RUNNING (Jenkins: public, private)" but it finished an hour ago.... is that normal?
[23:25] <xnox> no like at 19:10 the first time.