[08:26] <mitchell`> aloha.
[08:26] <mitchell`> When filing a bug report on Launchpad, are there any markdown-style formatting options?
[08:26] <mitchell`> For example if I want to mark something as a code block?
[08:26] <mitchell`> The help system and links on the page reveal nothing, but I thought I'd ask.
[08:27] <lifeless> not yet, poolie has a branch in progress that will add this
[08:29] <mitchell`> cool
[08:29] <mitchell`> thanks :)
[10:40] <effie_jayx> dobey: I mean to merge the debian directory from one branch with the source from upstream to build  the package
[10:40] <effie_jayx> but what happens to debian/changelog, does it stay untouched?
[11:10] <Laney> bigjools: we just tried a sponsored sync & it seems to have worked well, thanks! the only thing we've noticed besides UI is that the sponsoree doesn't get email
[11:10] <Laney> want a bug for that?
[11:11] <bigjools> Laney: have they ever got email?
[11:12] <bigjools> I mean, yes, please file a bug if you want something but I need to know if it's something that is missing from the old syncs
[11:12] <bigjools> compared to old syncs I mean
[11:15] <Laney> checking
[11:15] <geser> IIRC I got accepted mails from sponsored syncs (when they got processed by an AA)
[11:15] <bigjools> ok thanks
[11:16] <Laney> you always got mail by virtue of the workflow being bug-based
[11:16] <Laney> I don't remember if /soyuz/ sent you one
[11:17] <geser> bigjools: will check my mail archive when I'm home to be sure
[11:17] <bigjools> ok , cheers
[11:20] <Laney> bigjools: yeah, confirmed that you do
[11:24] <Laney> bug #902114
[11:24] <bigjools> thanks
[11:27] <Laney> yw
[12:42] <Forage> Are the builds still catching up after the failure? Or is there a different reason why my uploads have been waiting 15 hours so far?
[12:43] <bigjools> we are missing some builders, they'll be back this weekend
[12:43] <Forage> ah
[12:45] <Forage> I assumed everything was back to normal again, or at least did not expect such a big delay
[12:47] <Forage> by the way, if I look on the build status page, I get to see "Pending (2505)" as a status for each package. What does the 2505 mean? Is it a status code or e.g. position in a queue?
[12:58] <bigjools> it's the score
[12:58] <bigjools> the queue is not revealed anywhere, but it's based on the score and the order of entry
[13:07] <dobey> effie_jayx: in recipe builds? no, it is updated with an automated entry at time of build, but the bzr repository itself is not modified
[13:48] <glen> hi, can i somehow add links to mailing lists of teams to project page?
[13:48] <glen> i.e i'd like to have on project page: https://launchpad.net/eventum two lists
[13:48] <glen> https://launchpad.net/~eventum-developers and https://launchpad.net/~eventum-users
[13:49] <glen> is it possible?
[13:50] <glen> currently i added as text to project description
[17:46] <SpamapS> is there a bug in bug search right now?
[17:46] <SpamapS> https://bugs.launchpad.net/ubuntu/+bugs?field.searchtext=&field.status%3Alist=NEW&field.importance%3Alist=UNDECIDED&assignee_option=any&field.assignee=&field.bug_reporter=&field.bug_supervisor=ubuntu-server&field.bug_commenter=&field.subscriber=&field.component-empty-marker=1&field.status_upstream-empty-marker=1&field.omit_dupes.used=&field.omit_dupes=on&field.has_patch.used=&field.has_cve.used=&field.affects_me.used=&field.tag=-notserv&field.tags_c
[17:46] <SpamapS> That should only show bugs with ubuntu-server as the bug supervisor
[17:46] <SpamapS> but its showing *all* bugs in Ubuntu that are New/Undecided
[17:49] <SpamapS> Hm if I manually rebuild the search it works right..
[17:49] <SpamapS> but the only difference seems to be adding &orderby=-importance
[17:51] <lifeless> well, the search you want is timing out
[17:51] <SpamapS> It has worked a few times for me
[17:52] <SpamapS> and timed out a couple as well ;)
[17:52] <SpamapS> without a bug importance, it times out a lot more
[17:52] <SpamapS> like, always
[17:52] <SpamapS> :-/
[17:53] <SpamapS> OOPS-e8a2841c618a6644cb0281502950af4a
[17:54] <lifeless> SpamapS: you can look at the oops yourself :)
[17:54] <SpamapS> lifeless: that is *seriously* nice
[17:54] <SpamapS> wow
[17:54]  * SpamapS is distracted by the wonder of closed loop error inspection
[17:54] <lifeless> duh :P why do you think I spent such time on this :)
[17:55] <SpamapS> figured you were bored ;)
[17:56] <SpamapS> So, is it the addition of the order by that is causing this problem? The server team has been using this particular search for at least 20 months..
[17:57] <christoph_debian> hi! is there somewhere a confortable way to say "this package in sid/preceise builds fine in $someolder-ubuntu please build it in my ppa" without creating a handfull of different version/distribution uploads?
[17:57] <SpamapS> ORDER BY BugTask.datecreated LIMIT 76 OFFSET 0'
[17:58] <SpamapS> wouldn't bug ID be more efficient? :-P
[17:58] <lifeless> SpamapS: there is a bug for tjis
[17:59] <SpamapS> christoph_debian: you may be interested in the 'backportpackage' script which will take any .dsc and spray it out to as many releases as you want, automatically uploading to PPA :)
[17:59] <lifeless> SpamapS: short story, something went over a threshold and the planner went nuts
[17:59] <SpamapS> lifeless: that is a pretty gnarly SQL statement.. ;)
[18:00] <SpamapS> WHERE BugTask.assignee IN (SELECT team FROM teams)
[18:00] <SpamapS> that in particular seems like it might cause fits
[18:00] <SpamapS> Oh wait thats a subselect
[18:01] <SpamapS> oi
[18:01] <lifeless> welcome to the mire that is LP bug search
[18:01] <lifeless> SpamapS: teams is a CTE
[18:02] <lifeless> SpamapS: not a subselect
[18:03] <SpamapS> I don't speak deep SQL-eze
[18:03] <lifeless> common table expression
[18:03] <SpamapS> "one of those things that you can shoot yourself in the foot with"
[18:04] <lifeless> think 'inlinable function'
[18:04] <SpamapS> So, when are we moving to Solandra ? ;-)
[18:06] <SpamapS> lifeless: seriously, SQL's ad-hoc capability is a blessing and a curse.. is there any plan to go to a more scalable search method? SOLR would eat this for breakfast.
[18:07] <lifeless> I know
[18:07] <lifeless> yes there are, but no staff time to do it
[18:09] <christoph_debian> SpamapS: indeed that works
[18:11] <lifeless> SpamapS: notice at the bottom - we get tracebacks for each action
[18:11] <lifeless> SpamapS: so we can tell what triggered it
[18:11] <lifeless> :)
[18:11]  * lifeless smugs
[18:11] <christoph_debian> hm no not that way "PPA uploads must be for the RELEASE pocket."
[18:12] <lifeless> SpamapS: so what is this a search of (in english)
[18:13] <lifeless>  Limit  (cost=6159595.53..6159595.72 rows=76 width=925) (actual time=27815.295..27815.320 rows=76 loops=1)
[18:13] <SpamapS> lifeless: all bugs marked new on packages that have ubuntu-server as bug supervisor
[18:13] <lifeless> righto, this is the one daviey had blow up last week
[18:14] <christoph_debian> this backportpackage doesn't accept a pocket name if one doesn't directly upload it seems
[18:14] <SpamapS> 27815, is that seconds? so .. only 7 hours
[18:14] <lifeless> SpamapS: http://explain.depesz.com/s/CQn
[18:16] <lifeless> SpamapS: bug 892820
[18:18] <SpamapS> lifeless: so there is evil in ordering.. wouldn't ordering by whatever the clustered key of the table is be the most efficient method?
[18:18] <lifeless> SpamapS: pg doesn't cluster
[18:20] <christoph_debian> d backportpackage also doesn't support supplying a different gpg key
[18:21] <SpamapS> lifeless: at all? sad.
[18:21] <lifeless> SpamapS: you can, manually, but it ages right away
[18:22] <SpamapS> lifeless: thats one of the best tricks in InnoDB.. put the important stuff early in the PK and you can order by it.. lightning fast order by.
[18:22] <lifeless> SpamapS: bug heat, for instance, causes a rewrite of every bug tuple every few days
[18:22] <lifeless> SpamapS: we can do lightning fast order by - we can - the problem here isn't order by being costly
[18:23] <SpamapS> sort + nested loop then?
[18:23] <lifeless> SpamapS: look at the plan
[18:24] <SpamapS> lifeless: I have zero knowledge of tuning pg queries
[18:24] <lifeless> note the seq scan on bug task *inside* an inner loop
[18:24] <SpamapS> its like op ork gleep ..
[18:24] <SpamapS> OOH
[18:24] <lifeless> now
[18:24] <lifeless> its throwing that into a hash
[18:24] <SpamapS> That I grok. Yes, scans are bad, mmkay
[18:26] <lifeless> the table scan is crazy
[18:26] <lifeless> there are 194 bugs matching your criteria
[18:27] <SpamapS> Yeah
[18:27] <SpamapS> part of the reason we may not have seen this before is we usually kept that number < 30
[18:28] <lifeless> no, its because its making bad predictions
[18:28] <SpamapS> well I was thinking maybe the prediction was different w/ 30
[18:28] <lifeless> hmm
[18:28] <lifeless> your url is different somehow
[18:28] <lifeless> the ss CTE is gone
[18:29] <lifeless> the url in https://bugs.launchpad.net/launchpad/+bug/892820 may work if you leave the bug listing beta
[19:02] <lifeless> SpamapS: that search, can I get a screenshot of you setting it up on the advanced search page or something ?
[19:06] <SpamapS> lifeless: is it enough to say that the only changes to the default are to make sure only status 'New' is checked, and type in 'ubuntu-server' as the bug supervisor?
[19:06] <SpamapS> lifeless: thats exactly what I've just done to reproduce it.
[19:13] <lifeless> SpamapS: let me see
[19:15] <lifeless> SpamapS: https://bugs.launchpad.net/ubuntu timed out. Not a win.
[19:35] <SpamapS> lifeless: seems that even getting more specific.. only bugs with NEW and UNDECIDED ... times out. :(
[19:35] <SpamapS> OOPS-090c23f65f06474b1dab9c4b998ff34f
[19:36] <SpamapS> wait that one isn't fair. ;)
[19:36] <SpamapS> OOPS-5e15e62026c1471f9750455874cb0482
[19:56] <Forage> I got my first build on launchpad and it failed to compile. According to the received e-mail I was free to "contact a member of the Launchpad Buildd Administrators team" for help. Does this channel count? :-D
[19:56] <Forage> https://launchpadlibrarian.net/87089367/buildlog_ubuntu-oneiric-i386.empathy_3.2.2-0ppa1_FAILEDTOBUILD.txt.gz
[19:56] <Forage> As far as I can tell it failed at "FAILED [dpkg-buildpackage died]"
[19:57] <Forage> The question is why
[19:58] <Forage> Is it the two "undefined reference to" messages before or is "dpkg-buildpackage: error: debian/rules build gave error exit status 2" a separate error?
[19:58] <Forage> the two "undefined reference to" don't look like real errors
[19:59] <Duditz> hi all! please, after I do launchpad actions, when my karma is updated?
[20:11] <Duditz> .
[20:15] <dobey> Forage: the undefined reference errors are the problem
[20:16] <Forage> hmm, not good then
[20:16] <Forage> that would mean the source contains bugs, doesn't it?
[20:17] <Forage> I can't imagine it's a dependency or build issue
[20:17] <dobey> maybe
[20:18] <Forage> or maybe outdated patches
[20:18] <dobey> it's not linking to an X lib
[20:22] <Forage> hmm, no additional patches touching that file
[20:59] <Forage> bleh, can't find it
[21:01] <Forage> why is it that whenever I "just" want to give something a try, stuff is going wrong in a weird way?
[21:02] <Forage> x11 is linked as far as I can tell, and no changes where made in that area to make it break
[21:39] <micahg> Forage: #ubuntu-packaging would probably be more appropriate since it's an issue with the packaging and not the PPA itself
[22:07] <Forage> micahg: thanks for the hint, I'll give it a try
[22:09] <Forage> I tried xclaesse as well, since he did some empathy packaging for the telepathy ppa already, but he's afk