[00:12] <randomusername> hello?
[00:13] <randomusername> seems to be working
[00:13] <randomusername> hi there
[00:13] <wgrant> Hi.
[00:13] <randomusername> hi wgrant
[00:13] <randomusername> wgrant, can you tell me if are there any launchpad dev/support members here?
[00:14] <wgrant> Not many, since it's the weekend for most people, but there are still some.
[00:14] <wgrant> Ask your question, and someone will answer if they know.
[00:14] <randomusername> thanks
[00:14] <randomusername> "Thank you for your question concerning your Launchpad login that you are  experiencing. Case #number has been created and a support analyst  will get back to you within the next 24 hours."
[00:14] <randomusername> this was sent to me... two days ago
[00:15] <randomusername> tl;dr - after the 9/9 outage, my account got deleted
[00:15] <wgrant> What's the symptom?
[00:16] <wgrant> Can you no longer log in?
[00:16] <randomusername> when I logged in two days ago I went to an account that wasn't mine at all
[00:16] <wgrant> Ahh.
[00:17] <randomusername> and my account URL always point to OOPS
[00:17] <wgrant> What's the URL?
[00:17] <randomusername> my deceased username is ~tiagosilva
[00:17] <randomusername> I am now ~tiagosilva29 for some reason
[00:18] <wgrant> randomusername: So, ~tiagosilva doesn't OOPS for me.
[00:18] <wgrant> Does it for you?
[00:18] <randomusername> yeap
[00:19] <randomusername> I'm at home right now
[00:19] <randomusername> and at work is the same thing
[00:20] <randomusername> we browse via a proxy hosted in another country, and it's the same
[00:20] <wgrant> And you're logged in as ~tiagosilva29?
[00:20] <randomusername> when I log in I goto tiagosilva29
[00:20] <wgrant> And you get the OOPS when you're logged in?
[00:20] <slangasek> hi, I'm having persistent timeouts trying to file a bug against the ubuntu kernel package with apport; is there a known issue?
[00:21] <slangasek> oh n/m, apparently '/join #launchpad' was the magic rune to make it go through
[00:21] <slangasek> kthx ;)
[00:21] <randomusername> OOPS in ~tiagosilva ?
[00:21] <wgrant> randomusername: Yes.
[00:21] <randomusername> it's persistent
[00:21] <wgrant> slangasek: As in "Sorry, we couldn't connect to the Launchpad server" timeouts? or timeouts with an OOPS ID?
[00:21] <wgrant> Bah.
[00:21] <randomusername> logged in/out = same result, OOPS
[00:23] <randomusername> wgrant: I just logged in, and now I'm ~tiagosilva29
[00:23] <wgrant> randomusername: Ah, interesting.
[00:23] <wgrant> ~tiagosilva oopses on launchpad.net, but not edge.launchpad.net.
[00:23] <wgrant> OOPS-1714K1846
[00:23] <ubot5`> https://lp-oops.canonical.com/oops.py/?oopsid=1714K1846
[00:23] <randomusername> https://launchpad.net/~tiagosilva29 -> I just posted HELLO WGRUNT in that profile
[00:24] <randomusername> wgrant: I confirm that ~tiagosilva via edge is me
[00:25] <randomusername> aka the advice dog guy who hates easter eggs and translates more that it should
[00:25] <wgrant> Heh.
[00:27] <randomusername> as this issue occurred before?
[00:27] <wgrant> Not that I've seen, no.
[00:27] <wgrant> I'm poking around to see what changed.
[00:28] <randomusername> do you want me to stay out of launchpad for a while?
[00:28] <randomusername> or openid
[00:35] <wgrant> randomusername: So, I can't obviously see what's wrong.
[00:36] <wgrant> randomusername: You gave a reference above to a support case -- is that from the link on login.launchpad.net?
[00:36] <wgrant> If so, I suggest waiting for a response from them.
[00:36] <randomusername> Case #00012729 has been created and a support analyst will get back to you within the next 24 hours.
[00:37] <randomusername> got this two days ago on an email
[00:38] <wgrant> Can you add to that case the information we've established here? Your new and old usernames, the fact that your old one OOPSes on production with OOPS-1714K1846 but is fine on edge, and anything else that seems relevant?
[00:38] <ubot5`> https://lp-oops.canonical.com/oops.py/?oopsid=1714K1846
[00:39] <randomusername> wgrant: it doesn't just oops with -1714K1846
[00:39] <wgrant> Well, it will get a different code each time.
[00:40] <randomusername> ah, that I did not know
[00:40] <wgrant> 1714 identifies the day, K identifies the server, and 1846 is just a sequence number for that server today.
[00:40] <randomusername> where do I have to log in to insert what you posted above?
[00:40] <wgrant> I have no idea, sorry. I've never used ISD's support system.
[00:40] <wgrant> Maybe try replying to the email?
[00:41] <randomusername> reply -> noreply@canonical.com
[00:41] <randomusername> nogo
[00:41] <randomusername> so I'll just have to wait for an answer
[00:47] <randomusername> wgrant: gonna forward this to feedback@launchpad.net
[00:48] <randomusername> and hope for the better
[00:48] <wgrant> randomusername: Sounds like a reasonable idea.
[00:48] <wgrant> randomusername: Make sure you include all the information I suggested.
[00:49] <randomusername> all done
[00:51] <randomusername> info sent
[00:51] <randomusername> thanks for the help wgrant
[00:52] <wgrant> randomusername: Great. Hopefully you'll get a response soon.
[00:52] <wgrant> But maybe not until Monday.
[00:52] <randomusername> it's fine
[00:53] <randomusername> thanks again
[01:26] <yofel> is something wrong with the ppa buildds?  since a while ago I see it frequently that builds start delayed, like https://edge.launchpad.net/~yofel/+archive/ppa/+build/1954695 which is currently building for 15min whithout the buildd actually doing anything. Before that it tried to build on another one and was simply reset to 'needs building' after a while
[01:28] <wgrant> yofel: Known scaling issue. There have been a few fixes in the last couple of months, and the last bit might be happening next week.
[01:28] <yofel> k, thanks
[01:58]  * maxb looks askance at the kdepim code import which has now been running for 24 hours
[01:59] <maxb> Just how slow a machine is galapagos anyway? :-/
[02:01] <wgrant> maxb: The name might be a good hint.
[02:02] <wgrant> (There is a Galapagos Penguin. Penguin species were Canonical's first machine naming convention.)
[02:05] <maxb> Meanwhile the import on neumayer looks like it's got another day of "determining revprop revisions" before it starts fetching anything
[02:07] <maxb> I'm seriously tempted to keep poking it until it restarts on pear
[02:07] <wgrant> I think throughput would be increased if the russkaya/neumayer/galapagos were just disabled.
[02:14] <maxb> it would be entirely useful if ~vcs-imports still let you quiesce importds :-(
[02:27] <maxb> *right* having chased lp:kdepim onto pear, maybe it now has a hope of actually importing
[02:30] <wgrant> maxb: How did you manage that?
[05:07] <maxb> wgrant: bombarding the importds with requests to import other kde branches to trigger a sqlite locking fail, to get it off the ones I didn't want, then refreshing +code-imports/+machines a lot and choosing an opportune moment to retry
[05:21] <wgrant> maxb: Ahh, I forgot you could get them to fail like that.
[05:21] <wgrant> Handy.
[07:47] <MTecknology>  Try reloading this page in a minute or two. If the problem persists, let us know in the #launchpad IRC channel on Freenode.
[07:47] <MTecknology> :(
[07:47] <MTecknology> wgrant: you guys doing something?
[07:48] <MTecknology> you guys have two different try again pages?
[07:49] <wgrant> MTecknology: Which URL?
[07:49] <MTecknology> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+filebug/5197f7d4-bd6f-11df-a895-0019bbccd75c
[07:49] <wgrant> Ah.
[07:49] <MTecknology> I'm trying to file an apport bug
[07:49] <wgrant> Known issue.
[07:49] <wgrant> Keep trying.
[07:49] <MTecknology> oh
[07:49] <wgrant> It should work eventually.
[07:49] <wgrant> Possibly a firewall issue.
[07:49] <MTecknology> what's the issue with it?
[07:49] <MTecknology> on my side?
[07:50] <wgrant> No.
[07:50] <wgrant> Launchpad-side.
[07:50] <wgrant> Although the fact it's always kernel bugs may put an end to the firewall hypothesis.
[07:50] <MTecknology> lol.. I was going to file another kernel bug right after
[07:50] <MTecknology> any chance it's because of the absolutely massive file upload size?
[07:51] <MTecknology> this was 11.7MB
[07:51] <wgrant> I suspected that.
[07:51] <wgrant> But maybe not.
[07:51] <MTecknology> :P
[07:52] <MTecknology> *sniffle*
[07:52] <wgrant> Could you stop trying to file it for a sec, while I poke around?
[07:52] <MTecknology> sure
[07:54] <wgrant> :(
[07:54] <wgrant> The data from it is trivially small.
[07:55] <wgrant> So the error occurs after you enter the summary and description, and click the final submission button?
[07:56] <wgrant> MTecknology: ^^
[07:56] <MTecknology> ya
[07:56] <MTecknology> I get two different 'Please try again' pages too - one centered and one off to the left
[07:57] <wgrant> Interesting.
[08:01] <MTecknology> wgrant: I don't like this :( make it better so I do this -> :) <- instead
[08:02] <wgrant> Aha.
[08:02] <wgrant> Reproduced on staging.
[08:02] <wgrant> WTFery++
[08:02] <MTecknology> uh?
[08:02] <MTecknology> oh... disable redirect and pop off edge?
[08:03] <wgrant> No.
[08:03] <wgrant> Production is broken too.
[08:03] <MTecknology> ok, so you're playing?
[08:03] <wgrant> I am.
[08:04] <MTecknology> WTFery... I had to look that up on urban dictionary -_-
[08:04] <wgrant> Heh.
[08:06] <MTecknology> wgrant: should I keep trying it?
[08:06] <wgrant> MTecknology: I'd say so.
[08:06] <wgrant> Others have said that it works eventually, and it's easy enough for us to reproduce.
[08:08] <MTecknology> If I go to sleep and save the page, maybe I could try in the morning too?
[08:08] <wgrant> I'm not sure how long the token will last.
[08:13] <MTecknology> lifeless: You happen to know if I could retry in the morning?
[08:14] <MTecknology> maybe I'll give up and just file the bug myself - might be easier
[08:14] <lifeless> I don't know the delete frequency for unused attachments
[08:14] <lifeless> probablt a week or so
[08:14] <lifeless> AFACT this is just slow code
[08:15] <wgrant> But slow code shouldn't cause a timeout like this... unless it has lots and lots of non-SQL seconds.
[08:15] <lifeless> SQL time: 13812 ms
[08:15] <lifeless> Non-sql time: 345 ms
[08:15] <lifeless> Total time: 14157 ms
[08:15] <lifeless> Statement Count: 42
[08:15] <MTecknology> is that what you were talking about in -dev?
[08:15] <MTecknology> I'm too tired to comprehend it..
[08:16] <lifeless> wgrant: LaunchpadTimeoutError: Statement: 'SELECT BinaryPackagePublishingHistory.archive, BinaryPackagePublishingHistory.binarypackagerelease, BinaryPackagePublishingHistory.component, BinaryPackagePublishingHistory.datecreated,
[08:16] <lifeless> s305
[08:16] <lifeless> there is three seconds between
[08:16] <lifeless> SELECT SourcePackageName.id, SourcePackageName.name FROM SourcePackageName WHERE SourcePackageName.name = %s
[08:16] <lifeless> and
[08:16] <wgrant> lifeless: That probably isn't it. If it was an SQL timeout, the appserver would have been killed and responded in time.
[08:16] <lifeless> SELECT DISTINCT SourcePackagePublishingHistory.archive, SourcePackagePublishingHistory.component, SourcePackagePublishingHistory.datecreated, SourcePackagePublishingHistory.datemadepending, SourcePackagePublishingHistory.datepublished, SourcePackagePublishingHistory.dateremoved,
[08:17] <lifeless> note that staging runs at loads of 6 or more
[08:17] <lifeless> regularly
[08:17] <wgrant> ........ ew.
[08:17] <lifeless> timeouts there currently mean almost nothing
[08:17] <wgrant> Well, yes, I don't care about the real timeouts that I was able to see. I care about the request that hung past the proxy's threshold.
[08:17] <lifeless> I spent a day chasing 10 seconds in storm that is 1.4 seconds under normal load
[08:18] <wgrant> Which was probably *not* an SQL timeout.
[08:18] <lifeless> sure
[08:18] <lifeless> the ones I saw were though
[08:18] <wgrant> Hmmm.
[08:19] <MTecknology> aight.. I'm going to give up and file it myself in the morning
[08:19] <MTecknology> thanks for looking at it
[08:19] <MTecknology> g'night
[08:19] <MTecknology> or morning..
[08:19] <wgrant> Heh. Night.
[08:19] <MTecknology> acky..
[08:20] <MTecknology> I lost the description I wanted to send with
[08:21] <lifeless> sorry
[08:21] <lifeless> will be looking this monday
[08:21] <lifeless> as a priority
[08:21] <wgrant> Is the staging timeout 10s?
[08:21] <lifeless> yes
[08:23] <wgrant> Iiinteresting.
[08:23] <wgrant> Normal timeouts take 13s.
[08:23] <wgrant> The proxy timeout comes back after slightly over 10s.
[08:25] <wgrant> So there's something even more strange going on.
[08:25] <wgrant> And it's still intermittent, despite there only being one server.
[08:34] <lifeless> we need to talk to spm to get details on the trigger for that page
[09:04] <pmiller> I am trying to upload a source package to a ppa, without success... https://launchpad.net/~pmiller-opensource/+archive/ppa
[09:04] <wgrant> pmiller: What does it do when you try?
[09:05] <pmiller> wgrant: nothing happens.  The upload completes, but the packages don't appear.
[09:07] <wgrant> pmiller: What are the commands that you're using to build and upload the source package?
[09:08] <pmiller> Build: debuild -S -sa or debuild -S -sd, as per https://help.launchpad.net/Packaging/PPA/BuildingASourcePackage, neither seems to make a difference.  Then cd ..; dput ppa:pmiller-opensource/ppa *.changes
[09:09] <wgrant> pmiller: Is the changes file signed with 0xD0EDB64D?
[09:09] <pmiller> yes.
[09:11] <wgrant> pmiller: Can you upload the signed package somewhere that I can examine it?
[09:24] <pmiller> wgrant: http://libexplain.sourceforge.net/debian/x/
[09:27] <wgrant> pmiller: That sends a rejection email when I process it locally.
[09:27] <wgrant> Do you not have such an email?
[09:27] <wgrant> The signature goes through fine, so you should.
[09:27] <pmiller> wgrant: no email
[09:28] <wgrant> Well, let's see what happens if you fix the package.
[09:28] <wgrant> 2010-09-11 08:26:36 WARNING 	Unable to find distroseries: unstable
[09:29] <wgrant> You need to have an Ubuntu series name in the changelog, rather than a Debian one.
[09:29] <wgrant> Maybe the email is getting lost.
[09:32] <pmiller> wgrant: will try hacking changelog.  email for openpgp key update took hours, but it did arrive.  I've been trying to upload fro longer than that.
[09:32] <wgrant> pmiller: Hm, could you see where the delay was?
[09:34] <pmiller> wgrant: X-greylist: delayed 2431 seconds by postgrey-1.31 at mail.ivt.com.au; Sat, 11 Sep 2010 16:00:34 EST (I know the admin, I aks him about it)
[09:37] <wgrant> Ah, IVT.
[09:37] <wgrant> Interesting.
[09:39] <pmiller> wgrant: ooh, look!  email.
[09:40] <pmiller> wgrant: for "distroseries" it wants "lucid"?   (I usually upload to Debian, as you may have guessed.)
[09:40] <wgrant> pmiller: Yep.
[10:11] <pmiller> wgrant: most recent upload was accepted "currently building" yay
[10:14] <pmiller> wgrant: is there a way to request automatic backports, or do I have to make multiple uploads, one for each distroseries?
[10:39] <wgrant> pmiller: You can't request automatic backports. However, you can build on one series and copy the source and binaries to another.
[10:39] <wgrant> eg. lots of people build on hardy, then copy up to newer series.
[10:41] <pmiller> so long as there aren't library versioning issues, yes.
[10:42] <pmiller> of course, libexplain underlies some of the other projects I want in the ppa, so I'm acutely aware of library version issues.
[10:42] <wgrant> Right.
[10:42] <wgrant> If you want multiple sets of binaries, you need multiple uploads.
[10:43] <lifeless> wgrant: don't recipes provide a build-for-N releases?
[10:43] <wgrant> This is true.
[10:43] <pmiller> that's what I thought.  It just gripes me that they will differ by only one word: s/lucid/whatever/
[10:44] <pmiller> recipes are predicated on bzr branches, and despite lifeless' best efforts, I don't use bzr
[11:13] <lifeless> pmiller: export a tarball; bzr import that onto a branch; push
[11:13] <lifeless> pmiller: not trying to move you to bzr to do this :P
[11:17] <Ddorda> hey guys
[11:18] <Ddorda> is it a known bug that launchpad mailing lists can't read non-latin languages?
[13:04] <philip_stoev> guys, the search box for finding users when trying to assign a bug to a specific user in the bug database does not work
[13:04] <philip_stoev> no matches are returned even when the actual user id or user name is used
[13:09] <wgrant> philip_stoev: Unless you are the bug supervisor of the project, it will only find teams that you administer.
[13:09] <wgrant> philip_stoev: If you are a bug supervisor, it should find anyone.
[13:35] <philip_stoev> wgrant: ok got it, thanks
[14:47] <fta> if any admin is here, please kill this build: https://edge.launchpad.net/~chromium-daily/+archive/ppa/+build/1952454 (stuck for 3 days)
[14:53] <bilalakhtar> fta: You're right. terranova builder is slow
[14:53] <bilalakhtar> I have also faced it once
[14:53] <wgrant> This isn't slow.
[14:53] <bilalakhtar> plymouth, which takes 10 minutes for a build, took 4 hours on it!
[14:53] <wgrant> Something's broken.
[14:54] <bilalakhtar> wgrant: what about my plymouth case?
[14:54] <wgrant> Hmm. That is odd.
[14:54] <wgrant> I could see up to half an hour or so.
[14:54] <wgrant> But not four hours.
[14:55] <bilalakhtar> wgrant: I saw, towards the end, dpkg-deb was taking 15 minutes to just archive a minary!
[14:55] <bilalakhtar> *binary
[14:55] <bilalakhtar> And there were 8 binaries by that source
[14:55] <wgrant> The log can take up to 15 minutes to upate.
[14:55] <wgrant> You can't rely on it for timings like that.
[14:56] <bilalakhtar> wgrant: really? but still, 4 hours was too long! And the build succeeded, nevertheles
[14:56] <bilalakhtar> s
[14:56] <wgrant> Hmm. That's concerning.
[14:56] <bilalakhtar> then I deleted my PPA and created a new one, otherwise I would have shown you the logs
[15:26] <JoshBrown> The Launchpad mailing list footer is not removed when quoting: https://lists.launchpad.net/photostory-team/msg00005.html
[20:23] <jenkins> how do I release a package for different versions of ubuntu in the same ppa? I have tired another entry in the change log  (changing maverick to lucid) but that does not help
[20:27] <JoshBrown> jenkins: I don't use PPAs that much personally but I've heard there's a built-in Launchpad function to do this (unless the package needs to be different in different Ubuntu versions)
[20:28] <jenkins> I did have a play with the lauchpad build function but I decided i like the the dput wap
[23:02] <MTecknology> LP is making me cry :(
[23:07] <lifeless> the filebug thing?
[23:07] <lifeless> will be looking at that tomorrow; need LOSA support
[23:15] <penguin42> LOSA?
[23:15] <lifeless> launchpad operational sysadmin
[23:16] <penguin42> ah, guys with hamers
[23:32]  * JoshBrown says HAMMER TIME