[00:27] <wgrant> Ah, crap.
[00:29] <lifeless> where?
[00:30] <wgrant> Bug 575426 is bad.
[00:30] <mup> Bug #575426: copy-package broken <Soyuz:New> <https://launchpad.net/bugs/575426>
[00:35] <wgrant> I wish our data model didn't make reasonable things impossible :(
[00:36] <wgrant> Hm, and I imagine that check is going to be very slow.
[00:40] <maxb> copying packages has not fared well this rollout
[00:40] <maxb> see also bug 575450 bug 574552 :-(
[00:40] <mup> Bug #575450: +copy-packages nearly unusable due to timeouts <Soyuz:New> <https://launchpad.net/bugs/575450>
[00:40] <mup> Bug #574552: "different contents" error when unembargoing security update or copying packages <Soyuz:In Progress by stevenk> <https://launchpad.net/bugs/574552>
[00:41] <wgrant> Yeah, the fix for 574552 caused 575426
[00:41] <wgrant> It's a bit unfortunate that all the hashes are expired along with the actual file content.
[00:58] <wgrant> Can't something be done about the scanning delays? Like, say, restaggering all maverick branches?
[00:59] <wgrant> This is not actually an awesome way to get people to hate you less.
[08:10] <adeuring> good morning
[09:04] <mrevell> Helllo
[09:27] <jml> mrevell, hello :)
[09:28] <mrevell> Howdy jml. Don't forget to take blankets, a torch and some water, in case you get stuck in the tunnel.
[09:29] <jml> mrevell, good point!
[10:33] <henninge> What does it mean when I get this error message while trying Branch.getBzrBranch() locally?
[10:33] <henninge> UnsupportedProtocol: Unsupported protocol for url "lp-internal:///~stevea/thunderbird/main"
[10:33] <mwhudson> henninge: you haven't started a server
[10:33] <mwhudson> server = get_ro_server()
[10:34] <mwhudson> server.start_server()
[10:34] <mwhudson> try:
[10:34] <mwhudson>    ...
[10:34] <mwhudson> finally:
[10:34] <mwhudson>     server.stop_server()
[10:34] <henninge> mwhudson: thanks
[10:34] <henninge> mwhudson: get_ro_server is imported from where? (Saves me the grep)
[10:35] <henninge> I am running make harness, btw.
[10:35] <mwhudson> henninge: lp.codehosting.vfs
[10:35] <henninge> cheers
[10:35] <mwhudson> henninge: there's get_rw_server too
[10:38] <henninge> mwhudson: NotBranchError: Not a branch: "lp-internal:///~stevea/thunderbird/main".
[10:38] <henninge> missing something else?
[10:38] <mwhudson> henninge: is that a sample data branch?
[10:38] <henninge> yes ;)
[10:38] <mwhudson> there's probably not actually any underlying branch data
[10:38] <henninge> ah, ok. I'll push one, then.
[10:39] <mwhudson> maybe ./utilities/make-dummy-hosted-branches is out of date now
[10:39] <mwhudson> yes, looks like it
[10:54] <henninge> mwhudson: yes, that is ... ;I)
[10:54] <henninge> mwhudson: but I was able to push to ~mark/+junk/testdoc
[10:55] <henninge> mwhudson: when I tell an admin to try this on  a production server, will they need to start the server as well?
[10:56] <henninge> I'd think not, as it would access the production server.
[10:58] <mwhudson> henninge: no, you will
[10:58] <mwhudson> 'server' is perhaps a strange term
[10:58] <mwhudson> it's the thing that knows how to handle the lp-internal:/// URL scheme
[10:58] <henninge> ok, so it's a local thing
[10:59] <henninge> mwhudson: so, this should work on production? http://paste.ubuntu.com/428201/
[10:59] <henninge> it worked locally ... ;)
[11:00] <henninge> _hasPotteryCompatibleSetup calls getBzrBranch
[11:03] <mwhudson> oh ok :)
[11:03] <mwhudson> then yeah, that should be ok
[12:46] <maxb> Is there a date set for re-rollout yet? Not being able to copy packages between PPAs is rather awkward
[12:53] <noodles775> maxb: 14UTC is the info I've got, but afaik it won't include the fix for 575426 - StevenK has a patch which he'll get CP'd asap.
[12:54] <maxb> bug 575426
[12:54] <mup> Bug #575426: SHA1-based copy checking breaks when there are expired sources in the target <Soyuz:Triaged by stevenk> <https://launchpad.net/bugs/575426>
[12:55] <wgrant> And that still won't fix the inefficiency, AIUI.
[12:55] <maxb> ugh
[12:55] <wgrant> So timeouts will still occur for dailies.
[12:56] <wgrant> (the deciding factor is the number of source publications in the target archive with the same name)
[12:56] <maxb> Well, if it worked at all, I could while 1; do ./myscript; done until it works :-)
[12:59] <wgrant> But I suspect the extra load caused by that would finally let Launchpad fulfill its life-long dream, which it has been steadily recently: to service requests in greater than finite time.
[12:59] <noodles775> wgrant: afaiui, it'll not do the check if the source and dest. archives are the same... not sure if that's what you were refering to.
[12:59] <wgrant> Er, steadily approaching.
[13:05] <wgrant> noodles775: Right, but maxb needs inter-archive copies.
[13:12]  * StevenK waves his arms.
[13:12] <leonardr> bac, maybe you can address https://answers.edge.launchpad.net/launchpad-registry/+question/109701 or redirect it to someone more knowledgable?
[13:16] <leonardr> bac: i'm also wondering who has responsibility for scripts (bug 575615 and bug 575611). is it foundations?
[13:16] <mup> Bug #575615: scripts/gina.py is not logging to scriptactivity <Launchpad itself:New> <https://launchpad.net/bugs/575615>
[13:16] <mup> Bug #575611: cache-country-mirrors.py is not logging to scriptactivity <Launchpad itself:New> <https://launchpad.net/bugs/575611>
[13:16] <StevenK> The first is soyuz
[13:16] <StevenK> The second I'm not sure about
[13:16] <bac> leonardr: right, first is soyuz second is registry
[13:17] <leonardr> cool, thanks
[13:53] <wgrant> Is changing the project of a bug via AJAX meant to still hang forever?
[13:54] <beuno> wgrant, yes, still unfixed
[13:54] <beuno> deryck, knows about it
[13:55] <wgrant> But it's been like a year.
[13:55] <beuno> something about the URL changing
[13:55] <mars> BjornT, ping
[13:55] <poolie> hi wgrant, beuno, mars
[13:55] <wgrant> Hi poolie.
[13:55] <mars> hi poolie
[13:55] <beuno> heya poolie
[13:55] <mars> wgrant, :(
[13:56] <deryck> wgrant, a year is nothing.  There are 4 digit open bugs on Launchpad. :-)
[13:57] <wgrant> deryck: Yeah, but they're not generally obviously broken buttons that never worked :)
[13:57] <deryck> wgrant, depends on who you ask, doesn't it? :-)
[13:59] <beuno> deryck, while we're all complaining
[14:00] <beuno> millions of bugs are not filed per day
[14:00] <beuno> due to the person picker re-directing after you select
[14:00] <beuno> so you loose everything you wrote
[14:00] <beuno> is that on your radar at all?
[14:01] <wgrant> There's a person picker in the bug-filing workflow?
[14:01] <wgrant> (apart from the broken assignee input for bug supevisors)
[14:01] <deryck> beuno, that is.  I have about 3-4 filebug bugs that I have on my list of 10-15 bugs that need looking at in the short term.
[14:01] <deryck> wgrant, the broken assignee picker is what beuno means, I believe
[14:01] <beuno> yes
[14:02] <beuno> and hi deryck  :)
[14:02] <deryck> beuno, hello :-)
[14:03] <mars> sinzui, ping
[14:03] <sinzui> hi mars
[14:13] <BjornT> hi mars
[14:59] <bac> reviewers meeting starting soon
[16:04] <sinzui> wgrant, ping
[16:05] <sinzui> jelmer, ping
[16:06] <jelmer> sinzui: pong
[17:45] <rockstar> jelmer, ping
[17:45] <jelmer> rockstar: hey
[17:46] <rockstar> jelmer, where are the unittests for IArchiveSet?
[17:47] <jelmer> rockstar: are you sure there are unit tests?
[17:48] <rockstar> jelmer, no.  I guess I should have said "Where are the tests for IArchiveSet?"
[17:48]  * rockstar whimpers
[17:48] <rockstar> Please don't be doctests, please don't be doctests...
[17:56] <mrevell> night all
[17:57] <intellectronica> rockstar: it's not a doctest, it's a codestory :)
[17:57] <rockstar> intellectronica, those kinds of stories give me nightmares...  :)
[17:57] <intellectronica> lol
[17:59] <rockstar> jelmer, I don't see a test at all for it.
[18:05] <rockstar> jelmer, is it alright if I create a unittest for it?
[18:42] <bryceh> I'm seeing an error "ImportError: No module named tickcount" when trying to do rocketfuel-branch
[18:42] <bryceh> http://pastebin.ubuntu.com/428444/
[18:42] <bryceh> I did a complete re-checkout from scratch on a fresh machine, re-ran rocketfuel-branch and got the same error
[18:43] <maxb> bryceh: This is interesting. I wish I could figure out why
[18:43] <bryceh> I checked and I do have python-tickcount installed
[18:44] <bryceh> I cal also 'import tickcount' just fine from the python prompt
[18:44] <bryceh> s/cal/can/
[18:44] <salgado> bryceh, can you do that on python2.5 as well?
[18:44] <bryceh> bdmurray, did you run into this too?  ^^  Any tips?
[18:44] <maxb> bryceh: The difference will be if you use python2.5
[18:44] <bryceh> Python 2.6.5 (r265:79063, Apr 16 2010, 13:09:56)
[18:44] <maxb> Please could you check what version of python-support you have installed?
[18:45] <bryceh> yeah forgot to mention, both systems were lucid
[18:45] <bryceh> python-support:
[18:45] <bryceh>   Installed: 1.0.4ubuntu1
[18:45] <bryceh>   Candidate: 1.0.4ubuntu1launchpad2
[18:45] <bdmurray> bryceh: it's been quite some time since I set it up and I'm using a karmic virtual machine
[18:46] <maxb> bryceh: that'll be the problem
[18:46] <maxb> So the question is why don't you have the version from the launchpad ppa
[18:46] <bryceh> I'll do a dist-upgrade
[18:47] <bryceh> rocketfuel-setup looked like it completed successfully, not sure why it would have left that un-upgraded
[18:51] <maxb> bryceh: I think this is an initial-setup corner case. Established developers are so used to being up-to-date with the PPA, so the relevant dependency to enforce this never got added to meta-lp-deps
[18:52] <bryceh> ok, looks like that worked.  Thanks!
[19:13] <bryceh> hmm, alright I've done 'make run' and tried loading http://launchpad.dev.  It prompts me to confirm the certificate but then I get a 503 error
[19:13] <bryceh> [Wed May 05 11:11:49 2010] [error] (111)Connection refused: proxy: HTTP: attempt to connect to 127.0.0.1:8086 (localhost) failed
[19:13] <bryceh> [Wed May 05 11:11:49 2010] [error] ap_proxy_connect_backend disabling worker for (localhost)
[20:07] <bryceh> whoohoo got it
[20:37]  * mtaylor got some feedback from someone today that they were trying to find code on launchpad for something, and when they finally found a loggerhead view, they freaked out that the launchpad skin/branding/nav went away
[21:51] <jelmer> rockstar: sorry, was out for dinner
[21:51] <jelmer> rockstar: I think we would have serious concerns about having our code properly unit tested.
[21:52] <rockstar> jelmer, I found some integration tests, but you seem to lack unit tests in general.
[21:53] <jelmer> rockstar: We're usually pretty good about adding them for new code, but there's a lot of existing code that lacks tests.
[21:54] <thumper> mtaylor: that's a bit sad
[21:54] <thumper> mtaylor: I'd love to integrate more
[21:54] <thumper> mtaylor: but we need to make loggerhead scale better first
[21:57] <mtaylor> thumper: put it in the cloud
[21:57] <mtaylor> thumper: doesn't that make everything scale better magically?
[21:58] <thumper> mtaylor: geez, why didn't I think of that?
[21:58] <thumper> mtaylor: we are getting our own private cloud :)
[21:58] <thumper> mtaylor: sugar lined and everything
[21:58] <mtaylor> speaking of clouds... when are you guys going to get a cloud in the us so I don't have to deal with transatlantic lag all the time?
[22:00] <thumper> it seems like we could maybe have a read only slave over there...
[22:00] <thumper> with some app servers
[22:00] <thumper> multi-write masters seem to be complex :)
[22:00] <mtaylor> well sure - if you're using postgres :)
[22:00] <mtaylor> we'll hook you up with some spiffy new drizzle servers with jay's new replication stuff
[22:00] <thumper> and easy for dizzle yes?
[22:01] <mtaylor> of course!
[22:01] <mtaylor> we're a database for the cloud after all :)
[23:12] <wgrant> sinzui: Hi.
[23:49] <wgrant> LOSAs: Is buildd-manager running?
[23:49] <wgrant> It looks like it has been dead for perhaps seven hours.