text
stringlengths
15
499
[07:03] <lifeless> jtv: brand new world
[07:03] <jtv> lifeless: oh, I wasn't aware of that one. I just type "paste" and let my browser figure out the rest.
[07:04] <lifeless> sorry, paste.ubuntu.com
[07:04] <lifeless> well, open :)
[07:04] <lifeless> so, something is holding a lock on the branch
[07:04] <jtv> lifeless: I am.
[07:05] <lifeless> oh actually
[07:05] <jtv> A write lock. And when I unlock it, boom.
[07:05] <lifeless> that message says 'the lock was broken while open'
[07:05] <lifeless> so perhaps something deleted the branch?
[07:05] <lifeless> its in your teardown
[07:05] <lifeless> is it possible you're causing 'rm -rf', 'branch.unlock()' to happen, in that order?
[07:06] <henninge> spm: I'd like to put staging into read-only mode for a few minutes.
[07:06] <spm> henninge: ? why not just use the slave?
[07:06] <henninge> spiv: This is a patch that does that, can you apply it please and restart staging? https://pastebin.canonical.com/20128/
[07:06] <beuno> already on it's way to digg: http://digg.com/linux_unix/Canonical_releases_complete_source_code_for_Launchpad
[07:07] <henninge> spm: No, the fix I am QA'ing checks that configuration directive.
[07:07] <henninge> spm: The bug is specifically about read-only mode.
[07:08] <henninge> bug 380852
[07:08] <jtv> lifeless: it's the first teardown I do, afaik. I could try doing the unlock explicitly at an earlier stage; hang on.
[07:08] <henninge> bug #380852
[07:08] <ubot3> Malone bug 380852 in rosetta "translation page shows editable fields while in read only mode" [Medium,Fix committed] https://launchpad.net/bugs/380852
[07:08] <henninge> ah!
[07:09] <henninge> spm: is the restart still such a pain with buildout?
[07:10] <jtv> lifeless: you're right, it happens in the teardown but not if I unlock before then.
[07:10] <jtv> lifeless: maybe useBzrBranches puts its cleanup in the wrong place?
[07:11] <spm> henninge: patched and restarted
[07:12] <henninge> spm: great!
[07:12] <lifeless> jtv: use addCleanup rather than tearDown
[07:13] <jtv> lifeless: already made that change in an as-yet unlanded branch, though it wasn't enough to fix the problem.
[07:16] <henninge> spm: thank you very much! Everything works as expected.
[07:16] <spm> henninge: revert?
[07:16] <henninge> spm: Can you just revert or would you like a patch?
[07:16] <spm> revert is fine
[07:16] <henninge> spm: ;-)
[07:16] <henninge> spm: please do
[07:17] * henninge is blazing through the test plan here ...
[07:17] <spm> henninge: is back
[07:17] <henninge> spm: cool, that was a great help!
[07:18] <VK7HSE> just in the throws of setting up a local copy of LP here... \o/
[07:19] <jtv> VK7HSE: I hope that's "throes," not "exception throws" :)
[07:19] <VK7HSE> yeah something like that !!! ;)
[07:20] <\sh> what did I read? LP is opensourced? Congrats to everybody involved...good job :)
[07:21] <jtv> I guess the word is out then
[07:21] <VK7HSE> https://dev.launchpad.net/Getting
=== beuno changed the topic of #launchpad-dev to: Launchpad Development Channel | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: http://identi.ca/notice/6403823
=== beuno changed the topic of #launchpad-dev to: Launchpad Development Channel | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: https://dev.launchpad.net/Getting
[07:23] <beuno> COPYNPASTE FAIL
[07:23] <beuno> should go to bed
[07:25] <mwhudson> failures in buildbot :(
=== al-maisan_ is now known as al-maisan
[07:26] <mwhudson> all seemingly to do with uncollectable garbage :(
[07:26] <jml> mwhudson, actually no.
[07:27] <mwhudson> no?
[07:27] <jml> mwhudson, one of them is to do with the move of bzr to buildout.
[07:27] <mwhudson> jml: oh "yay" ?
[07:27] <jml> mwhudson, it ought to be fixed when my branch is finished running through ec2test.
[07:27] <mwhudson> jml: oh ok
[07:28] <jml> except that's going to db-devel
[07:28] <jml> hmm.
[07:30] <henninge> jtv: any clue how to best QA bug #388473 ?
[07:30] <ubot3> Malone bug 388473 in rosetta "POTMsgSets on page not in ascending sequence order" [High,Fix committed] https://launchpad.net/bugs/388473
=== lifeless changed the topic of #launchpad-dev to: Launchpad Development Channel | https://dev.launchpad.net/ | Please use #launchpad for support. | https://launchpad.net/~launchpad-dev | Get it: https://dev.launchpad.net/Getting | http://paste.ubuntu.com/
[07:31] <jtv> henninge: in a case like this I'd just make sure that the page works properly, and let the oops reports show that the bug is gone.
[07:31] <henninge> jtv: cool, can do that ;)
[07:32] <jml> mwhudson, and hey, that just finished with ec2test. I'll retarget the branch for devel & land appropriately
[07:32] <kfogel> mwhudson, jml: yeah, somewhere we need to explain our four trunks thing, in the dev wiki...
[07:32] <thekorn> congrats everybody!
[07:32] <kfogel> thekorn: hey there! Thank you. Enjoy :-).
[07:32] <jml> kfogel, yes. but not today :)
[07:32] <kfogel> jml: indeed
[07:36] <emilis_info> how can I remove all changes introduced by rocketfuel-setup?
[07:37] <emilis_info> I didn't expect it to mess up my Apache+PHP setup
[07:39] <jml> emilis_info, good question!
[07:39] <thekorn> wow, and soyus and codehosting too, awesome
[07:40] <thekorn> you must have a good PR manager, nice bluff ;)
[07:40] <jml> emilis_info, there's isn't an automatic way of doing it, probably best to look at the script itself.
[07:40] <emilis_info> heh
[07:40] <LaserJock> thekorn: that's what I was thinking ;-)
[07:41] <emilis_info> I think you should recommend installing it on a separate / virtual machine, because it touches so many places it can break something for a lot of people
[07:44] <noodles775> Morning
[07:44] <jtv> hi noodles775!
[07:45] <noodles775> Hiya jtv, gee you've been around early and late lately? QA to finish?
[07:45] <henninge> jtv: even better, there is a link in the sequence bug that fails on production but succeeds on edge - QAed!
[07:46] <jtv> noodles775: yeah, QA's been a real bottleneck for the past few months. Finally got a full end-to-end QA on production though, so pretty cool.
[07:46] <jtv> noodles775: I've worked about 2.5 working days so far this week. :)
[07:46] <noodles775> Sheesh
[07:46] <jtv> henninge: perfect.
[07:47] <jtv> henninge: that reminds me... I do _not_ plan to do much more today. I finished at 5 AM and got back to work at 11 AM, and the most crucial thing is done.
[07:47] <henninge> jtv: you are tough!
[07:47] <jtv> henninge: would you mind having a very quick standup soon and then doing without me?
[07:47] <jtv> henninge: I'm stupid.
[07:47] <henninge> jtv: what is the crucial thing?
[07:47] <henninge> jtv: I don't mind at all!
[07:48] <jtv> henninge: there was yet another bzr hiccup. Tickling yet another part of bzrlib that Launchpad hadn't touched yet.
[07:49] <henninge> oh, bzrlib
[07:51] <jtv> yes, when you reference a directory you need to look up or create an id for it. And first you need to do the same for its parents.
[07:52] <jtv> But the lookup doesn't work for directories you just created, only for ones that were already in the committed branch.
[07:52] <henninge> spm: how much longer will you be around?
[07:52] <jtv> So if you had n files in the same directory d, where d was not in the committed branch, I'd create d n times.