[05:03] <lifeless> hmm, no stub yet
[05:03] <lifeless> this is annoying
[05:06] <jamesh> lifeless: did you disable the problem test?
[05:07] <lifeless> jamesh: yes but the librariangc tests are hanging
[05:07] <jamesh> ah
[05:07] <lifeless> and as stub wrote that end to end I want to walk through it with him rather than guessing
[05:25] <lifeless> has ubuntu main moved to malone yet ?
[05:26] <jamesh> the bugzilla migration hasn't occurred yet
[05:26] <jamesh> but they are accepting bugs in either, iirc
[05:26] <lifeless> sweet
[05:27] <lifeless> fink is a 'distro' :)
[05:35] <jamesh> lifeless: what do you think about how to handle the gpgme issue?
[05:35] <jamesh> or more specifically, the pyme problem
[05:35] <lifeless> pyme blows in so many ways
[05:35] <lifeless> terrible module name
[05:35] <lifeless> limited functionality
[05:35] <lifeless> -swig-
[05:36] <lifeless> I think that short term just fix-what-we-have
[05:37] <lifeless> and that we could start a gpgme python module using handcrafted or pyrex bindings
[06:30] <lifeless> spiv: ping
[06:30] <spiv> lifeless: pong
[06:30] <lifeless> your review queue is saturated AFAICT
[06:31] <spiv> Heh, the duplicate doesn't help ;)
[06:31] <lifeless> 3 in needs-review
[06:31] <lifeless> ok, 2 is alreight :)
[08:22] <SteveA> spiv: ping
[08:29] <SteveA> jamesh: ?
[08:37] <jamesh> SteveA: yeah?
[08:38] <SteveA> jamesh: up for a phone call?
[08:38] <SteveA> or even voip
[08:39] <jamesh> I don't have voip set up
[08:39] <jamesh> but sure
[08:39] <SteveA> ok
[09:02] <jamesh> lifeless: got time to test another potential fix for the pyme bug?
[09:06] <jamesh> https://chinstrap.ubuntu.com/~dsilvers/paste/fileZgK1z9.html <- a minimal patch that might fix it
[09:09] <lifeless> jamesh: waiting for stub
[09:09] <lifeless> jamesh: will test after that
[09:09] <jamesh> lifeless: okay.
[09:26] <lifeless> spiv: ping
[09:41] <stub> Well.. I think that window is dead for a while longer
[09:42] <stub> lifeless: Your paste is still pasting, one slow line at a time. You got a patch to that file?
[09:42] <stub> c/patch/path/
[09:43] <lifeless> stub: sorry, no.
[09:43] <lifeless> uhm, I might be able to make one
[09:43] <lifeless> yes
[09:43] <lifeless> fotxt
[09:57] <SteveA> stub: ping?
[10:02] <SteveA> stub: ping?
[10:03] <lifeless> stub: hes looking at some logs 
[10:03] <lifeless> SteveA: ^^
[10:03] <SteveA> okay
[10:04] <SteveA> stub: i'd like to do a voip call sometime today
[10:04] <lifeless> jamesh: no fixy
[10:04] <lifeless> jamesh: how can I tell I applied it correctly ?
[10:11] <stub> SteveA: I'm online
[10:12] <jamesh> lifeless: check to see if _wrap_delete__gpgme_key() in gpgme_wrap.c calls delete__gpgme_key() rather than free()
[10:14] <lifeless> _wrap_delete_gpgme_key_t_p  ?
[10:14] <jamesh> no
[10:15] <lifeless> there is no _wrap_delete__gpgme_key
[10:15] <lifeless> oh, hmm there is
[10:15] <jamesh> double underscore
[10:15] <lifeless> delete__gpgme_key(arg1);
[10:15] <jamesh> is there also a _wrap_delete__gpgme_key()?
[10:15] <jamesh> yeah.
[10:16] <lifeless> yes
[10:17] <jamesh> okay.  That should get rid of the chance of pyme calling free() on gpgme_key's
[10:18] <lucasvo> hi, Is it a known bug, that if table get too long they are placed over the right column?
[10:18] <lucasvo> see: https://launchpad.net/distros/ubuntu/dapper/+source/fwbuilder/+pots/review-dapper-fwbuilder-1/de/+translate
[10:20] <jamesh> lucasvo: I don't see a bug report for it, but it is probably a side effect of the new CSS based column layout
[10:21] <jamesh> lucasvo: please file a bug
[10:21] <lucasvo> jamesh: ok, I will do it
[10:26] <lucasvo> Bug #5831
[10:26] <Ubugtu> Malone bug #5831: New CSS design places long lines in the content over the right column In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/5831
[10:26] <jamesh> lucasvo: you can reproduce it on a lot of pages by upping the font size sufficiently
[10:26] <jamesh> but it is more obvious on those translation pages
[10:27] <lucasvo> jamesh: one can not work like this 
[10:28] <jamesh> lucasvo: does it make the page unusable, or just inconvenient?
[10:28] <lucasvo> it doesn't makes it unusable but very strange
[10:29] <lucasvo> if you try hard you can read 
[10:29] <lucasvo> but it is not easy
[10:29] <jamesh> lucasvo: I'm not saying it shouldn't be fixed; just trying to work out signficant it is
[10:29] <jamesh> ctrl+- might help (but makes things harder to read)
[10:29] <lucasvo> I could try out and submit a fixed css
[10:29] <lucasvo> yeah
[10:30] <sivang> morning all
[10:32] <lifeless> jamesh: if its not easy to read, people wont try
[10:32] <lifeless> jamesh: different from ugly
[10:37] <jamesh> lifeless: I was offering a work around (make the fonts smaller).  Nothing more than that
[10:38] <jamesh> I agree that it should be fixed
[10:38] <lifeless> jamesh: cool. same page then
[10:43] <carlos> morning
[10:43] <lifeless> jamesh: so - disable the tests, or do you want another backtracec ?
[10:44] <lifeless> jamesh: (assuming it may have changed)
[10:44] <jamesh> lifeless: if it's easy for you to get another backtrace, it would be useful
[10:45] <jamesh> (if it has changed, that would be interesting)
[10:45] <lifeless> first off, a valgrind run
[10:45] <jamesh> but disabling the test is probably the right decision at this point
[10:45] <lifeless> running
[10:58] <lifeless> .win goto jamesh
[10:59] <jamesh> same stack traces
[10:59] <jamesh> thanks
[11:03] <lifeless> ok
[11:03] <lifeless> so, free() of that was not the cause
[11:04] <SteveA> spiv: around?
[11:05] <lifeless> the lifeless(tm) paging service is online, if needed
[11:08] <ddaa> hey
[11:08] <ddaa> jamesh: on a scale from 4 to 5, how busy are you nowadays?
[11:10] <jamesh> I've got a few things queued up, but may be able to fit some things in
[11:10] <ddaa> I was wondering whether you'd be able to fix the bug in the BranchNavigation when Branch.owner is a team.
[11:11] <ddaa> Since I'm obviously not going to be able to do it next week...
[11:11] <jamesh> ddaa: sure.  Assign the bug to me and I'll look at it
[11:11] <ddaa> weren't you the one who commented on it?
[11:11] <jamesh> yeah
[11:12] <ddaa> bug 5583
[11:12] <Ubugtu> Malone bug #5583: Branch link 404s In: launchpad (upstream), Severity: Normal, Assigned to: David Allouche, Status: New https://launchpad.net/bugs/5583
[11:12] <ddaa> okay, assigning that to you
[11:15] <ddaa> spiv: do you expect to be blocked by optional-branch-title before January 9th?
[11:19] <lifeless> hes having dinner
[11:19] <lifeless> courtesy the lifeless(tm) paging service
[11:21] <ddaa>  /msg lifeless_paging set spellcheck
[11:26] <lifeless> stub: pg 8 specific bug ? 
[11:26] <lifeless> stub: that test failed again
[11:26] <lifeless> stub: but it did not hang this time
[11:26] <lifeless> 1 and only one failure
[11:31] <lifeless> who here has seen intermittent failures on lib/canonical/launchpad/ftests/../pagetests/foaf/32-setpreferredemail.txt
[11:33] <ddaa> lifeless: the check_merge yesterday succeeded (except for a few failure in some completely unrelated areas) I'm pushing the bzn
[11:33] <lifeless> ddaa: thanks
[11:33] <ddaa> I'm pushing bzrsync fix to /home/warthogs/archives/david/launchpad/newbzr right now
[11:35] <lifeless>     +   Module canonical.launchpad.browser.person, line 1354, in emailFormSubmitted
[11:35] <lifeless>     +     self._setPreferred()
[11:35] <lifeless>     +   Module canonical.launchpad.browser.person, line 1508, in _setPreferred
[11:35] <lifeless>     +     assert emailaddress.person.id == self.context.id
[11:35] <lifeless>     + AssertionError
[11:35] <lifeless> thats the failure
[11:35] <lifeless> which suggests to me that context.id is probably None
[11:36] <SteveA> a newly created person?
[11:36] <lifeless> shouldn't be
[11:37] <lifeless> the page test is setting the preferred email on an existing person
[11:37] <lifeless> *I think* 
[11:37] <lifeless> post /people/no-priv/+editmails
[11:38] <lifeless> VALIDATED_SELECTED=foo%40baz.com&SET_PREFERRED+Set...
[11:38] <lifeless> looks like a correct post to me
[11:38] <lifeless> though I'm not sure about the url.
[11:38] <lifeless> how can we tell if its a new person ? Is there some marker ?
[11:39] <SteveA> i don't think the ids are None
[11:39] <SteveA> i think it's more likely that the ids differ
[11:39] <SteveA> and are non-none
[11:39] <SteveA> can the assert also give output as to what the ids are?
[11:39] <SteveA> that would help a lot
[11:40] <lifeless> would assert emailaddress.person.id == self.context.id, "differening ids in emailaddress.person.id(%d) == self.context.id(%d)" % (emailaddress.person.id, self.context.id)
[11:40] <lifeless> work ?
[11:40] <lifeless> I don't use the two value assert much
[11:41] <jamesh> lifeless: yes
[11:41] <jamesh> the second argument only gets evaluated if the first arg evaluates to True
[11:41] <lifeless> jamesh: yes that will work, or yes you have seen this error ?
[11:41] <jamesh> the assert will work
[11:42] <jamesh> lifeless: I've only seen the error in a PQM failure message
[11:42] <jamesh> (for a branch that merged successfully the second time)
[11:42] <lifeless> jamesh: garh unreliable tests garh
[11:43] <jamesh> so it would be good if we can get more information out of the next failure
[11:43] <lifeless> well
[11:43] <lifeless> its blocking pqm at the moment
[11:44] <lifeless> because I want to get a clean run before committing
[11:48] <bradb> stub: around?
[11:49] <stub> bradb: yo
[11:49] <bradb> stub: Just wanted to get clearance to land a branch that needs a small bit of data migration.
[11:49] <stub> bradb: This the XXX: comments one? Sure.
[11:50] <bradb> stub: This is the bug status changes.
[11:50] <bradb> Basically, Rejected is getting a different ID.
[11:50] <bradb> It's currently 40, but will become 17.
[11:50] <bradb> So, when this patch is rolled out into product, that update to BugTask.status will need to happen at the same time
[11:50] <bradb> .
[11:50] <bradb> s/product/production/
[11:51] <bradb> Will that be okay?
[11:51] <stub> ok. Please put a .sql script in database/schema/pending doing UPDATE BugTask SET status=17 WHERE status=40 and commit it at the same time
[11:52] <bradb> stub: sure, will do (I'd also note that there is no test data needing migration.)
[11:53] <stub> It this update to change the sort ordering of bugtasks?
[11:53] <bradb> yeah
[12:21] <lifeless> SteveA: AssertionError: differring ids in emailaddress.person.id(59) == self.context.id(52)
[12:22] <SteveA> next, what is the person.id when you run that test locally
[12:22] <lifeless> person.id where ?
[12:23] <bradb> 32-setpreferredemail.txt just failed for me locally. The error message sounds to me like the same thing you guys are discussing.
[12:23] <SteveA> lifeless: when you run the test locally, at the time of that assert statemet, there are two person references, which have the same id.  what is that id?
[12:24] <SteveA> i think what the assertion error is saying is, the code is trying to set a perferred email for someone, but that email address belongs to someone else
[12:25] <lifeless> SteveA: ECONTEXT. I really dont know what you mean. I don't see 'two person references'.
[12:25] <SteveA> emailaddress.person is one
[12:25] <SteveA> self.context is another
[12:25] <SteveA> they are both Person objects
[12:25] <lifeless> ok.
[12:25] <lifeless> i get thet
[12:25] <SteveA> the code is, i expect, trying to check that the owner of the email address is the same as the person who is trying to set it as their prefered
[12:26] <lifeless> I fall apart on 'which have the same id'
[12:26] <lifeless> because quite clearly; they dont
[12:26] <SteveA> the do, when you run them locally
[12:26] <lifeless> so I was looking for some local variables in that code block that would have the same value
[12:26] <SteveA> they don't on pqm
[12:27] <stub> It would be interesting to see repr(emailaddress.person) and repr(context), and possibly id(emailaddress.person) and id(context) if they are not shown
[12:27] <lifeless> SteveA: they *sometimes do* locally.
[12:27] <lifeless> SteveA: bradb is suffering a failure on this right now
[12:30] <lifeless> ok, another test run with stubs questions
[12:30] <lifeless> bradb: can you try to capture the value when they are the same ?
[12:30] <lifeless> bradb: i.e. print out the values when the are the same, and run test -vv so we can figure out where that should be
[12:32] <lifeless> bradb: you have a --story supporting runner ?
[12:32] <lifeless> bradb: thats on my todo - gimme a patch I'll finish & test it & shove it in rf for everyone
[12:36] <bradb> lifeless: I'll give you the patch for --story support shortly. Just make sure to credit appropriately. ;)
[12:38] <lifeless> bradb: of course
[12:46] <bradb> lifeless: hm, slight problem. SteveA pointed out that I can make a change to upstream Zope 3's test.py for this, so I'll have to tweak the patch before it is something that can be checked in.
[12:46] <bradb> s/can make/can't make/
[12:47] <lifeless> bradb: did you change it, or you 'can change it' ?
[12:47] <lifeless> bradb: we're still blocked on a new z3.
[12:47] <lifeless> bradb: If this *can* be done without changing the z3 test runner that is much better - for now.
[12:48] <bradb> lifeless: IOW, the only patch I can give you for story support involves hacking Z3's test.py right now. Otherwise I have to tweak it (I don't yet know if that'll be trivial or painful) to patch our own test.py.
[12:48] <SteveA> i'm fine with the patch being a change to our zope3
[12:48] <SteveA> when up get the new zope3, we'll briefly lose --story
[12:48] <SteveA> and then we'll have to get it back again
[12:48] <lifeless> bradb: ok, throw the patch my way
[12:49] <bradb> lifeless: are you ok with my just sending you the two needed files?
[12:49] <lifeless> bradb: sure
[12:49] <lifeless> bradb: if I can't figure it out, I'll let you know
[12:52] <bradb> lifeless: sent.
[12:52] <lifeless> danke
[12:52] <bradb> np
[12:55] <bradb> lifeless: lib/zope/app/tests/test.py was the test.py that I patched. You run it like python test.py --story=dirname.
[01:02] <lifeless> stub: 
[01:02] <lifeless> AssertionError: differing ids in emailaddress.person.id(<Person at 0x2aaab69a1650>,46912696360912,59) == self.context.id(<Person at 0x2aaab6c6ded0>,46912696382736,52)
[01:02] <lifeless> thats repr, id, foo.id
[01:09] <stub> So what has to be happening is that 30-mergepeople.txt is succeeding, and merging the foo account into the no-priv account. But when 32-setpreferred-email.txt is run, it isn't seeing the recent database changes for some reason.
[01:09] <lifeless> ahh
[01:09] <lifeless> race conditions I expect
[01:10] <stub> Possibly 32-setpreferred-email.txt is using SQLObject cache from the previous request (merge person uses direct SQL, so we need to ensure SQLObject caches are removed)
[01:11] <stub> The race condition shouldn't be possible unless test 32 is running before test 30 has committed or rolled back. (but not possible doesn't mean it isn't happening)
[01:12] <lifeless> ok
[01:12] <lifeless> so the end of 30-
[01:13] <lifeless> checks that it gets a perint by name of 'myemail'
[01:13] <lifeless> thats sqlobject yah ?
[01:13] <daf> SteveA: perhaps you could weigh in on https://launchpad.net/products/launchpad/+bug/3057?
[01:13] <Ubugtu> Malone bug #3057: "You're already logged in" message should never occur In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/3057
[01:13] <stub> perint?
[01:14] <lifeless> afgeifq?
[01:14] <stub> (19:13:15) Robert Collins (lifeless): checks that it gets a perint by name of 'myemail'
[01:15] <lifeless> person set
[01:16] <stub> Yer - we can stick a test at the end of 30- to ensure that SQLObject land sees the correct person is linked to that email address.
[01:17] <stub> And maybe a 31- test that just does that (ensuring the sanity of the data as viewed through SQLObject)
[01:17] <lifeless> I'm waaay tired
[01:17] <lifeless> can you take over ?
[01:18] <stub> Sure. Is anything landing through PQM at the moment? Or is this only intermittent when run through PQM
[01:18] <lifeless> (by which I mean, get it so that ~/tests/launchpad passes make check_merge)
[01:18] <lifeless> pqm is disabled until this is resolved
[01:18] <lifeless> it happens on every make check_merge on balleny
[01:18] <lifeless> regular as clockwork
[01:19] <stub> ok. How do I reenable PQM if I nut it out? Just switch the cronjob back on?
[01:19] <lifeless> so this is essentially a manual commit to pqm: commit in that dir, and bzr push to ~/archives/rocketfuel/launchpad/devel
[01:20] <lifeless> then bzr push to sftp://chinstrap.ubuntu.com/home/warthogs/archives/rocketfuel/launchpad/devel
[01:20] <lifeless> then just switch the cronjob back on
[01:20] <stub> ok
[01:21] <lifeless> thanks
[01:21] <lifeless> I'll check in in the morning and see where its at
[01:21] <lifeless> if you get stuck outside the 'make check_merge pass' bit, just email me where you did get up to
[01:37] <daf> who owns Project?
[01:38] <carlos> daf, Which Project?
[01:38] <daf> the class Project
[01:38] <carlos> daf, anyone from the launchpad team
[01:39] <carlos> daf, the Register does not have an specific developer working on it
[01:39] <daf> ah
[01:48] <carlos> Is there a planned date to move Ubuntu's bugs from bugzilla to malone?
[01:51] <daf> is there an owner for the specs code?
[01:51] <carlos> daf, Mark wrote it, not sure if he maintains it too 
[01:53] <jordi> carlos: I thought that migration was ready to happen any time now. I wonder what's holding it
[02:05] <daf> stub, spiv: perhaps one of you could comment on bug 347 -- I suspect it may be out of date by now
[02:05] <Ubugtu> Error: I cannot access this bug
[02:05] <Nafallo> linuxdcpp is missing in malone. source is in dapper, so could someone add it? :-)
[02:05] <stub> carlos: Bugzilla migration is blocked on a production Gina run
[02:06] <carlos> stub, oh, ok
[02:11] <daf> SteveA: I think bug 676 could do with your input too
[02:11] <Ubugtu> Malone bug #676: need persistent logins In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/676
[02:21] <carlos> stub, Why the Warning and Error notifications have the same UI?
[02:21] <carlos> stub, shouldn't it be different?
[02:21] <carlos> at least I expect them to be different...
[02:23] <daf> hurrah for people putting OOPS numbers in bug reports
[02:24] <stub> carlos: If you mean the notifications stuff, I don't see why you would want too apis. All you are doing is creating a message to show the user - the levels just control rendering.
[02:25] <stub> carlos: What do you expect to work differently between a warning and an error?
[02:25] <carlos> stub, I'm using self.request.response.addWarningNotification and self.request.response.addErrorNotification
[02:25] <carlos> and both have the same UI when rendered
[02:26] <stub> Ahh... I see. Because mpt wanted it that way ;)
[02:26] <carlos> That makes me move the Warning to Info....
[02:26] <carlos> will talk with mpt when he's online...
[02:26] <stub> carlos: That is incorrect - don't break your code to work around what you perceive as a UI issue
[02:27] <stub> carlos: Set the correct level and if you think it renders badly, open a bug with a url or screenshot of your page demonstrating the issue.
[02:27] <carlos> ok
[02:58] <lucasvo> I get reports about bugs 2 times, is this normal? :D
[02:59] <lucasvo> I made a bugreport on launchpad, I added a comment, that I can reproduce it on several machines, it got assigned, and I got the mail 2 times...
[02:59] <daf> sounds odd to me
[03:00] <daf> when you get duplicates, are they identical?
[03:00] <daf> or just nearly the same?
[03:00] <lucasvo> daf: yes
[03:00] <lucasvo> daf: identical at the same time
[03:00] <daf> same Message ID?
[03:00] <daf> hmm
[03:00] <ddaa> sorry for the time off, I tried to find a xorg configuration that does not destroy the built-in panel when using an external display...
[03:01] <daf> lucasvo: could you file a bug?
[03:01] <daf> lucasvo: https://launchpad.net/products/malone/+filebug
[03:01] <ddaa> Incomplete success. It seems that turning on UseFBDev is nicer to the built-in panel, but I was not unable to find a way to turn it off and still using the external display
[03:01] <lucasvo> Message-Id: <20051216131736.24631.27612.launchpad@canonical@gangotri.ubuntu.com>
[03:02] <lucasvo> Message-Id: <20051216131736.24631.27612.launchpad@canonical@gangotri.ubuntu.com>
[03:02] <lucasvo> they are the same
[03:02] <daf> cool, you can put that in the bug report
[03:03] <daf> it might help us track it down
[03:03] <lucasvo> daf: ok
[03:03] <daf> thanks
[03:03] <lucasvo> silly, creating bugreport because bugreport I created is wrong:D
[03:04] <daf> well, Malone has lots of bugs :)
[03:08] <lucasvo> https://launchpad.net/products/evolution/+bug/5837?lpnotification=4T1Ufd1zXd1EBl0KJNI9sqRreuQ
[03:08] <Ubugtu> Malone bug #5837: double delivery of mails about statuschanges of bugs In: evolution (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/5837
[03:09] <lucasvo> ah ok
[03:12] <bradb> lucasvo: Would it be possible for you to attach each message to that bug report?
[03:12] <lucasvo> ah, I made it in evolution...
[03:12] <lucasvo> bradb: yes
[03:14] <lucasvo> done
[03:14] <bradb> lucasvo: Thanks. /me looks
[03:15] <lucasvo> bradb: thank you! I am not the one who fixes it ;D
[03:15] <bradb> lucasvo: I only see one message attached though.
[03:15] <bradb> er, wait
[03:16] <lucasvo> bradb: hm, isn't it a file with multiple messages?
[03:16] <bradb> Sorry, I was being blind.
[03:16] <lucasvo> np
[03:19] <daf> bradb: if I subscribe someone else to a bug, do they get a notification?
[03:20] <bradb> daf: Yes.
[03:20] <daf> groovy
[03:22] <bradb> daf: Did you get two copies of the message whose body looked like:
[03:22] <bradb> Public bug report changed:
[03:22] <bradb> https://launchpad.net/malone/bugs/5831
[03:22] <bradb> Changed in: Launchpad (upstream)
[03:22] <bradb>        Assignee: (unassigned) => Matthew Paul Thomas
[03:22] <bradb>          Status: New => Accepted
[03:22] <Ubugtu> Malone bug #5831: New CSS design places long lines in the content over the right column In: launchpad (upstream), Severity: Normal, Assigned to: Matthew Paul Thomas, Status: Accepted https://launchpad.net/bugs/5831
[03:23] <daf> bradb: I can't see mails for bug 5831
[03:23] <Ubugtu> Malone bug #5831: New CSS design places long lines in the content over the right column In: launchpad (upstream), Severity: Normal, Assigned to: Matthew Paul Thomas, Status: Accepted https://launchpad.net/bugs/5831
[03:26] <Gijon> hello everybody
[03:26] <Gijon> can someone help me throuh the launchpad website
[03:26] <daf> sure
[03:26] <Gijon> i'v just registed
[03:26] <daf> that's a good start :)
[03:27] <Gijon> and i want to know how i can translate the gnomaber
[03:27] <Gijon> gnomebaker
[03:27] <daf> if you go to https://launchpad.net/products/gnomebaker
[03:27] <daf> you should see a "Translations" link on the right hand side
[03:27] <Gijon> got it
[03:28] <bradb> lucasvo: So are basically all bugs you're involved in causing double notifications to be sent on status changes?
[03:28] <daf> Gijon: then it's just a matter of choosing your language
[03:28] <Gijon> i've come to a page where i can see al languages
[03:28] <lucasvo> bradb: I didn't get this bugchanges twice
[03:28] <Gijon> and i see the name of the translaters
[03:29] <daf> Gijon: ok, wihch language are you interested in translating to?
[03:29] <Gijon> to portuguese 
[03:29] <Gijon> from portugal
[03:29] <Gijon> the real deal
[03:30] <lucasvo> bradb: but it looks like I really got it twice(not just only evolution) because on webmail it is the same
[03:30] <daf> Gijon: hey, careful, we have Brazilians in here :)
[03:30] <Gijon> lol
[03:30] <Gijon> daf, are u a brasilian?
[03:31] <daf> no, I'm not
[03:31] <daf> so, if you click on the "Portuguese" link, it'll take you to the translation page
[03:31] <bradb> stub: If I give you a message ID and a recipient, are you able to tell me how many times we sent that mail out from production?
[03:32] <Gijon> daf, yes now i see
[03:32] <bradb> lucasvo: Do you ever get seemingly duplicate email from any other sources?
[03:32] <daf> SteveA: what do you think about bug 3735?
[03:32] <Ubugtu> Malone bug #3735: Login page should always have the same URL In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/3735
[03:32] <stub> bradb: I don't think we have that info in any logs I have access to. You need elmo or Znarl to chase that.
[03:32] <Gijon> daf : but isn't there like a file .po that you can download and translate offline??
[03:32] <SteveA> bug 5839
[03:32] <Ubugtu> Malone bug #5839: in bug mail, the bug description should be in the body In: malone (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/5839
[03:32] <daf> Gijon: sure: click on Portuguese, then click the Download link
[03:33] <bradb> elmo: ping (question about one or more emails sent from production)
[03:33] <Gijon> daf, thanks
[03:34] <daf> Gijon: de nada
[03:34] <Gijon> daf, falas portugues?
[03:34] <daf> no :(
[03:34] <Gijon> daf, but you know how to say thanks..
[03:34] <daf> I can say "bom dia" and that's about it :)
[03:36] <Gijon> daf, that's not bad..
[03:36] <Gijon> daf, why did you learn how to speack a few words?
[03:36] <daf> well, I work with all these Brazilians...
[03:37] <SteveA> daf: on 3735, it is needinfo
[03:37] <SteveA> bug 3735
[03:37] <Ubugtu> Malone bug #3735: Login page should always have the same URL In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: NeedInfo https://launchpad.net/bugs/3735
[03:38] <daf> SteveA: you mean we want to know if it works with Amazon?
[03:39] <SteveA> yes
[03:40] <SteveA> this supports mpt's proposition that it is a browser bug
[03:40] <SteveA> even if we change login pages to not have +login on the end, they will still be in multiple locations
[03:43] <daf> hmm, where are the Brazilians?
[03:43] <daf> Internet trouble?
[03:43] <SteveA> public holiday?
[03:43] <SteveA> aparently not
[03:44] <SteveA> looks like telefonica has gone arseovertit
[03:44] <SteveA> Kinnison: phone call?
[03:45] <SteveA> or voip if pos
[03:45] <SteveA> phone then?
[03:45] <Kinnison> sure, *checks* noone is on the phone
[03:45] <Kinnison> landline please
[03:47] <ddaa> bradb: is that intentional that bugmail for the Launchpad Admins is now sent to the individual email addresses?
[03:48] <ddaa> Instead of the lanchpad mailing list
[03:48] <ddaa> that screws up my mail filtering, now I cannot tell Launchpad Admin bugspam from stuff I'm explicitely interested in :(
[03:48] <daf> who owns the support tracker?
[03:48] <bradb> ddaa: Email will get delivered individually to team members if there is no team email set.
[03:50] <ddaa> bradb: sure, I see admins has no contact address anymore
[03:50] <ddaa> SteveA: can I set the contact address of admins to launchpad@lists.canonical.com?
[03:55] <ddaa> Mh... nevermind that...
[03:56] <ddaa> bradb: is there no way to prevent us from receiving all the launchpad bugmail through launchpad-bugs AND direct email?
[03:56] <ddaa> short of setting the contact address for the the team to /dev/null...
[03:57] <bradb> ddaa: Several people have suggested having Malone be s-m-r-t enough to deliver only one bugmail in that case. I tend to maybe agree. IIRC, SteveA suggested that it's the user's problem.
[03:57] <bradb> Another possible alternative would be to include a footer which states precisely why you get each piece of mail.
[03:58] <bradb> "You are getting this email because you are a member of the A Team, and the A Team is subscribed to this bug"
[04:02] <bradb> ddaa: Maybe I should open a bug saying that it could be a preference?
[04:05] <daf> can somebody remind me where the account merge stuff is?
[04:06] <stub> daf: canonical.launchpad.database.person
[04:06] <daf> I mean, in the UI
[04:09] <ddaa> bradb: I'm okay with receiving email from launchpad-bugs and direct mail.
[04:09] <bradb> ok
[04:10] <ddaa> I'm complaining about mails committed recently, that I'm not subscribed to, and for which I received mails in my inbox
[04:10] <ddaa> Supposedly because I'm a launchpad admin...
[04:10] <daf> bradb: for bug triage, a "New bugs" link would be handy
[04:11] <ddaa> Well, I guess I would want an option for "do not send me bugmail for admins"
[04:11] <daf> Satus: New ==> not-triaged-yet
[04:11] <bradb> daf: You mean in addition to the "Untriaged Bugs" link, which already includes New bugs?
[04:11] <ddaa> But I still want mail for stuff that's assigned to me or that I'm subscribed to.
[04:12] <bradb> er, it actually reads just "Untriaged" but same difference
[04:13] <bradb> ddaa: Presumably the LP admins are subscribed to the bugs you're getting spammed on, right?
[04:14] <daf> bradb: doh!
[04:14] <daf> I didn't see that :)
[04:14] <daf> I think I was expecting it to be called "New bugs"
[04:15] <daf> er
[04:15] <bradb> daf: Interesting.
[04:16] <bradb> This reminds me of the time I showed kiko the "bugmail settings" screen for source packages and he said that I should have included a list of all the pkg bug contact on that page as well.
[04:17] <bradb> I had done that, but it was in one of the three portlets on the page.
[04:19] <ddaa> bradb: yes, like all Launchpad bugs...
[04:37] <Kinnison> ciau
[04:40] <kiko> hello there
[04:44] <kiko> daf, you bugspammer yo
[04:49] <SteveA> hi mpt 
[04:49] <kiko> daf, matsubara: I think bug 5815 is a dupe.
[04:49] <Ubugtu> Malone bug #5815: Timeout Error  In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: Accepted https://launchpad.net/bugs/5815
[04:49] <mpt> hi SteveA 
[04:49] <kiko> it's the same bug as +upstream
[04:49] <jbailey> SteveA: Glad to. =)  It's a reasonably slow day, so it's all good/
[04:49] <SteveA> mpt: jbailey and i are discussing bug 3578
[04:49] <Ubugtu> Malone bug #3578: System errors in launchpad should provide information In: launchpad (upstream), Severity: Normal, Assigned to: Matthew Paul Thomas, Status: NeedInfo https://launchpad.net/bugs/3578
[04:49] <carlos> mpt, hi
[04:50] <kiko> hey carlos 
[04:50] <carlos> kiko, hi
[04:50] <kiko> no PQM love eh?
[04:50] <SteveA> i am wondering if jbailey finds many of his points have been addressed with the new error pages
[04:50] <jbailey> SteveA: I think so far that's fine for a 404 page.  My guess (said without any statistical support) that most 404's are going to be bad links, or typos, so not worthy of a bug report.
[04:50] <SteveA> mpt: i think we need some more explicit text on the error pages saying how to report a bug
[04:50] <carlos> mpt, the problem with the translation form are not the suggestions but the first column of the table
[04:50] <SteveA> jbailey: the bad links are often problems with launchpad
[04:50] <carlos> mpt, takes too much space
[04:51] <SteveA> jbailey: such as the "specs" menu item taking you to a 404 page for an object
[04:51] <jbailey> SteveA: I find this the same problem was when people tell me to mail a self-addressed stamped envelope to them to get information.  I'm never sure if thats' really enough, or do I need to include a letter along with it saying what I want, or what..  It's just a feeling of being lost and unsure how to proceed.
[04:51] <jbailey> Ah, okay.
[04:52] <carlos> mpt, "I tried to fix that but I fucked the form so I will send you an email when I finish with the form so you can fix it, ok?
[04:52] <salgado> so, we agreed on "Non-method attributes and properties should use underscores". is that true for attributes that refer to database columns too? if so, should we name the database column with underscores?
[04:52] <mpt> carlos, ok
[04:52] <carlos> kiko, test block it....
[04:53] <carlos> mpt, also, I want to discuss with you the fact that the notification messages that we show in our forms render the Warning and the Errors with the same color and icon....
[04:53] <carlos> mpt, why do you think it should be done that way? (stuart told me that you decided to do it that way)
[04:53] <mpt> SteveA, I never put info in the 404 page about reporting bugs, because kiko's error reports are a better way of finding and prioritizing 404s
[04:53] <SteveA> salgado: please raise that on the launchpad mailing list, with some examples
[04:53] <SteveA> mpt: okay
[04:53] <mpt> carlos, no, I didn't.
[04:53] <carlos> oh
[04:53] <jbailey> SteveA: That's why I proposed the text saying "Hey, if you got this page and you think you're not supposed to, please send us an <a href="">email</a> saying that you got an error, reference number OOPS-B282.  We'll have all the rest of the information."
[04:54] <SteveA> mpt: do you know if there's a page in launchpad that is giving a real error?
[04:54] <carlos> mpt, so do we agree that the UI should be different?
[04:54] <mpt> carlos, it's just something I haven't gotten around to fixing
[04:54] <mpt> yes :-)
[04:54] <carlos> ok
[04:54] <carlos> mpt, thank you ;-)
[04:54] <jbailey> SteveA: So I think I'm looking for the assertion that this is actually enough information.  Or the notion that you'll let me know if I need to provide more, but generally that I won't need to, or something.
[04:54] <carlos> mpt, is there a bug report for it?
[04:54] <SteveA> jbailey: i'm going to find a real error, so we can see what it looks like now
[04:54] <mpt> carlos, I don't think so
[04:55] <carlos> mpt, I will file one as the translation form will start showing both kind of messages soon and it looks confusing
[04:55] <mpt> ok
[04:58] <mpt> There don't seem to be open bug reports on pages that cause errors directly, but bug 5757 is fairly harmless and easy to reproduce
[04:58] <Ubugtu> Malone bug #5757: Oops from making a second fix request for the same product In: malone (upstream), Severity: Normal, Assigned to: Diogo Matsubara, Status: Accepted https://launchpad.net/bugs/5757
[04:58] <mpt> SteveA, ^^
[04:59] <mpt> e.g. go to https://launchpad.net/products/malone/+bug/5757/+upstreamtask and enter "malone"
[04:59] <Ubugtu> Malone bug #5757: Oops from making a second fix request for the same product In: malone (upstream), Severity: Normal, Assigned to: Diogo Matsubara, Status: Accepted https://launchpad.net/bugs/5757
[05:00] <SteveA> https://launchpad.net/products/malone/+bugs-untriaged?batch_start=20&batch_end=40
[05:00] <mpt> "Please report this as a bug, after checking that it isnt reported already. In your bug report include the URL of this page, and what you were trying to do. (If this error is preventing you from reporting bugs, e-mail us instead, with Bug in the Subject line.) The reference for this error is OOPS-B285. Please include it in your bug report or email."
[05:00] <SteveA> the url https://launchpad.net/products/malone/+bugs-untriaged?batch_start=20&batch_end=40
[05:00] <SteveA> gives an oops
[05:01] <SteveA> salgado: there's a shipit oops too
[05:01] <salgado> SteveA, really? can you reproduce it?
[05:02] <SteveA> salgado: no, it depends on the user logged in, for myrequest
[05:02] <mpt> The only thing I can see that needs changing in that message, IMO, is merging the error code paragraph with the what-to-include-in-your-bug-report paragraph
[05:02] <SteveA> salgado: look at 2005-12-16/54528.A239
[05:02] <SteveA> mpt: we don't need the URL of this page.  we need only the oops
[05:02] <jbailey> mpt: I think there needs to be the message that reporting the reference number is enough information.
[05:02] <SteveA> mpt: and a description of what they're doing
[05:03] <mpt> ah, excellent
[05:03] <SteveA> mpt: in plain english
[05:03] <SteveA> mpt: the oops report such as the one i just pointed salgado at gives us a lot of information about what the user is trying to access
[05:04] <SteveA> i don't see why "from" is italicised
[05:05] <mpt> for the same reason it would be emphasized if spoken
[05:06] <mpt> it's talking about the catch-22 case
[05:07] <mpt> So, (1) fix the text of the Oops page
[05:08] <mpt> (2) remove the paragraph containing the error code from the 404 page
[05:08] <SteveA> i would emphasize "preventing" or "preventing you" and "reporting"
[05:08] <SteveA> i would not emphasize "from" in my spoken dialect
[05:08] <mpt> okay
[05:08] <SteveA> kiko-fud: i found this a useful grep:  grep -L "Exception-Type: \(NotFound\|RequestQueryTimedOut\|RequestExpired\)" *
[05:08] <SteveA> don't remove the oops code from the 404 page
[05:08] <SteveA> i think the 404 page is okay
[05:08] <daf> mpt: I've just drafted a spec, and there are a couple of points that could use your input
[05:09] <daf> mpt: would you mind taking a look?
[05:09] <mpt> daf, sure
[05:09] <daf> mpt: https://wiki.launchpad.canonical.com/DistroReleaseUINaming
[05:10] <mpt> SteveA, the only benefit of encouraging 404 victims to report bugs is that it makes it easy for them to be notified of our progress in fixing it
[05:10] <mpt> at least, the only benefit I can think of
[05:10] <daf> who's in charge of CoC stuff?
[05:11] <daf> kiko-fud: 5815 is a dupe of what?
[05:11] <mpt> SteveA, the disadvantages being that we have to deal with 404 bug reports that may not include referers, may be ranty, may be noisy disproportionately to their frequency, etc
[05:12] <SteveA> if we were being very smart, we'd show an OOPS code in a 404 only when we have a Referer from launchpad
[05:12] <SteveA> but, that's too tricky
[05:12] <SteveA> let's just leave them in, but not particularly encourage error reporting, as now
[05:12] <mpt> okay.
[05:13] <daf> SteveA: the referer check sounds like a good idea to me -- perhaps we should open a low-priority bug on that?
[05:13] <mpt> daf, DafyddHarries4? What happened to the other three?
[05:14] <SteveA> daf: no, i think no
[05:14] <daf> ok
[05:14] <SteveA> daf: it is bad news to put too much smarts in an error page
[05:14] <daf> mpt: I can't remember how to merge my account with the one that owns DafyddHarries
[05:14] <daf> SteveA: good point
[05:14] <SteveA> and we can find out about such pages with a referer from analyzing the oops logs
[05:16] <mpt> daf, is that a Launchpad bug that it's not obvious enough?
[05:16] <SteveA> so, seems that we're still getting an IntegrityError and an SQLObjectNotFoundError somewhere
[05:16] <mpt> daf, I like that spec
[05:17] <daf> mpt: great, thanks
[05:17] <daf> mpt: sorry, that what's not obvious enough?
[05:17] <mpt> how to merge your account isn't obvious enough
[05:18] <salgado> stub, around?
[05:18] <stub> salgado: barely
[05:18] <daf> oh, probably
[05:18] <stub> salgado: wassup?
[05:18] <salgado> stub, I'm gonna be quick. privmsging you
[05:18] <daf> salgado: do you know who owns the CoC code?
[05:19] <bradb> I have been unable to reconnect to irc.freenode.net since my laptop spontaneously combusted.
[05:20] <mpt> bradb, that's Freenode's fault
[05:20] <mpt> unless your laptop combusted days ago
[05:20] <mpt> http://freenode.net/news.shtml
[05:21] <salgado> daf, it was written by cprov, but I don't think he's maintaining it
[05:21] <SteveA> bradb: the url https://launchpad.net/products/malone/+bugs-untriaged?batch_start=20&batch_end=40
[05:21] <bradb> SteveA: thanks
[05:22] <daf> salgado: ah, thanks
[05:22] <salgado> daf, I'm going to take responsibility for it, but I've been having some higher priority tasks that's preventing me from doing so
[05:22] <daf> sure, I just want to know who to assign bugs to :)
[05:22] <bradb> mpt: Weird. I had similar problems the last time my laptop died (that time having run out of battery.)
[05:23] <SteveA> daf: OOPS-B245 is a soyuz UI one
[05:23] <daf> SteveA: does it have a bug?
[05:24] <Nafallo> oh, could someone assign malone 1801 to carlos? :-)
[05:24] <Ubugtu> Malone bug #1801: Show Suggestions In: rosetta (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/1801
[05:24] <carlos> Nafallo, I'm changing the translation form atm...
[05:25] <Nafallo> so that means it's fixed soonisch? :-)
[05:25] <daf> carlos: is the bug valid?
[05:25] <SteveA> daf: i do not know
[05:26] <carlos> daf, yes, it is. Is a feature request more than a bug ;-)
[05:26] <daf> SteveA: is this one you encountered yourself?
[05:26] <carlos> Nafallo, I think it's doable, so yes, I will try to add it now.
[05:27] <Nafallo> carlos: nice. good thing I did the reminder then :-)
[05:27] <daf> who is doing ShipIt work these days?
[05:27] <carlos> Nafallo, well, I was planning to go over all bug reports to fix/implement the easy ones
[05:27] <Nafallo> :-)
[05:28] <carlos> Nafallo, but it's a good thing that you pointed me to it anyway ;-)
[05:28] <Nafallo> the only reason I remembered it was a comment today, so it landed on top in my inbox :-)
[05:28] <daf> Znarl: we have an open bug saying that launchpad@ubuntu.com doesn't work
[05:29] <daf> Znarl: apparently we use this address in some of the mail we send out
[05:31] <SteveA> daf: no, i was grepping through today's oops reports
[05:31] <SteveA> looking for obvious programming errors
[05:31] <daf> ah, ok
[05:31] <SteveA> rather than the more subtle timeouts and not founds
[05:33] <SteveA> daf: you should file an RT report
[05:33] <SteveA> daf: rather than getting admins on irc
[05:33] <daf> how do I do that?
[05:34] <SteveA> i'll tell you in a private message
[05:34] <jbailey> SteveA: I wonder how many of us did. =)
[05:35] <daf> SteveA: do you know who maintains ShipIt?
[05:35] <SteveA> daf: yes
[05:35] <SteveA> it is salgado
[05:35] <daf> thanks
[05:38] <daf> mpt: what did you think about the UI Changes section?
[05:45] <mpt> daf, I liked it all
[05:45] <mpt> iirc someone else (ddaa, perhaps) suggested exactly the same presentation you did
[05:46] <daf> I mean the /!\ bit
[05:46] <mpt> e.g. Ubuntu 5.10 (Breezy)
[05:46] <Ubugtu> Error: Could not parse XML returned by Ubuntu Bugzilla bugzilla: not well-formed (invalid token): line 88, column 76
[05:46] <daf> ok, in that case I'll just remove the note and send it for review
[05:47] <daf> mpt: you don't happen to know where the person merge page is, do you?
[05:47] <mpt> no, sorry :-)
[05:47] <mpt> salgado should know
[05:47] <salgado> daf, launchpad.net/people
[05:47] <salgado> daf, there's a link there
[05:48] <daf> ah, so there is
[05:48] <daf> thanks
[05:48] <daf> "Constraint not satisfied"
[05:48] <daf> maybe it's because name82 doesn't have an email address
[05:49] <salgado> it could be
[05:50] <salgado> the "Constraint not satisfied" is the standard error message when something is not in a vocabulary. I filed a bug asking for some way to be able to specify a custom error message  on the field that uses the vocabulary
[05:52] <daf> how do accounts without email addresses get merged?
[05:52] <salgado> they don't
[05:52] <daf> indeed, if you have no email address, it's pretty hard to log in, yes?
[05:53] <salgado> it's impossible. you can only login if you have a validated email address
[05:53] <daf> and you can't create a validated email address if you're not logged in? :)
[05:55] <salgado> exactly. what we need is something that allow admins to merge accounts without going through the normal workflow
[05:55] <salgado> there's a bug open for that too
[05:55] <daf> aha, I can use my admin powers to change name82's wikiname, and then I can reclaim DafyddHarries for myself
[05:55] <daf> hurrah
[05:59] <daf> SteveA: https://launchpad.net/products/launchpad/+spec/distroreleasenaming
[06:02] <salgado> SteveA, I don't seem to be able to parse the traceback that was pasted on bug 5808. can you give me some help on it?
[06:02] <Ubugtu> Malone bug #5808: Shipit NotFoundError In: shipit (upstream), Severity: Normal, Assigned to: Guilherme Salgado, Status: New https://launchpad.net/bugs/5808
[06:05] <SteveA> salgado: it is saying that there is an expression in a page template like context/quantityx86
[06:06] <SteveA> and that the ShippingRequest object that is 'context' doesn't have a quantityx86 attribute or __getitem__ key
[06:06] <SteveA> the name in the page template expression might be not 'context' but something else
[06:06] <SteveA> so, find the likely template, adn look for quantityx86 in it
[06:06] <salgado> SteveA, right, but why doesn't it tell me where is the whatever/quantityx86, like I've seen in many tracebacks?
[06:07] <SteveA> because the traceback doesn't say so
[06:07] <SteveA> there is special code in the warnings logging code to do this
[06:07] <SteveA> but that code is not hooked up to the tracebacks you see there
[06:12] <SteveA> daf: https://launchpad.net/products/launchpad/+spec/distroreleasenaming looks good to me
[06:12] <daf> SteveA: great, what happens next?
[06:13] <SteveA> i ask certain other people to take a look at it
[06:13] <kiko-fud> daf, bug 5815?
[06:13] <Ubugtu> Malone bug #5815: Timeout Error  In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: Accepted https://launchpad.net/bugs/5815
[06:13] <kiko-fud> daf, if you look at my error report summary from last week i listed the bug #
[06:13] <daf> bradb: I've found myself wanting to annotate a bug with something that says "this is easy to fix"
[06:13] <bradb> daf: Like a keyword or something?
[06:14] <daf> yeah
[06:14] <daf> like...
[06:14] <kiko-fud> I use [trivial]  in the summary :)
[06:14] <daf> a TAG!
[06:14] <bradb> daf: I brain dumped keywords with SteveA this week in .lt.
[06:14] <daf>  <kiko> daf, matsubara: I think bug 5815 is a dupe.
[06:14] <Ubugtu> Malone bug #5815: Timeout Error  In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: Accepted https://launchpad.net/bugs/5815
[06:14] <bradb> We have a simple solution, coming to a theatre near you.
[06:14] <daf> oops
[06:14] <daf> bradb: great
[06:14] <kiko-fud> me?
[06:15] <bradb> salgado: Has that optimization of the people-related vocabs landed yet?
[06:16] <salgado> bradb, two weeks ago, I think. why?
[06:16] <bradb> salgado: Er, by "landed" I meant "been rolled out".
[06:16] <bradb> I guess it's been rolled out then?
[06:17] <bradb> I'm trying to figure out why bug 5847 is happening to daf
[06:17] <Ubugtu> Malone bug #5847: editstatus page is too long In: malone (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/5847
[06:17] <salgado> yes, it was rolled out more than a week ago
[06:17] <bradb> hm
[06:18] <daf> bug 4031 seems especially prone to timing out
[06:18] <Ubugtu> Malone bug #4031: GTK+ french translation request takes too long error In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/4031
[06:18] <daf> ironic, really
[06:19] <daf> yeah, there it goes again (OOPS-A290)
[06:19] <daf> I'm trying to reassign it to rosetta and assign it to carlos at the same time
[06:19] <daf> perhaps it's the product lookup being slow
[06:20] <daf> hmm, just changing the product works
[06:20] <bradb> daf: Do you get the timeout on that page from doing something other than changing the assignee + product?
[06:20] <daf> aha, it still times out if I just assign it
[06:20] <daf> (to carlos)
[06:20] <SteveA> daf: thanks for writing the release UI spec.  i approve it.
[06:21] <daf> thanks!
[06:21] <carlos> daf, The person lookup takes a bit...
[06:21] <salgado> daf, what text do you enter to assign it to carlos?
[06:21] <carlos> daf, what are you using as my account?
[06:21] <carlos> :-P
[06:22] <daf> "carlos"
[06:22] <daf> it's worked before
[06:22] <carlos> daf, use my email
[06:22] <carlos> it's faster
[06:22] <carlos> it works but if the server is busy it timeouts...
[06:22] <daf> hmm
[06:22] <salgado> carlos, actually, the lookup is almost instantaneous. what takes time is to retrieve more than 10000 people whose name match 'carlos'. :)
[06:22] <daf> heh :)
[06:23] <carlos> salgado, oh
[06:23] <daf> perhaps a LIMIT is in order?
[06:23] <daf> or trying to look up Person.name first?
[06:24] <salgado> that's a problem with the vocabs, it's nothing specific to the people vocab
[06:24] <daf> ah
[06:24] <daf> tricky
[06:25] <bradb> Yeah, it's looking like every timeout is happening in a people-related query
[06:25] <bradb> e.g.
[06:25] <bradb> Exception-Value: (('\n        SELECT DISTINCT Person.id, Person.displayname FROM Person\n            LEFT OUTER JOIN EmailAddress ON Person.id = EmailAddress.person\n         WHERE \n        (teamowner IS NOT NULL OR (\n            teamowner
[06:25] <bradb> IS NULL AND password IS NOT NULL AND\n            merged IS NULL AND EmailAddress.status = 4\n            ))\n         AND Person.id = 98 ORDER BY displayname, id',), {})
[06:26] <daf> perhaps the bug I filed should be retitled "person vocabs can be too slow"?
[06:27] <bradb> daf: That might be it. I'll add some information to that bug report.
[06:28] <bradb> As a starting point for possible optimization(s)
[06:28] <salgado> anybody has read access to the production or staging db?
[06:28] <daf> bradb: great
[06:29] <SteveA> salgado: i may be able to help
[06:30] <salgado> SteveA, can you run https://chinstrap.ubuntu.com/~dsilvers/paste/fileBVYP0C.html and tell me how long it takes?
[06:31] <salgado> bradb, daf, these problems are usually tricky, because there might be a query that takes 7 seconds but is always the first one to be executted, and then the other ones have only one second to run and the traceback will never give you the one that's actually eating up all the time
[06:31] <daf> bradb: looks like it's a dup, then: https://launchpad.net/products/launchpad/+bug/5415
[06:31] <Ubugtu> Malone bug #5415: Launchpad times out while searching for a person In: launchpad (upstream), Severity: Normal, Assigned to: Guilherme Salgado, Status: New https://launchpad.net/bugs/5415
[06:32] <daf> salgado: mmm, good point
[06:32] <salgado> we need to store all queries issued during a transaction, and show them when that transaction times out
[06:32] <daf> that would be good
[06:32] <SteveA> salgado: https://chinstrap.ubuntu.com/~dsilvers/paste/filekbzM2V.html
[06:33] <salgado> having only the last one is useful, but in some cases we need all of them
[06:33] <daf> if there was timiing information for each query, even better!
[06:33] <salgado> SteveA, thanks!
[06:33] <SteveA> salgado: please file a bug with that "store all queries issued" idea, and assign to jamesh
[06:34] <salgado> bradb, that query you got in the traceback took 75.571 ms on production.
[06:34] <daf> SteveA: bug 5758 looks like a SteveA sort of bug
[06:34] <Ubugtu> Malone bug #5758: vserver (Ubuntu) - util-vserver: merge new debian version In: util-vserver (Ubuntu), Severity: Normal, Assigned to: MOTU Merge Team, Status: Fixed https://launchpad.net/bugs/5758
[06:34] <salgado> SteveA, sure
[06:34] <SteveA> on staging
[06:35] <SteveA> salgado: i don't know if it is so significant though
[06:35] <SteveA> salgado: we need to consider the whole request anyway
[06:35] <SteveA> so, also subscribe stub to the bug report
[06:35] <SteveA> and we'll get some discussion going
[06:35] <salgado> SteveA, sorry, I didn't get what could not be so significant
[06:36] <salgado> could be not so significant, even
[06:36] <daf> SteveA: maybe bug 5765 too
[06:36] <Ubugtu> Malone bug #5765: Absence of fmt:datetime for code generated datetime.datetime instances In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/5765
[06:37] <SteveA> often the solution to having too many queries in the request is to refactor
[06:37] <SteveA> and so recording that until the timeout may or may not help
[06:37] <SteveA> so, it needs some discussion to see if it will be a good way to proceed
[06:37] <SteveA> daf: why bug 5758?
[06:37] <Ubugtu> Malone bug #5758: vserver (Ubuntu) - util-vserver: merge new debian version In: util-vserver (Ubuntu), Severity: Normal, Assigned to: MOTU Merge Team, Status: Fixed https://launchpad.net/bugs/5758
[06:39] <daf> SteveA: sorry, bug 5768
[06:39] <Ubugtu> Malone bug #5768: enumvalue doesn't work with security proxied objects In: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New https://launchpad.net/bugs/5768
[06:44] <daf> wow, a bug from an OS/2 user using Launchpad
[06:45] <SteveA> bug 5768
[06:45] <Ubugtu> Malone bug #5768: enumvalue doesn't work with security proxied objects In: launchpad (upstream), Severity: Normal, Assigned to: Steve Alexander, Status: New https://launchpad.net/bugs/5768
[06:46] <daf> oh, Google have released the code behind Google Talk
[06:47] <mpt> very helpful he was too, daf
[06:47] <mpt> mailed me lots of screenshots :-)
[06:47] <daf> looks like the bug is waiting on you, though :)
[06:47] <daf> should it be Accepted, at least?
[06:47] <mpt> When I triaged it, Accepted didn't mean what it means now
[06:48] <daf> what does it mean now?
[06:48] <daf> I use it to mean "somebody has looked at it and thinks there's something that needs fixing"
[06:48] <bradb> It means "Confirmed"
[06:48] <daf> what has been confirmed?
[06:48] <mpt> daf, before I was using it to mean "I'm going to be fixing this bug"
[06:49] <mpt> now it's about to be renamed "Confirmed"
[06:49] <mpt> the existence of the bug
[06:49] <daf> ok
[06:50] <bradb> lifeless: around?
[06:52] <bradb> bradb@oxygen:~/canonical/malone-smallfixes $ cat .bzr/x-push-data
[06:52] <bradb> chinstrap:/home/warthogs/archives/bradb/launchpad/malone-smallfixes/
[06:52] <bradb> bradb@oxygen:~/canonical/malone-smallfixes $ bzr push
[06:52] <bradb> bzr: ERROR: No push location known or specified.
[06:53] <daf> right, I'm all triaged out
[07:01] <kiko> Kinnison, ping?
[07:02] <kiko> daf, the voip part of it?
[07:04] <Kinnison> kiko: yo dude, you just caught me
[07:05] <kiko> Kinnison, that's because I'm lucky
[07:05] <Kinnison> kiko: lucky 8
[07:05] <kiko> Kinnison, is there a reason we're not generating Contents files via apt-ftparchive?
[07:05] <Kinnison> kiko: yes, it takes fucking ages
[07:05] <kiko> Kinnison, and is there a plan for providing the daily_singing_key.asc?
[07:05] <Kinnison> kiko: Even james only creates them periodically
[07:06] <Kinnison> the signing key will be provided when we switch
[07:06] <kiko> Kinnison, 'provided'?
[07:07] <kiko> and Release.gpg?
[07:07] <sivang> daf: OS/2 ? is anybody still using it? :)
[07:07] <Kinnison> kiko: When we have the signing keys from the old archive
[07:07] <Kinnison> kiko: Release.gpg is made by signing the Release files using the archive key
[07:08] <carlos> sivang, many banks
[07:09] <carlos> sivang, at least in Spain
[07:09] <kiko> Kinnison, is there code that will sign those files now? I don't care about them being the same -- I care about them existing, right now.
[07:09] <kiko> and I want to make sure we know what needs to be manually tweaked when rolling this baby out
[07:09] <SteveA> mpt: hello
[07:09] <Kinnison> kiko: I have a shell lump which does it, I ought to put it as a stage at the end of the publisher
[07:10] <SteveA> mpt: so, i think we have three types of "portlet" box
[07:10] <mpt> hi SteveA 
[07:10] <SteveA> we have one kind that is for menus and sitemaps
[07:10] <SteveA> we have one kind that shows carefully-formatted data
[07:10] <SteveA> we have one kind that gives narrative text
[07:10] <kiko> Kinnison, "ought" to? 
[07:10] <SteveA> can you make each kind have a different css class?
[07:10] <Kinnison> kiko: Well, in the katie world, it's done at the end of cron.daily
[07:10] <SteveA> and then make the one with narrative text have underlined links?
[07:10] <Kinnison> kiko: which is shell
[07:11] <Kinnison> kiko: but I think it ought to be in the python
[07:11] <mpt> SteveA, ok
[07:11] <SteveA> althoiugh...
[07:11] <SteveA> it isn't that simple
[07:11] <kiko> Kinnison, I see. are there bugs filed on this?
[07:11] <SteveA> consider the "open source people" box on the front page
[07:11] <SteveA> the top paragraph there is narrative
[07:11] <SteveA> and under that is a table
[07:11] <Kinnison> kiko: I'd tell you if I could navigate the new launchpad layout
[07:12] <daf> kiko: yes, libjingle
[07:12] <SteveA> as inconsistent as it seems, i'd expect to see the link within the text as underlined
[07:12] <kiko> Kinnison, or an item on the soyuz deployment wikipage?
[07:12] <daf> http://code.google.com/apis/talk/index.html
[07:12] <SteveA> and the links in the table of people not so
[07:12] <Kinnison> kiko: bug 3540 is the release.gpg placeholder bug
[07:12] <Ubugtu> Malone bug #3540: publisher (upstream) - Missing Release.gpg In: launchpad-publisher (upstream), Severity: Normal, Assigned to: Daniel Silverstone, Status: New https://launchpad.net/bugs/3540
[07:12] <kiko> Kinnison, okay.
[07:12] <Kinnison> and it should be on the wiki page
[07:12] <kiko> jordi, carlos, I need some time with you today to finish you performance review
[07:12] <kiko> ok.
[07:13] <Kinnison> yep 'tis on the wiki page
[07:13] <kiko> I'll add notes on the bits we are missing in the archive with bug numbers.
[07:13] <sivang> carlos: ah I see, interesting :)
[07:13] <carlos> kiko, I will be busy in 15 minutes or so....
[07:13] <carlos> kiko, do you want to talk now or when I'm back?
[07:13] <kiko> carlos, and afterwards?
[07:13] <kiko> when will you be back?
[07:14] <carlos> well, I'm not leaving my home, but I will have a visit. I suppose that it should not take more than one hour
[07:14] <kiko> Kinnison, let me ask you one more thing
[07:14] <kiko> carlos, okay, ping me when you are back.
[07:14] <carlos> kiko, ok
[07:14] <SteveA> mpt: can you sort out something that makes sense with styles or considering what elements text is inside?
[07:14] <kiko> Kinnison, one last item
[07:15] <mpt> SteveA, yes, I can special-case <p>
[07:15] <kiko> Kinnison, can you give me an idea of what the situation are with the commandline tools that need to be built for deployment?
[07:15] <kiko> Kinnison, I know that queue is in cprov's hands so that's fine
[07:15] <Kinnison> so, cprov has queue
[07:15] <kiko> right
[07:15] <Kinnison> I have an override changer which I need to push, then I'll mail cprov so he can look at it
[07:16] <Kinnison> we talked about sync porting
[07:16] <kiko> this is for post-publishing changes right?
[07:16] <Kinnison> and various other low-priority tools
[07:16] <Kinnison> yeah, this is post-publishing
[07:16] <Kinnison> I've got nowhere near a working version of any of the low-priority tools
[07:16] <cprov> Kinnison: good, It won't be a big issue
[07:17] <kiko> Kinnison, and the buildd master tools are actually web-based?
[07:17] <SteveA> mpt: thanks
[07:18] <Kinnison> kiko: yep as far as I can tell
[07:18] <kiko> so there's no commandline buildd stuff
[07:19] <Kinnison> cprov has a twisted shell thing
[07:19] <Kinnison> but other than that, it should all be web-ui
[07:20] <kiko> Kinnison, why does our queue builder need to support PackagesArchSpecific?
[07:20] <cprov> kiko: something are already done in buildd Web UI, like reset build, other require some redesign of builddmaster, like those that requires XMLRPC access to the slaves
[07:21] <cprov> s\are\is
[07:21] <kiko> can't it just be done with some SQL-Fu and an early-return?
[07:21] <Kinnison> kiko: PaS is used because we can't always trust the Architecture line
[07:21] <kiko> sure
[07:21] <Kinnison> kiko: elmo or Kamion will be able to explain the finer points of it
[07:21] <kiko> but we know that from the database, right?
[07:21] <cprov> Kinnison: buildd-monitor (twisted shell) is kind of dead atm
[07:22] <Kinnison> the db trusts the packages
[07:22] <kiko> I see
[07:23] <kiko> ok
[07:23] <kiko> I'll spend some time updating the SoyuzRunsUbuntuTaskList page today
[07:23] <kiko> and I'm officially nodding okay to a gina run
[07:23] <Kinnison> cool
[07:23] <kiko> 1.3 years after writing its initial version
[07:24] <Kinnison> You've done so much work on her dude, you deserve a big prize
[07:24] <kiko> so jamesh, you'll have packages soon
[07:24] <kiko> hopefully I can collect that prize in sexual favors from her
[07:24] <kiko> but she's such an uptight babe
[07:24] <kiko> Kinnison, ah.
[07:24] <Kinnison> kiko: given how many men have been at her, do you want to risk it?
[07:25] <kiko> that's half the fun
[07:25] <kiko> Kinnison, any issue with us running Gina over everything + dapper?
[07:25] <kiko> Kinnison, also, note that we are publishing arm for breezy, which is kinda crack
[07:25] <kiko> did your database patch include a DAR for breezy-arm?
[07:25] <Kinnison> mine didn't, I'm fairly sure
[07:26] <kiko> so wtf
[07:26] <kiko> is your db patch in rf?
[07:27] <Kinnison> I mailed it to stub because he wanted to run it on production etc
[07:27] <kiko> it should be in RF always
[07:27] <Kinnison> it'd be far too confusing for people
[07:27] <kiko> doesn't need to be sampledata
[07:28] <Kinnison> given stub modified them slightly, I'd ask him to add them to RF if I were you
[07:28] <kiko> yeah.
[07:30] <Kinnison> hmm, 18:30
[07:30] <Kinnison> time for me to go wash my hair and get stuff in the car
[07:30] <Kinnison> kiko: you have the mail from me to warthogs about how to reach me
[07:30] <Kinnison> kiko: yes?
[07:30] <kiko> I should have yes
[07:30] <kiko> thanks.
[07:30] <kiko> see you.. what day are you back?
[07:32] <Kinnison> umm 9th
[07:32] <Kinnison> but I imagine I'll be suffering withdrawl before then and I'll wave around on here a bit
[07:32] <kiko> your help in january will be vital
[07:32] <kiko> so thanks
[07:32] <kiko> at any rate, have a good time. thanks for the answers
[07:32] <cprov> Kinnison: hope we can reach you by phone ;) see you and have a nice time
[07:33] <Kinnison> cprov: yep, have a good time yourself, and give janice a big hug and a kiss from me and rob
[07:33] <cprov> ehe
[07:33] <kiko> my dogs are fighting
[07:33] <Kinnison> ciau all
[07:33] <cprov> Kinnison: will do, we may call you sometime to wish happy new year ;)
[07:34] <Kinnison> cprov: mail me with the pt_BR for "Merry Christmas and a happy new year" if you would?
[07:35] <cprov> Kinnison: sure, bye
[07:56] <mpt> Kinnison, Feliz Natal, e ... something ... ano qui a vem
[07:58] <mpt> Kinnison, Feliz Natal, e Feliz Ano Novo
[07:58] <mpt> (though that's "Happy Christmas", obviously -- the translation for "Merry" is something complicated about being full of lights)
[08:00] <kiko> SteveA, will you deliberate on salgado's question?
[08:18] <mpt> whee, only 13 pagetest errors left
[08:18] <mpt> time to cause some more
[08:27] <LarstiQ> heh
[08:36] <carlos> kiko, I'm back
[08:37] <carlos> mpt, I can give you my pagetest errors if you are bored....
[08:37] <carlos> ;-)
[08:37] <kiko> carlos, that's nifty. Let me email you and I'll call you a few minutes afterwards. 
[08:37] <carlos> ok
[09:12] <kiko> carlos, I'm adding some notes on the soyuz deployment and then I'll ring. are you okay with taking a call?
[09:12] <carlos> yes
[09:13] <carlos> but I didn't get your email...
[09:13] <kiko> really?
[09:13] <carlos> yes
[09:14] <kiko> damn.
[09:14] <kiko> that throws a spanner in the works
[09:22] <jordi> kiko: I have 10 minutes now, if not it probably needs to wait until monday or sunday night
[09:23] <kiko> jordi, no chance for tomorrow morning?
[09:23] <jordi> nope, I'll be... in the olive tree fields.
[09:24] <jordi> I need to go pick olives. You know, the makers of the olive gold sometimes need to do the shitty job.
[09:28] <jordi> kiko: I need to leave
[09:28] <jordi> tty on Sunday=
[09:29] <kiko> hmmm.
[09:29] <kiko> ok.
[10:25] <mpt> """it must be a valid inside Launchpad context"""
[10:25] <mpt> ohhhhhh really
[10:41] <kiko> not really
[11:51] <mpt> wow, this is close
[11:55] <mpt> 35 projects
[11:55] <mpt> 39 products pretending to be projects
[11:55] <mpt> 3 translation teams pretending to be projects
[11:56] <mpt> 1 distro pretending to be a project
[11:56] <mpt> 1 commercial Web site pretending to be a project