[12:25] <seb128> how comes than some people still comment on bugzilla?
[12:27] <dholbach> seb128: are the comments on bugzilla and launchpad in sync?
[12:27] <seb128> dholbach: ??
[12:28] <dholbach> which bug is it?
[12:28] <seb128> dholbach: bugzilla should be read-only, I guess that some people still have the login activated from before the migration or something
[12:28] <seb128> dholbach: not going to make a difference, you are not a bugzilla admin :)
[12:28] <dholbach> ...
[12:28] <seb128> http://bugzilla.ubuntu.com/show_bug.cgi?id=19777 got one
[12:28] <Ubugtu> Ubuntu bug 19777: "Program freeze on selecting directory using gnome file chooser" Product: Ubuntu, Component: gtk+2.0, Severity: normal, Assigned to: seb128@ubuntu.com, Status: NEEDINFO
[12:29] <dholbach> I just wanted to know if launchpad and bugzilla have the same comments in those cases.
[12:29] <seb128> http://bugzilla.ubuntu.com/show_bug.cgi?id=15372 got one saturday
[12:29] <Ubugtu> Ubuntu bug 15372: "alts_toggle breaks instead of overriding ralt -> l3" Product: Ubuntu, Component: xkeyboard-config, Severity: major, Assigned to: daniel.stone@ubuntu.com, Status: REOPENED
[12:29] <seb128> dholbach: there is no gateway between both
[12:29] <seb128> dholbach: the import was a one run stuff and bugzilla is supposed to be locked (if I got it right)
[12:29] <dholbach> I see. :/
[12:32] <mpt> oh dear
[12:32] <mpt> so Bugzilla login was disabled, but that didn't stop people who had checked the "Remember me" checkbox or whatever it was
[12:34] <mpt> Fortunately that particular comment was mostly noise ;-)
[12:35] <seb128> yeah, but I've got 3 on saturday too
[12:35] <seb128> and I'm not subscribed to every bugs on it
[12:36] <mpt> jamesh?
[12:36] <jamesh> mpt: yeah?
[12:37] <mpt> can you turn off bug editing for real, instead of just turning off logging in? :-)
[12:37] <mpt> (on bugzilla)
[12:38] <jamesh> mpt: I don't know
[12:39] <jamesh> I used the method to disable editing that kiko suggested
[12:39] <mpt> well it seems not to be working
[12:39] <jamesh> it'd probably be worth changing the db permissions at this point, actually
[12:40] <mpt> jamesh, a cheap and nasty way to fix the problem right now would be to delete process_bug.cgi and enter_bug.cgi
[12:42] <jamesh> mpt: I don't have the ability to do any of these things on my own though
[12:42] <mpt> who does?
[12:42] <jamesh> we'd need elmo or Znarl
[12:42] <mpt> ... followed up later by hacking show_bug.cgi to display the data as text rather than form fields
[12:42] <mpt> ok, so it's an rt thang
[12:43] <mpt> what's the rt address?
[12:43] <jamesh> I just use the web interface
[12:43] <jamesh> https://rt.admin.canonical.com/
[12:44] <mpt> ok, I'll request that now
[12:44] <dholbach> good night everybody.
[12:44] <mpt> except it wants a "username"
[12:44] <jamesh> the login is the same as the one we use for https://chinstrap.ubuntu.com
[12:44] <jamesh> iirc
[12:44] <jamesh> and the canonical wiki
[12:44] <mpt> ok
[12:45] <elmo> eww eww
[12:45] <elmo> please use mail
[12:45] <elmo> to create tickets
[12:45] <elmo> otherwise they come from a generic user and replies don't go to anyone
[12:45] <elmo> just mail rt@admin.canonical.com
[12:45] <mpt> ok
[12:45] <elmo> jamesh: did you see my question earlier about the activity log in malone?
[12:46] <jamesh> elmo: no
[12:46] <de_wizze> do we raise web design questions here ?
[12:46] <elmo> 18:48 < elmo> jamesh: do you think it'd be possible to add something to the activity log for all the bugs imported from bugzilla?
[12:47] <mpt> de_wizze, what kind of Web design questions?
[12:47] <de_wizze> how about default setting concerns
[12:47] <mpt> about the design of Launchpad? yes
[12:47] <jamesh> elmo: I suppose so.  What do you think would be appropriate?
[12:47] <mpt> about Web design in general? no :-)
[12:47] <lifeless> allYouBugsBelongToUs
[12:47] <elmo> jamesh: just any note that amounts to "imported from bugzilla on dd-mm-yyyy" or whatever?
[12:47] <de_wizze> Web Design - the launchpad page has horizontal scroolbars in IE
[12:48] <elmo> [orthogonally-ish, in general I'd love to see activity log more widely used, last I checked, it doesn't record outgoing emails] 
[12:49] <mpt> de_wizze, every page?
[12:49] <mpt> or just the front page?
[12:49] <mpt> or just some other page?
[12:49] <de_wizze> every page that I have been to so far
[12:50] <lifeless> if we are whinging about malone, where is the 'malone spams me when I change something' bug on the fixlist ?
[12:50] <elmo> oh the non-subscriber stuff?
[12:51] <elmo> that might be me :/
[12:51] <lifeless> ?
[12:51] <de_wizze> I think is has to do with the fact that a width of 100% is hard coded into the table class named "header"
[12:51] <elmo> lifeless: oh, nm, you're talking about something else, I guess.  do you mean the fact you get mail?
[12:52] <elmo> lifeless: or the fact you get "post from a non-subscriber" whines from mailman?
[12:52] <lifeless> elmo: no, I mean that say I open a bug in malone on bzr, malone then tells the members of the bzr-developers team, which includes me.
[12:52] <lifeless> rather than telling the *other* members.
[12:52] <elmo> ah, ok, sorry, don't mind me then
[12:52] <mpt> de_wizze, assuming you mean IE for Windows, I don't have it, so any help you can provide in debugging that problem would be great :-)
[12:52] <lifeless> likewise, if you open a bug, you always get a mail, which users count as spam.
[12:52] <lifeless> we had a guy yesterday in #bzr
[12:53] <lifeless> we asked him to file a bug
[12:53] <lifeless> he signed up to lp
[12:53] <lifeless> filed the bug
[12:53] <lifeless> changed the status
[12:53] <lifeless> complained about spam, and changed his account's email to point to an address he never ever uses
[12:53] <lifeless> asked if there was some way he could 'deregister'
[12:53] <elmo> lifeless: err, you can't fix that
[12:53] <jamesh> lifeless: #8 on https://wiki.launchpad.canonical.com/MaloneRunsUbuntuTaskList
[12:54] <lifeless> elmo: its a bug, not a feature, its very fixable.
[12:54] <elmo> well, in the sense, that you're inevitably going to get some mail when you report a bug
[12:54] <lifeless> jamesh: sweet.
[12:54] <de_wizze> mpt: ok will do
[12:54] <elmo> either when it's closed or a request for more information
[12:54] <lifeless> elmo: in bugzilla, when *I* change a bug I'm subscribed to, *I* dont get mail.
[12:54] <elmo> if someone reacts that violently to a confirmation mail ...
[12:54] <mpt> elmo, it's the "don't mail me about stuff *I* did" bug
[12:54] <lifeless> elmo: no, it was the 5 follow ons from lp-itself.
[12:54] <elmo> meh, ok
[12:55] <elmo> (still think he's overreacting tho :P)
[12:55] <jamesh> lifeless: you mean 5 emails for 5 changes, or something else?
[12:55] <lifeless> I haven't quite gotten pissed off enough to blacklist lp mails to me
[12:55] <de_wizze> now about the new placement of the logout button in the upper right corner ...
[12:55] <mpt> bug 548
[12:55] <Ubugtu> Malone bug 548: "don't send mails about changes to the people doing the change" Fix req. for: malone (upstream), Severity: Normal, Assigned to: Brad Bollenbach, Status: Confirmed http://launchpad.net/bugs/548
[12:55] <lifeless> jamesh: open the bug, change the priority, change the severity, update the description etc
[12:55] <mpt> bug 1350
[12:55] <Ubugtu> Malone bug 1350: "Change notifications should be batched" Fix req. for: malone (upstream), Severity: Wishlist, Assigned to: Nobody, Status: Unconfirmed http://launchpad.net/bugs/1350
[12:56] <lifeless> the first one is not needed - he knows he filed the bug, and none of the rest are needed either.
[12:56] <lifeless> mpt: yeah, both of those.
[12:56] <mpt> and making more changes possible in a single step in the Web interface will help too
[12:58] <jamesh> lifeless: okay, so 5 form submits.  That's the "normal" malone bug spam, rather than a new issue
[12:58] <lifeless> jamesh: I did not claim it was new ;)
[12:58] <lifeless> jamesh: I just asked where it was in the priority list, which mpt answered well
[01:00] <mpt> de_wizze, report bugs about Launchpad in general at https://launchpad.net/products/launchpad/+filebug
[01:01] <mpt> seb128, RT ticket sent
[01:03] <seb128> thank you
[01:05] <jamesh> lifeless: any progress in getting rocketfuel changes to sync back to chinstrap?
[01:05] <lifeless> Using saved location: sftp://chinstrap.ubuntu.com/home/pqm/archives/rocketfuel/launchpad/devel/
[01:06] <lifeless> [=                                           ]  fetch revision  12/562 129:42:30
[01:06] <jamesh> would it be worth switching to rsync push in the short term?
[01:07] <lifeless> jamesh: no, that will cause serious confusion
[01:07] <jamesh> fair enough
[01:07] <lifeless> jamesh: rsync is not safe for concurrent access to a branch, and there are cron scripts doing exactly that
[01:08] <lifeless> its 500 revs because the fixed push on thursday was aborted too
[01:08] <elmo> lifeless: I'm sorry about killing pqm, I was sure you once told me pqm was safe to kill, and didn't realise that had changed
[01:09] <lifeless> elmo: no apologies needed.
[01:09] <lifeless> generally it *should* be ok, I suspect there is a bug in bzrlib with respect to leaving stale locks
[01:09] <lifeless> so I have that on my todo list to fix
[01:10] <lifeless> but right now we seem to have pqm killing as a fix for 'its not running my merge' - which is a known clear condition, thus my mail explaining things to the relevant folk.
[01:28] <jamesh> lifeless: is that 129 hour ETA for the push accurate?
[01:29] <lifeless> jamesh: no
[01:29] <jamesh> good :)
[01:29] <lifeless> its probably about twice that
[01:31] <jamesh> I wonder if rsyncing the branch to chinstrap and then doing a local push would work better
[01:39] <lifeless> 08:07 < lifeless> then we would need to run bzr in ssh on chinstrap
[01:39] <lifeless> 08:07 < lifeless> thats getting awfully complex
[01:39] <lifeless> 08:07 < lifeless> compared to just fixing the root cause.
[01:39] <lifeless> 08:07 < lifeless> and time spent on that is time not spent on the root cause.
[02:03] <de_wizze> I have a simple question ... what was the reason for creating launchpad?
[02:11] <spiv> de_wizze: To help open source projects collaborate.
[02:12] <spiv> de_wizze: See also the introductory text at https://wiki.launchpad.canonical.com/
[02:15] <jamesh> the other reason was to provide infrastructure to help develop Ubuntu
[02:19] <de_wizze> ok ... so one reason is in fact to help bridge and help coordinate upstream development and dispersion of projects .. correct ?
[02:21] <jamesh> yep
[02:28] <de_wizze> it kinda like what Conary strives to do, but on a broader scope? will they be or could they be complimentary?
[02:56] <mpt> de_wizze, broader scope, because Launchpad covers bug tracking, translations, bounties etc as well as package management
[03:00] <de_wizze> thats wonderful ...
[03:30] <skyblue> hi all
[03:30] <skyblue> I'm trying to merge some duplicate accounts with mine, but one of those accounts belong to an e-mail address I don't have access to anymore
[03:30] <skyblue> what am I supposed to do now?
[03:31] <skyblue> (I guess that address is bouncing now)
[03:32] <jamesh> could someone with admin access perform an administrative person merge for skyblue?
[03:33] <skyblue> jamesh: thanks
[03:56] <skyblue> for anyone who may want to take a look at it, my correct/active account name is: "ealtin" and I'd like to merge "enver-altin" and "enver-altin-frontsite" to it.
[03:58] <jamesh> mpt/spiv: maybe you could help skyblue? You are both on the admins group
[03:58] <elmo> jamesh: if it's webbased, point me at the url/right way and I can
[04:00] <jamesh> elmo: I was under the impression that it was web based.  I'll see if I can find the URL
[04:01] <elmo> there's no obvious linksfor it from a person's /people/ page, so I'm assuming it's magic URL
[04:03] <jamesh> maybe it is something stub does directly
[04:04] <stub> There is a magic URL somewhere...
[04:04] <stub> I can't really do it directly
[04:04] <elmo> stub: hey - how did mizuho move without any DNS changes? :/
[04:05] <spiv> jamesh: I don't know the magic URL, sorry
[04:05] <stub> elmo: Because the only external DNS name being used to access it is librarian.*.com, so just needed to change the apache conf.
[04:06] <elmo> oh, right, I'd entirely forgotten the apache front end was on another machine, sorry
[04:06] <elmo> stub: do you use librarian.*.com internally too?
[04:08] <stub> Sometimes :-) Generally the real host name is used because we need to access the upload port, so going via apache is silly
[04:09] <elmo> ok, then I'd like to create a librarian.internal for that?
[04:09] <elmo> I'd like to avoid hardcoding the penguin/antartic hostnames in services as much as poss
[04:09] <stub> skyblue: What is the email address that you no longer have access to?
[04:10] <skyblue> stub: One is enver.altin@frontsite.com.tr
[04:10] <skyblue> stub: The other is somehow got incorrectly imported from somewhere, possibly something like enver.altin@frontsite.com
[04:11] <skyblue> stub: My current preferred address is ealtin@parkyeri.com if it helps.
[04:15] <stub> skyblue: What happens if you try and merge one of those accounts? Does the system tell you it is sending an email for validation or something?
[04:17] <stub> skyblue: Merge will hopefully work now anyway
[04:25] <skyblue> stub: Yeah, it does. It says I'll need to follow the link submitted to me.
[04:25] <stub> Bug 1281
[04:25] <skyblue> stub: but the problem is, I don't have access to these addresses.
[04:26] <stub> skyblue: Try again now - I've removed those email addresses from those accounts. I think merge will either work or you will get an OOPS error
[04:26] <skyblue> stub: thanks, will do.
[04:28] <skyblue> stub: Accounts don't seem to exist at all. They don't show up in search results.
[04:28] <stub> Ahh....
[04:29] <skyblue> stub: that bad? :)
[04:30] <stub> skyblue: If there is no other reason to merge the accounts that remove the duplicates, then we are done. If those accounts had extra rights (owner/maintainer of products, bugs, packages, whatever), then that still needs to be sorted.
[04:31] <skyblue> stub: I guess they were gathered from translations and bug databases of some other projects.
[04:32] <jamesh> if that's the case, then they might get recreated on future data imports
[04:32] <skyblue> stub: I'm not really sure if they had some rights, but even if they had, they are not really important.
[04:32] <skyblue> jamesh: Yeah, possibly.
[04:33] <jamesh> you really want those addresses associated with the main account, but marked as OLD
[04:33] <skyblue> jamesh: if you say so.. :)
[04:34] <jamesh> maybe not: the description of OLD says "... nor should we associate new incoming content from that email address with that person."
[04:35] <skyblue> jamesh: apparently yes. associating them and marking as OLD looks good to me, but that's going to be a bit painful I guess :)
[04:36] <skyblue> how do translated content get upstream?
[04:37] <skyblue> are they merged to upstream repositories automatically by launchpad?
[04:46] <jamesh> upstreams need to perform the merges
[05:22] <lifeless> jamesh: up to a review ?
[05:26] <lifeless> jamesh: my launchpad/story branch, now in your review queue.
[05:51] <dilys> Merge to devel/launchpad: [trivial]  use re.escape to prevent process-email from crashing. (r3004: Bjorn Tillenius)
[05:54] <jamesh> lifeless: okay.  I've also got BjornT's TicketTrackerEmailInterface branch to finish reviewing (it is currently on Steve's queue, but I agreed to take it)
[05:59] <lifeless> jamesh: sure
[06:00] <lifeless> jamesh: it should be trivial though
[06:10] <jamesh> lifeless: btw, I put my branch with the newline fixups here: http://people.ubuntu.com/~jamesh/pqm-newlines/
[06:12] <jamesh> lifeless: I had a bit of trouble working out how to fit a test for this into the existing test framework, so there is no explicit test for the newline handling of test suite output
[06:14] <lifeless> I would create a script which will run a precommit command that fails
[06:14] <lifeless> in the unittest tests
[06:14] <lifeless> that is
[06:42] <lifeless> jamesh: actually
[06:43] <lifeless> jamesh: I would break it into two
[06:43] <lifeless> a test that a formatting function does the right thing with sample 'output from a script'
[06:43] <lifeless> and a test that a failing script calls the formatting method (by using a failing script and checking its called)
[06:45] <jamesh> lifeless: well, the formatting function (gather_output) is now just a ''.join(output) call
[06:45] <jamesh> lifeless: the rest of the changes were to normalise the items added to the output list to end in newlines where appropriate
[06:45] <jamesh> previously gather_output was effectively '\n'.join(output)
[06:47] <lifeless> sure
[06:47] <lifeless> not critiquing your work, just saying how I would test it
[06:50] <jamesh> I also altered the test for catching bzr conflicts a bit
[06:50] <jamesh> since it was failing with current bzr
[06:51] <jamesh> bzrlib seemed to have trouble listing conflicts for ID-only conflicts
[06:51] <jamesh> the test tried to merge two branches that had both added a file called README
[06:51] <jamesh> I changed the test to try and merge two branches that had edited an already existing file called README
[06:57] <stub> lifeless: Do you know how long it takes between pqm announcing a new revision being committed and that revision being commited to bellany:~pqm/archives/launchpad ?
[06:59] <jamesh> stub: the ETA was 129 hours earlier today
[07:00] <stub> jamesh: That would be to chinstrap - I'm interested in balleny
[07:00] <stub> (which is currently up to r3003)
[07:08] <jamesh> stub: do you remember what config settings you changed for the launchpad-bugs list to get it to accept LP emails?
[07:08] <jamesh> stub: I put what I remembered at the bottom of https://wiki.launchpad.canonical.com/BugzillaImportProcess
[07:09] <stub> I  appended it to one of the bug reports. You need to add an 'allow' spam filter for bugs.launchpad.net
[07:10] <stub> But that involves someone with access to ubuntu-bugs actually making the change :-/
[07:13] <jamesh> judging by http://lists.ubuntu.com/archives/desktop-bugs/2006-January/thread.html, it looks like desktop-bugs was just set to unmoderated
[07:13] <jamesh> given the rolex spams
[07:16] <lifeless> stub: that should be almost no latency, except that I have a lock while I push the archive to chinstrap
[07:17] <lifeless> stub: I've stopped the push
[07:17] <lifeless> stub: lets see now
[07:24] <lifeless> stub: it should be there
[07:25] <lifeless> ring me if needed, doctor visit time
[07:28] <stub> np
[09:03] <dooglus> hi.  I'm trying to register the irssi bugtracker with malone, but it's a "Flyspray" tracker, which malone doesn't know about.
[09:04] <dooglus> http://flyspray.rocks.cc/ -- Flyspray is an uncomplicated, web-based bug tracking system for assisting with software development.
[09:06] <stub> launchpad will be going down in just over 5 minutes for its regular update. Estimated down time is 20 mins.
[09:06] <Burgundavia> dooglus, file a bug, product launchpad
[09:06] <dooglus> Burgundavia: ok, thanks
[09:09] <dooglus> is it possible to automatically subscribe to any bug I comment on?
[09:09] <Burgundavia> dooglus, I wish it just did that
[09:09] <Burgundavia> mpt_, ^ two up, implement this feature
[09:14] <dooglus> the 'flyspray' bug is here: https://launchpad.net/products/launchpad/+bug/28738
[09:14] <Ubugtu> Error: Could not parse data returned by Malone: Connection to Malone bugtracker failed: HTTP Error 503: Service Unavailable
[09:15] <dooglus> will launchpad go down every day for its regular update?
[09:15] <Burgundavia> dooglus, not everyday
[09:16] <jamesh> stub: btw, there were a few duplicate bug tracker objects created during the bugzilla import
[09:16] <stub> dooglus: It is a weekly update
[09:17] <jamesh> stub: I cleaned it up a bit by moving the watches to a single bug tracker, but it left a few empty bugtrackers that would be good to delete
[09:17] <stub> ok. We can have a look after the update
[09:20] <dooglus> I noticed that gnome is listed twice in the bugtrackers
[09:21] <dooglus> what does the update update?
[09:22] <Burgundavia> dooglus, the production version of launchpad, from the testing version
[09:22] <jamesh> dooglus: that's what I was just talking to stub about
[09:22] <jamesh> dooglus: when migrating bugs from bugzilla, we created watches on external trackers where appropriate
[09:23] <jamesh> dooglus: while most of the references to upstream Gnome bugs used the correct http://bugzilla.gnome.org hostname, some used theincorrect http://bugs.gnome.org URLs
[09:50] <sivang> morning all
[09:55] <dooglus> hrm.  now I see: DatabaseException
[09:55] <dooglus> A server error occurred. 
[09:56] <dooglus> should I report it?
[10:10] <stub> dooglus: Sorry about that - final stages of the upgrade. Should all be fine now.
[10:11] <dooglus> also, when it came back up again, the first few pages it generated for me said "I'll be going down very very soon"...
[10:11] <dooglus> I also have a problem with the page layout: http://librarian.launchpad.net/1518542/Screenshot-3.png
[10:14] <jamesh> stub: the bug trackers that should be deleted are "auto-bugs.gnome.org" and "auto-bugzilla.abiword.com"
[10:34] <stub> jamesh: Nuked
[10:44] <lifeless> jamesh: so, any chance to peek at that review ? (I'm keen, I know)
[11:50] <Kinnison> lifeless: what's the situation on pqm pushing changes to chinstrap?
[11:50] <Kinnison> lifeless: in particular the built tree?
[11:50] <lifeless> Kinnison: its pushing at the moment
[11:50] <lifeless> Kinnison: it had a stale lock, then when it was pushing to fix it got killed giving it another stale lock
[11:51] <Kinnison> lifeless: yum
[11:51] <lifeless> so, in another 80 odd hours it will have pushed all the backlog
[11:52] <Kinnison> urgh
[11:52] <Kinnison> lifeless: how come sftp over a gigabit lan is still so goddamn slow?
[11:53] <lifeless> haven't we had this discussion ?
[11:53] <Kinnison> I don't remember why, if at all, you said we're not using bzrtools/rsync-push though
[11:54] <lifeless> rsync does not lock
[11:54] <lifeless> our branches are multi user
[11:54] <lifeless> nuff said
[11:54] <Kinnison> righty, yep
[11:56] <stub> They are multi user? If that is many readers/single writer that could be fixed by rsyncing to temporary chinstrap and then localfs pushing to the final location (?)
[12:00] <lifeless> 08:07 < lifeless> then we would need to run bzr in ssh on chinstrap
[12:00] <lifeless> 08:07 < lifeless> thats getting awfully complex
[12:00] <lifeless> 08:07 < lifeless> compared to just fixing the root cause.
[12:00] <lifeless> 08:07 < lifeless> and time spent on that is time not spent on the root cause.
[12:01] <ddaa> hey lifeless
[12:03] <lifeless> stub: its been out since before thursday
[12:03] <lifeless> and in fact, local disk would not be all that much faster
[12:03] <stub> I've pushed head a few times to /tmp on chinstrap. An rsync could be a good idea as a temporary solution.
[12:03] <lifeless> I'm stracing the process, and the link is largely idly
[12:03] <lifeless> *idle*
[12:09] <daf> mpt_: yo
[12:11] <Kinnison> lifeless: urgh, so it's mostly diddling weaves?
[12:12] <lifeless> Kinnison: unpacking inventories
[12:12] <lifeless> although this is interesting:
[12:12] <lifeless> pqm@chinstrap:~$ cat sftp-strace | uniq | wc -l
[12:12] <lifeless> 3454
[12:12] <lifeless> pqm@chinstrap:~$ cat sftp-strace | wc -l
[12:12] <lifeless> 3518
[12:12] <lifeless> I suspected there was some duplicate effort, and now I have some proof ;)
[12:13] <lifeless> or not
[12:13] <lifeless> bah
[12:13] <lifeless> just new lines being written.
[12:13] <Kinnison> hehe
[12:13] <Kinnison> silly lifeless
[12:13] <Kinnison> One problem with stracing python is that the gc often means that the "same" operation looks different on a plain "uniq"
[12:13] <Kinnison> also, try sort sftp-strace | uniq | wc -l
[12:14] <Kinnison> rather than cat
[12:14] <lifeless> oh, right
[12:14] <lifeless> this is stracing the sftp server btw
[12:14] <lifeless> not python
[12:14] <lifeless> python is the other strace
[12:15] <lifeless> voila
[12:15] <lifeless> there is duplicate work.
[12:15] <lifeless> cat sftp-strace |  sort | uniq -d | wc -l
[12:15] <lifeless> 1470
[12:19] <Kinnison> lifeless: is this using jamesh's sftp-using-openssh stuff?
[12:20] <lifeless> dunno, would not expect that to matter
[12:20] <Kinnison> probably not, just wondered
[12:21] <daf> hi salgado 
[12:22] <salgado> hey daf
[12:22] <daf> is it you who's looking after shipit?
[12:22] <salgado> yep
[12:22] <daf> I found some weirdness in shipit-reports.pt
[12:23] <daf> near the bottom, there's some broken HTML
[12:23] <matsubara> good morning!
[12:23] <salgado> hmmm. /me checks
[12:24] <salgado> daf, where's it exactly?
[12:25] <daf> lines 39-46
[12:25] <daf> there's no start to that table
[12:25] <daf> oh, wait
[12:25] <salgado> line 11
[12:26] <daf> yeah, I see it now
[12:26] <daf> it's just a missing </tbody> tag
[12:26] <daf> fixed
[12:26] <salgado> all shipit pages have that table because of the ubuntu css files we have to use
[12:26] <salgado> the </tbody> might be missing in other shipit pages too
[12:29] <lifeless> night all
[12:29] <Kinnison> night lifeless
[01:02] <Kinnison> shelve is good, but I also miss baz undo/redo
[01:02] <daf> yeah
[01:03] <daf> I miss being able to shelve renames
[01:03] <daf> but I just did a hunk-by-hunk review of 123 changes to 77 files that would have taken me much longer with baz
[01:04] <Kinnison> heh
[01:04] <daf> I also miss your cake
[01:04] <Kinnison> So you used it as a diff hunk browser?
[01:04] <Kinnison> daf: aah, one little bite and anyone is hooked
[01:04] <daf> well, I just made a lot of changes
[01:04] <Kinnison> daf: If you like, I can assist you in the preparation of another cake
[01:04] <daf> some of which I'm confident in, others not
[01:05] <daf> I want to commit the safe ones now, and keep the rest for later
[01:05] <daf> where's the recipe?
[01:06] <daf> I suppose one advantage of this kind of cake is that you can take as long as you like in making it
[01:06] <daf> I think most of the intermediate stages are quite stable
[01:09] <kiko> man
[01:09] <kiko> 3 days with no head
[01:09] <kiko> wtf
[01:10] <Kinnison> kiko: you must have been bursting
[01:10] <Kinnison> oh, you mean source code
[01:10] <kiko> I mean having a small heart attack
[01:10] <Kinnison> daf: Well, there's one intermediate stage which is stable
[01:10] <daf> *snort*
[01:10] <Kinnison> daf: then there's the "cake is baked, can I resist eating it" stage
[01:10] <Kinnison> which I don't define as stable 'cos I'm greedy
[01:10] <daf> :)
[01:16] <daf> I was thinking in terms of making the almond paste and boozed fruits separately
[01:16] <daf> then again, impatience will have its part to play
[01:16] <kiko> something about bullets
[01:18] <daf> you can count on Ian not to mince words
[01:27] <kiko> daf, can you check if http://shipit.ubuntu.com/ works for you?
[01:28] <daf> works in what sense?
[01:28] <daf> I get a login page
[01:28] <kiko> log in and view order request form
[01:29] <daf> I see the form
[01:29] <kiko> thanks.
[01:29] <daf> browser was complaining about some parts being loaded over an insecure connection
[01:29] <daf> which is annoying
[01:30] <kiko> stupid lack of certificate and the ubuntu css we use
[01:30] <kiko> salgado, can we just not make a copy of the ubuntu css and put it in as shipit.css instead of referring to it?
[01:30] <daf> having validate certificates for user-facing sites would be nice
[01:31] <daf> I don't really care about wiki.canonical.com and such, but I'm sure having to view a certificate warning when visiting launchpad.net is off-putting
[01:35] <kiko> well
[01:35] <kiko> launchpad.net itself has a cert
[01:35] <kiko> www.launchpad.net doesn't however
[01:35] <daf> !
[01:35] <kiko> valid cert anyway
[01:35] <kiko> you know that it requires having two IPs right?
[01:35] <daf> oh, right
[01:35] <daf> SSL fuckage
[01:36] <kiko> yes, let's drink to fuck
[01:50] <ddaa> kiko... man...
[01:50] <kiko> hey ddaa
[01:50] <kiko> what did I do now?
[01:50] <ddaa> you are answering mails 5 weeks old, with only minimal context...
[01:50] <kiko> really?
[01:51] <kiko> which one in particular?
[01:51] <kiko> I did some cleaning out of my email yesterday
[01:51] <ddaa> I'll come around answering your message on the "Add branch action link" thread when I read the launchpad mailing list, I'll have more context then.
[01:51] <kiko> so you may have gotten some of that
[01:51] <ddaa> I'm doing the inbox now, and that's a bit off-putting :)
[01:53] <ddaa> kiko... I guess you did NOT mean bug 959
[01:53] <Ubugtu> Error: Could not parse data returned by Malone: 'NoneType' object has no attribute 'group'
[01:53] <ddaa> which is an inkscape bug...
[01:53] <kiko> 929
[01:54] <ddaa> Oh right, I see.
[01:54] <ddaa> My position is that one should be WONTFIX.
[01:55] <ddaa> As I explained in the last comment.
[01:55] <daf> looks like Ubugtu is buggy
[01:55] <kiko> I think it needs to be fixed
[01:55] <ddaa> kiko: the _only_ sane way to fix it is using <wbr>.
[01:55] <kiko> and given your suggestion is getting rid of the RHS portlets..
[01:56] <ddaa> and that's only sane if you accept using invalid html.
[01:56] <ddaa> and even then, that can only be an half-assed fix that will require pervasive changes.
[01:57] <kiko> why will it require pervasive changes?
[01:58] <ddaa> because it will need support in anything that can display user input
[01:58] <kiko> putting it in DPoT is enough
[01:58] <ddaa> ?
[01:59] <ddaa> DPoT: ENOENT
[01:59] <kiko> welcome to launchpad
[01:59] <daf> Displaying Paragraphs of Text
[01:59] <ddaa> Yeah, so breaking will happen at random places...
[02:00] <daf> making that support zero-width breaking spaces would help
[02:00] <ddaa> !!!!
[02:00] <ddaa> DUH DON'T!
[02:00] <daf> last I checked though, browser support was poor
[02:00] <ddaa> anyway it's WRONG
[02:00] <daf> why?
[02:00] <ddaa> Because it breaks copy-pasting
[02:00] <kiko> I remove myself from this conversation
[02:01] <daf> hmm, I'd want the browser to strip them when copying
[02:01] <daf> I don't get <b> tags when I copy bold text
[02:01] <ddaa> I've seen sabdfl copy-pasting an URL that was using zero-width space for line breaking.
[02:01] <daf> iz browser bug
[02:01] <ddaa> Because <b> is a tag. not a character. And besides you should get it if you paste to a word processor.
[02:01] <daf> whatever, <wbr>
[02:02] <daf> is that an IE extension?
[02:03] <ddaa> Ha, right. That's less wrong because browser bugs mean it will NOT be copy-pasted. But it's not valid HTML, and anyway I would not like my urls to get arbitrary broken in the middle of words.
[02:03] <ddaa> Whatever, you guys do whatever you think is best. I have made my opinion clear on that matter.
[02:04] <daf> well, if zwbs worked well in browsers, I'd advocate it
[02:04] <ddaa> copy-pasing zwbs is _correct_ behaviour
[02:05] <ddaa> stuffing it at random places in user data is what is usually called data corruption
[02:06] <daf> ok
 would seem the way to go if it was standard
[02:07] <sivang> what's the quickets way to file a bug on malone from hitting launchpad.net ?
[02:07] <kiko> sivang, navigate products or distros?
[02:07] <daf> I rely on the fact that https://launchpad.net/products/malone/+filebug is in my often-used-URL list
[02:08] <sivang> daf: I need to book mark that, everytime I hit the main page it takes me some minutes of orientation :)
[02:08] <sivang> kiko: yes, and then search for malone
[02:08] <sivang> btw, when searchign for malone I also got https://launchpad.net/products/launchpad , is that used for anythying?
[02:09] <daf> yes, that's where most LP bugs are filed
[02:09] <kiko> sivang, that's used for lots
[02:09] <sivang> kiko: for launchpad wide stuff and not launchpad/$COMPONENT stuff?
[02:09] <sivang> like as in infra, utilities, etc
[02:09] <ddaa> daf: I think wbr would be acceptable if it was standard and defined as being a display-only hint.
[02:10] <daf> indeed
[02:10] <daf> removing the right hand column is only a partial solution
[02:10] <daf> you can still get lines which require side-scrolling
[02:10] <daf> side-scrolling is evil
[02:11] <ddaa> but in the current state of affair, I think it's best to just let long urls run past the left end of the window, anyway, I do not think you like getting 5 lines of goo in the middle of paragraph of text just because somebody pasted a bugzilla url.
[02:12] <daf> another option is to turn http/foo.com/bar into a link with abbreviated text
[02:12] <ddaa> Yes, it's evil, but I do not think that if a string would require side-scrolling, you actually want to read it.
[02:12] <ddaa> That the basic idea. If a word is too long to fit in the window, it's too long for a human to read anyway.
[02:12] <daf> I think long URLs are the most common case of long lines
[02:12] <ddaa> Do you actually _read_ long URLs?
[02:13] <daf> no
[02:13] <daf> but if there's a list of them, I might compare the ends
[02:13] <daf> or rather, compare them
[02:13] <kiko> I think we need to be more practical if we are going to move anywhere
[02:14] <ddaa> in which case it is easier to do?
[02:14] <daf> well, URLs contain useful information sometimes
[02:14] <ddaa> if they are wrapped of if they are aligned, each on a line?
[02:14] <daf> kiko: I vote for abbreviating URLs in link text as the next step
[02:14] <kiko> copying and pasting the text of the URL is a use case I'm willing to drop
[02:14] <kiko> right-click or G (in lynx) for the original link
[02:15] <ddaa> kiko: I really want to display URLs in the branch page.
[02:15] <kiko> ddaa, you will be able to -- wrapped or abbreviated.
[02:15] <daf> I don't see any comment from mpt on 929
[02:16] <daf> or any discussion about the possibility of using the overflow property
[02:16] <ddaa> we'll see if I can hold it.
[02:17] <daf> kiko: let's ping mpt on it
[02:17] <kiko> sure
[02:17] <daf> we have a few options, he's probably best at picking the least bad
[02:19] <sivang> there, https://launchpad.net/products/malone/+bug/28768
[02:19] <Ubugtu> Error: Could not parse data returned by Malone: 'NoneType' object has no attribute 'group'
[02:20] <sivang> now I wonder if there's a quick way to fle a bug against bzrtools upstream, without following the package
[02:23] <ddaa> sivang: tell abentley in #bzr :)
[02:23] <sivang> ddaa: heh
[02:24] <ddaa> besides, there does not seem to be a bzrtools product in launchpad.
[02:24] <sivang> ddaa: yes, seems so, do you think there should be ? or should we tell him to add the branch under https://launchpad.net/products/bzr ?
[02:25] <ddaa> Whatever you think is best. Probably ask abentley, he's upstream after all. Maybe he would be interested in having a separate bug tracker and stuff.
[02:25] <ddaa> If upstream is not interested in having a launchpad project, it's probably best to put the branches in the bzr product.
[02:25] <sivang> ah, bzrtools is not canonical related?
[02:26] <LarstiQ> policy until now has been, bzrtools bugs to the mailinglist, afaik
[02:26] <LarstiQ> sivang: correct
[02:26] <ddaa> No, it's entirely community contributed.
[02:26] <sivang> s/bzrtools/abently/
[02:26] <LarstiQ> sivang: but ask abently
[02:26] <sivang> bah... I give up :-)
[02:26] <ddaa> well... he was actually proposed a position here, like, many times
[02:26] <sivang> too far for me for now
[02:27] <ddaa> but he likes it where he is now, so though luck :)
[02:27] <sivang> I see :) sometimes we just need to make the best out of what we have
[02:27] <LarstiQ> sivang: yeah, I feel sorry for you that you only have ddaa 
[02:28] <ddaa> LarstiQ: actually, no, they have mpool, and lifeless, and jblack
[02:28] <ddaa> I'm not actually a significant bzr dev.
[02:28] <ddaa> I just try to convince people to implement the stuff I would like :)
[02:29] <ddaa> and I hang around the bzr guys at conferences, trying to sound like I know what I'm talking about.
[02:29] <LarstiQ> haha :)
[02:29] <LarstiQ> ddaa: a far better response than biting, thank you :)
[02:29] <daf> is pqm wedged -- it's got 6 things in the queue and the last merge was 9 hours ago
[02:30] <LarstiQ> ddaa: I rather appreciate your comments though
[02:34] <ddaa> it's easy
[02:34] <ddaa> you just need to learn how to make gnome packages in your sleep
[02:35] <LarstiQ> heh
[02:35] <ddaa> (that was re: sivang is now known as seb64-wannabe)
[02:35] <daf> elmo: can you kick pqm?
[03:00] <ddaa> mh... according to jblack, the gplv3 draft no longer has any obnoxious web service clause...
[03:07] <kiko-afk> daf, elmo: are you sure pqm is hung?
[03:07] <kiko-afk> lifeless scorched me yesterday because what's happening is probably bzr push taking 90 hours (no joke)
[03:09] <daf> !!!
[03:09] <ddaa> zat's better than 90 days, as for importd2bzr...
[03:09] <ddaa> (please ignore this, just venting)
[03:09] <daf> is the fact that rsyncing something to chinstrap is taking forever?
[03:10] <daf> efnewkjfn
[03:10] <daf> is the fact that rsyncing something to chinstrap is taking forever related?
[03:10] <bradb> pqm will need serious optimization to reach the point of being unusably slow
[03:10] <ddaa> bradb: do you really mean that? What is it now?
[03:10] <bradb> ddaa: Yeah, I most certainly mean that. ;)
[03:10] <kiko> daf, it might be. don't touch pqm for the next 90 hours
[03:11] <bradb> ddaa: I think kiko and daf have some idea of what I mean ;)
[03:11] <daf> this is plain rsync, not bzr push
[03:11] <bradb> daf: the mirroring is still done with sftp isn't it?
[03:12] <spiv> daf: according to "vmstat 1" on chinstrap, there's a lot of IO happening, but I suspect that's due to check-pending-reviews.py
[03:12] <daf> aha
[03:12] <spiv> bradb: yeah, it is.
[03:12] <daf> it's an ssh bug
[03:12] <daf> I think
[03:13] <daf> I'm using ControlMaster auto and I had an existing ssh to chinstrap which was suspended
[03:13] <daf> killing that made rsync work again
[03:58] <ddaa> daf: I tried using ssh session muxing at a point for importd
[03:58] <ddaa> daf: the conclusion of the test was "it just does not work"
[03:59] <daf> I had problems with it
[03:59] <ddaa> IIRC there was a fd leak or something that just caused all new connections to fail past a certain point, but apparently that was only caused by the specific loads applied by baz.
[03:59] <daf> upgrading to dapper ssh helped
[03:59] <ddaa> I guess they have fixed some things.
[03:59] <daf> then I got very confused because it stopped port forwardings working
[03:59] <daf> then my rsync broke
[03:59] <daf> now I'm thinking it's not worth the trouble
[04:00] <ddaa> *nod* I'll check it again in a couple of years, when the *bsd folks had the time to squash the bugs
[04:01] <salgado> BjornT, around?
[04:01] <ddaa> they just spoiled us with openssh, now are just disappointed if ssh is not 99.9999999999 reliable
[04:38] <Panda^> hi everyone
[04:39] <Panda^> is there someone of the launchpad's technical stuff? I've a problem loggin in...
[04:51] <carlos> Panda^, go ahead and if we can help, we will help you
[04:56] <Panda^> carlos, tnx. I can't rember my account's password... but the email address that I used to create my account no longer exists...
[04:57] <carlos> hmm, Panda^ what's your account's name?
[04:58] <Panda^> I registered another new account, but I can't merge it with the old because it ask for the old password that I can't remember
[04:59] <Panda^> giampaolo@openmagazine.it
[04:59] <Panda^> carlos, giampaolo@openmagazine.it is the one I don't remeber the password
[05:00] <Kinnison> that one only has a single email address registered with it
[05:00] <Panda^> yes
[05:00] <carlos> Panda^, and the new email is giampaolo.bozzali@gmail.com ?
[05:00] <Panda^> carlos, exactly
[05:00] <Kinnison> openmagazine.it's MX accepts mail for giampaolo@openmagazine.it
[05:01] <Kinnison> Can't you convince them to forward the address to you temporarily?
[05:01] <Panda^> no, i can't
[05:02] <carlos> salgado-lunch, Kinnison aren't we allowed to force accounts merge as admins?
[05:02] <Kinnison> I don't think we can do that per-se
[05:02] <Panda^> I can't receive any email from openmagazine account
[05:03] <Kinnison> What's your new account name?
[05:03] <Panda^> the new is: giampaolo.bozzali@gmail.com
[05:05] <daf> kiko: do you remember which bug crashed on context/status/title (context.status == None)?
[05:06] <Kinnison> Hmm, I can find nothing on the old account which can match to the new one apart from public name information
[05:06] <Kinnison> and I wouldn't trust that per-se
[05:06] <bradb> daf: That sounds like something that would have happened before stub did the status data migration. It would have happened with a Rejected bug.
[05:07] <Kinnison> I don't think I have the right to make a policy decision on sorting this out
[05:07] <Kinnison> Panda^: I can think of various ways to help you out but I don't feel I have the right to do it without checking with someone else first
[05:07] <daf> bradb: this is for binary packages, not bugs
[05:07] <bradb> oh
[05:08] <daf> haha, found it: bug 5390
[05:08] <daf> I filed it myself
[05:08] <daf> (google site:launchpad.net wins)
[05:08] <bradb> indeed
[05:11] <carlos> Kinnison, I think the easier solution, if we have permissions as admins is to mail both accounts
[05:12] <Kinnison> carlos: we can't mail one of them
[05:12] <Kinnison> carlos: because the only registered address is no longer functional
[05:14] <Panda^^> as admins you can't change an account's password?
[05:14] <Kinnison> Even if I can, I can't directly trust that you are who you say you are
[05:14] <Kinnison> since there's no useful personal data on the obsolete account
[05:14] <Kinnison> hence I said that I don't feel I have the right to make a policy decision on this
[05:15] <Kinnison> Do you see that I have no way of being sure that you *are* the person who registered the original account?
[05:16] <Panda^^> of course
[05:16] <Kinnison> I'm sorry to be so hard-lined about this, but imagine how much trouble we'd be in if you were trying to trick us
[05:17] <Kinnison> carlos: Do we have a way of seeing when someone last logged into launchpad?
[05:18] <Panda^^> Kinnison, I understand perfectly...
[05:18] <Kinnison> Panda^: Do you know when you last did anything with the old account?
[05:18] <Panda^^> Kinnison, I don't remember exactly... surely 3 o 4 months ago
[05:19] <carlos> Kinnison, Hmm, I think we have that stored in our database, but I'm not sure
[05:19] <carlos> Translation Suggestion  	2005-11-24
[05:19] <carlos> yeah
[05:20] <carlos> that's what I was looking at
[05:20] <Kinnison> that's two months ago
[05:20] <Kinnison> Panda^^: what were you translating?
[05:20] <Panda^^> gdesklets
[05:20] <Kinnison> true enough
[05:21] <Panda^^> tnx
[05:22] <Kinnison> You never signed a CoC, never supplied a key for ssh or gpg
[05:22] <Kinnison> No hackergotchi, no info
[05:23] <kjcole> Ubugtu help
[05:23] <Kinnison> No hints of you in the wiki
[05:23] <Panda^^> yes, because I never had much time to spend... but in thoose days I would like to translate something
[05:23] <Kinnison> urgh
[05:23] <Kinnison> Okay, I think there's little to lose by allowing the merge
[05:24] <Kinnison> Now, I need an email address for you, which isn't one you already used in launchpad
[05:24] <Panda^^> Kinnison, few minutes ago I registered a new account with giampaolo.bozzali@gmail.com
[05:25] <Kinnison> Yeah, that's used now
[05:25] <Kinnison> do you have any other email addresses?
[05:25] <Panda^^> no
[05:25] <Kinnison> okay, one sec
[05:26] <Kinnison> I've sent you an email, please do what it says
[05:27] <Panda^^> ok
[05:28] <Kinnison> If you're not a registered IRC user, then instead of /msg, do it here in-channel
[05:30] <kiko> daf, no.
[05:32] <kiko> my god
[05:34] <Kinnison> Okay, looks like it's a bug in our protocols
[05:34] <Kinnison> namely we have no contingency for when a user loses their password and control of their only registered email address at one time
[05:34] <kiko> Kinnison, known problem, just mail stub.
[05:35] <Kinnison> kiko: Right
[05:35] <Kinnison> Panda^^: Will you please email stuart.bishop@canonical.com with all the details of your two accounts and which you want to keep etc?
[05:35] <Panda^^> ok
[05:35] <Kinnison> Carry on with your new account for now
[05:35] <Kinnison> and hopefully Stuart can sort it out
[05:35] <Kinnison> He's our database manager
[05:36] <Panda^^> ok
[05:36] <Panda^^> I'll do soon
[05:36] <Panda^^> thanks anyway for your time!
[05:36] <Kinnison> No problem
[05:36] <Kinnison> kiko: is there a bug about this already?
[05:36] <kiko> I believe so
[05:38] <BjornT_> hi salgado, i'm around now
[05:38] <Kinnison> Panda^^: subscribe to bug 1281 if you want to know when it gets fixed
[05:39] <Panda^^> Kinnison, ok
[05:43] <salgado> hi BjornT_, I remember seeing you discussing a problem with the enumvalue namespace in tales. is it a known problem that it doesn't work?
[05:43] <Panda^> Kinnison, sorry... can you tell again the bug number?
[05:43] <Kinnison> 1281
[05:43] <Panda^> tnx
[05:45] <BjornT> salgado: yes, bug 5768
[05:48] <salgado> BjornT, right, I thought that simply adding a "<class class="canonical.lp.dbschema.MirrorContent">\n <allow attributes="name title description items" />\n </class>" in lp/configure.zcml would solve it, but it didn't
[05:49] <salgado> BjornT, if it's a quick fix I'd like to fix it now, but if it's more involved I'd prefer to not spend time on it now, as I have other priorities. what do you think?
[05:58] <daf> kiko: no what?
[05:58] <kiko> daf, I didn't remember. :)
[06:04] <BjornT> salgado: hmm, that should have worked. that probably means that it's not an easy fix. i tried it as well, and couldn't get it to work.
[07:19] <kiko> matsubara, did you notice that stub closed bug 5573?
[07:22] <matsubara> kiko: yep, it's on pqm's queue.
[07:24] <matsubara> kiko: anyway, I had 2 bugs fixed with that patch. I'll remove the 5573 fixes and commit again.
[07:24] <kiko> matsubara, did he use your patch? or what?
[07:26] <matsubara> kiko: nope, I'm reading his patch now. He did a completely new patch
[07:27] <kiko> matsubara, and how does it work?
[07:30] <matsubara> kiko: he added the sftp to the valid_webref arguments, did lots of new tests, and changed the db constraint to check the scheme returned by the urlparse()
[07:32] <kiko> is it similar to what you did?
[07:33] <Kinnison> ciau lunchpadders
[07:33] <matsubara> kiko: a little bit different, and he has lots of tests, which I don't.
[07:33] <kiko> Kinnison, ahm
[07:34] <kiko> can I talk to you before you go?
[07:34] <Kinnison> kiko: if it'll take < 2 minutes, My dinner is almost cooked
[07:34] <kiko> Kinnison, you are in charge of one of the remaining parts of soyuz testing
[07:34] <kiko> Kinnison, and I think you might as well find out what it is so you can suffer over it in due course
[07:34] <kiko> dinner or not
[07:35] <Kinnison> heh, you make it sound like a chore, rather than the joy that it will surely be
[07:35] <Kinnison> go ahead
[07:35] <kiko> matsubara, salgado: don't tell me. let me guess. 90 hours ETA to mirror to chinstrap?
[07:36] <kiko> Kinnison, so mark would like us to run a dapper-mirror based on gina source-only uploads for dapper.
[07:36] <Kinnison> kiko: go on
[07:36] <mdke> how is the launchpad, bug-mailing-list situation being resolved?
[07:36] <kiko> I won't have time to do it, and neither with cprov with the current set of tasks -- we have the full-dapper test to do.
[07:37] <kiko> Kinnison, so you have been picked out as the man to coordinate and get it done.
[07:37] <kiko> it shouldn't be difficult
[07:37] <Kinnison> seems fairly sane
[07:37] <kiko> but I don't know how the buildd situation will be
[07:37] <kiko> and you only have threeish days
[07:37] <Kinnison> Right
[07:38] <Kinnison> Are we doing this on staging or dogfood?
[07:38] <kiko> Date: Wed, 11 Jan 2006 13:04:47 +0000
[07:38] <kiko> that's the email, from Mark, you are looking for
[07:38] <Kinnison> urgh, evo is currently sat thrashing
[07:38] <kiko> it will require: one buildd. mirroring uploads to staging. publishing.
[07:38] <kiko> and you can't use dogfood (or its builders I imagine)
[07:38] <Kinnison> can you mail me to remind me?
[07:39] <cprov> Kinnison: DF will be busy on dapper building, so the 3 builders we have 
[07:39] <Kinnison> that way I'll have it marked 'new' in my inbox
[07:39] <kiko> nope
[07:39] <kiko> it's your task
[07:39] <kiko> good luck!
[07:39] <kiko> you won't have anything else to worry about for the next set of days
[07:39] <kiko> so save your inbox as inbox-pending-soyuz
[07:39] <kiko> and then the emptiness in it will fill your heart with the soyuz release vibe
[07:40] <Kinnison> I'll get right on this starting tomorrow first thing
[07:40] <mdke>  if possible, i'd like to add the ubuntu-doc mailing list as default QA for ubuntu-docs bugs, but I'm not sure if it will be possible, will that work?
[07:42] <Kinnison> ciau all
[07:42] <kiko> thanks Kinnison 
[07:42] <kiko> mdke, I can help you, but let me talk to cprov first, bbias
[07:43] <bradb> Given an IBinaryPackageName and an IDistribution, what's the easiest way to get the IDistributionSourcePackage? (Assuming a binary package's name is the same across all arch's, which is reasonable, AIUI.)
[07:43] <mdke> kiko, sure, thanks. Doesn't need to be done immediately
[07:44] <kiko> bradb, it is not simple, though you can get the latest information.
[07:44] <bradb> Is there an example somewhere?
[07:44] <kiko> bradb, you need to look at the publishing tables and figure out what was the latest binary package with that name published in the distribution, then look up its build
[07:45] <kiko> binarypackagerelease actually has a distributionsourcepackagerelease method
[07:45] <bradb> Yeah, many steps to getting there though, it seems.
[07:45] <bradb> i.e., even BPR
[07:45] <kiko> so it's a matter of looking for publishing records for that binary package name
[07:46] <bradb> Okay, that makes sense, I'll try it.
[07:47] <kiko> it will help if you understand how secure*publishing work
[07:47] <kiko> they are not difficult -- basically logs of what was published with statuses
[07:47] <\sh> kiko: what was the result of the discussion with nkour and the gajim lp page?
[07:56] <kiko> carlos, how's it going?
[07:56] <kiko> \sh, I chose to ignore the subject completely. :)
[07:56] <kiko> \sh, honestly, here's my assessment of the problem
[07:57] <kiko> a) we don't currently do very well in using the "official/not official" status for upstreams. We're going to start working on that -- BjornT is working on some bugwatch features that may tie in to that.
[07:57] <kiko> b) in the meantime we can use the text in the upstream page to make a clear statement of "not using launchpad for bugs or translations". I can reassign the project to you if you like.
[07:58] <kiko> c) malone (and launchpad in general) need to grow a way to better describe the upstream/package split. so far none of the solutions proposed have held their ground though
[07:58] <kiko> hey carlos?
[07:58] <\sh> kiko: well..I don't mind...if nkour is ok with it, then do it. 
[07:59] <kiko> \sh, what's your launchpad username?
[07:59] <carlos> kiko, hi
[07:59] <kiko> how's it going man?
[07:59] <\sh> kiko: shermann 
[07:59] <carlos> kiko, Fine, thanks
[07:59] <kiko> carlos, wanna chat a bit on the phone?
[07:59] <carlos> kiko, Can we do that later?
[08:00] <carlos> I need to go out in 10 minutes
[08:00] <kiko> carlos, you tell me what time
[08:00] <carlos> will be back in two hours
[08:00] <carlos> 21:00 UTC
[08:00] <kiko> sounds good.
[08:00] <carlos> two hours from now
[08:00] <kiko> you ping me
[08:00] <carlos> kiko, ok
[08:03] <kiko> salgado, how's this MM monster looking?
[08:04] <kidem> how long does it take CD to get to the address?
[08:07] <salgado> kiko, I finished working on that file list thing (the one people has to provide for non-Archive mirrors) and added some missing validators. now I started working on the script to probe the mirrors; there's 3 new questions on the spec that need answering. maybe you know the answer...
[08:08] <salgado> cprov, it'd be good if you could take a look at these questions too (^^)
[08:08] <cprov> salgado: ok
[08:09] <kiko> sounds good.
[08:09] <kiko> matsubara, dude, bug 28768 is a disgrace 
[08:09] <Ubugtu> Malone bug 28768: ""Search for Products" on the "Welcome to Launchpad" page does nothing." Fix req. for: malone (upstream), Severity: Normal, Assigned to: Nobody, Status: Unconfirmed http://launchpad.net/bugs/28768
[08:11] <kiko> pagetest plus ultra
[08:16] <kiko> salgado?
[08:16] <kiko> here's my take on your questions
[08:16] <kiko> I think DM.enabled is what tells us that the mirror is activated.
[08:17] <kiko> I think enabled = False means don't probe until an admin (distro owner) resets the flag
[08:17] <kiko> I am looking for the code for you
[08:18] <matsubara> kiko: Isn't it a launchpad bug? anyway, just confirmed it and assigned to myself. 
[08:18] <kiko> matsubara, it is a launchpad bug -- a hideous one.
[08:18] <kiko> how can anyone work without a @#@!!@# tip of launchpad
[08:18] <kiko> bradb, did stub do a permanent mirror or was it one-shot?
[08:19] <bradb> kiko: one-shot
[08:19] <bradb> You WISH it were so convenient dude!
[08:19] <[1] Ben> hello i just downloaded Ubunutu Flight 3 for x64 and have a problem.
[08:19] <kiko> [1] Ben, I suspect you want the #ubuntu channel
[08:19] <[1] Ben> ill try that thank you
[08:20] <kiko> enjoy
[08:37] <kiko> salgado, elmo is the person who can talk to you about the tool, but I don't think the code is findable, see #canonical
[08:38] <salgado> kiko, yeah, I saw that. :-(
[08:49] <cprov> salgado: spec fixed
[08:52] <lifeless> well, we have  futex lock
[08:52] <lifeless> an actual hang of python
[08:53] <lifeless> jamesh has a cpu bound process running
[08:53] <lifeless> probably the pending branches script
[08:54] <kiko> lifeless, can we set up an rsync mitigation branch somehow?
[08:54] <lifeless> kiko: it wont mitigate it to any significant degree
[08:54] <lifeless> the amount of IO being done is very low
[08:55] <kiko> lifeless, I don't understand. stub made a mirror of a branch and brad was able to merge off it and work
[08:55] <kiko> why wouldn't that be enough for the rest of us?
[08:55] <lifeless> thats because bzr is very good at mesh behaviours
[08:56] <lifeless> wow, this is a strange bt
[08:56] <lifeless> #276 0x0000000000000000 in ?? ()
[08:56] <lifeless> #277 0x0000000000000000 in ?? ()
[08:56] <lifeless> #278 0x0000000000000000 in ?? ()
[08:57] <kiko> lifeless?
[09:07] <cassidy> hi. Is it possible to track a specific product ? Receive a mail notification on new bug, support request, etc  ?
[09:08] <bradb> cassidy: It's possible for bugs, with a bit of coordination.
[09:09] <bradb> cassidy: In its current design, a product has one bug contact.
[09:09] <bradb> cassidy: If that contact were a team to which you belonged, you could get email about the bugs on that product.
[09:10] <bradb> If the team had a mailing list email addressed, you'd have to be subscribed to that ML.
[09:10] <bradb> Does that help?
[09:12] <cassidy> bradb: yes thanks ! Where can i find the bug contact adresse ?
[09:12] <bradb> cassidy: /products/$name/+editbugcontact. Eh, that's kind of a bug I should fix.
[09:13] <cassidy> bradb: the "not allowed" is a bug so i assume ;)
[09:14] <bradb> cassidy: Yeah, the UI shouldn't allow you to click links that bring you to "not allowed" pages.
[09:14] <bradb> cassidy: Which product are you interested in?
[09:14] <cassidy> bradb: xchat-gnome
[09:14] <cassidy> should be on the desktop-team
[09:15] <bradb> cassidy: When you say product, do you mean a package in a distribution?
[09:16] <bradb> Some people mix those terms. The mention of desktop-team made me want to confirm we're talking about the same thing.
[09:16] <cassidy> both. Actually x-g is a specific product if i understand well ?
[09:16] <bradb> It's a product too, yeah.
[09:17] <cassidy> bradb: i'm not yet very familiar with launchpad so yes, there is good chance than i mix those terms ;)
[09:17] <cassidy> should also be a package since it's Dapper, doesn'it ?
[09:17] <bradb> cassidy: You can sub to the package: https://launchpad.net/distros/ubuntu/+source/xchat-gnome/+subscribe
[09:18] <cassidy> bradb: great ! thanks a lot
[09:19] <bradb> cassidy: No prob. As for the upstream, it looks like https://launchpad.net/people/trowbrds is your contact to talk about setting up contacts.
[09:20] <cassidy> bradb: yes, i know him :)
[09:20] <bradb> great. Have I given you enough to point him in the right direction for setting things up upstream?
[09:23] <cassidy> bradb: what things have to be changed ?
[09:24] <kiko> lifeless?
[09:24] <bradb> cassidy: Probably 1. add a team with an ML address, 2. set that team to the bug contact: https://launchpad.net/products/xchat-gnome/+editbugcontact
[09:25] <dilys> Merge to devel/launchpad: [trivial]  use re.escape to prevent process-email from crashing. (r3005)
[09:25] <lifeless> kiko: ?
[09:27] <cassidy> bradb: no problem. i'll tell to him. But why is it needed if we can add yourself on https://launchpad.net/distros/ubuntu/+source/xchat-gnome/+subscribe ?
[09:27] <kiko> lifeless, I'm still waiting for a mitigation strategy for us having no rocketfuel tip.
[09:27] <lifeless> kiko: I dont have one, I'm getting you one as fast as I can
[09:28] <kiko> lifeless, what's wrong with rsync, for the Nth time?
[09:29] <bradb> cassidy: That may be a design flaw. This feature is fairly new.
[09:30] <bradb> cassidy: One line of reasoning might say that a product would tend to map to one ML, that people could subscribe to, whereas a package in a distribution wouldn't, so individual package subscriptions help.
[09:31] <lifeless> kiko: I can rsync a bit image over, but that wont make the merge into the archive significantly faster AFAICT and will introduce race conditions as the locks are not mutually exclusive at the moment
[09:31] <kiko> lifeless, but at least us, developers, can do a merge from it?
[09:32] <lifeless> kiko: isn't there a copy stub pulled across ?
[09:32] <kiko> lifeless, it's old already.
[09:32] <kiko> it's not being updated
[09:32] <kiko> and THAT would be a mitigation strategy if kept up-to-date
[09:32] <lifeless> erm, pqm is blocked until the push completes
[09:32] <cassidy> bradb: in fact we use gnome bugzilla in upstream, so i think we just care about package related bugs
[09:32] <lifeless> that r3005 above has *not* hit the rocketfuel master copy.
[09:33] <bradb> cassidy: Ah, yeah, sounds correct. :)
[09:33] <lifeless> I'll rsync across a spare copy for you now, *that* I can do.
[09:33] <kiko> lifeless, and cron it?
[09:34] <lifeless> sure, but if you get corrupt branches merging from it, dont blame me
[09:34] <kiko> if you cron over an rsync of a spare copy -- that's a mitigation strategy
[09:34] <cassidy> bradb: i just would like receive mail notification when users use Malone to report bug on x-g (so it will be on the package i suppose)
[09:34] <lifeless> nor ask me to support them.
[09:34] <kiko> lifeless, make it rsync at a known time and mail launchpad-list
[09:34] <bradb> cassidy: Yep, subscribing to the package, not the product, is the right option for you.
[09:34] <kiko> like once every 2 hours, at the hour and 15 minutes, and we can avoid syncing at that time
[09:35] <cassidy> bradb: ok. i'm beginning to understand how it works :)
[09:35] <kiko> it sucks, but it sucks less than the competition
[09:35] <bradb> cassidy: cool
[09:35] <cassidy> bradb:  https://launchpad.net/distros/ubuntu/+source/xchat-gnome/+subscribe is about package so everything should be fine now ?
[09:36] <bradb> cassidy: Yep.
[09:37] <cassidy> bradb: great. isn't it possible to subscribe on all package stuff: new version published, support request, ... ?
[09:38] <bradb> cassidy: Not yet.
[09:39] <bradb> cassidy: We had a specification going on that: https://wiki.launchpad.canonical.com/PackageSubscriptions but only the bug bits have been implemented so far.
[09:39] <lifeless> kiko: chinstrap:/home/warthogs/archives/temporary-rocketfuel
[09:39] <lifeless> kiko: its just the lp branch that is horked.
[09:39] <kiko> lifeless, well, that's a relief. 
[09:39] <cassidy> ok ok. Thanks for for your help bradb
[09:39] <kiko> lifeless, can you mail the launchpad list with the sync times as I requested?
[09:39] <bradb> cassidy: no prob
[09:39] <lifeless> kiko: patience dude I'm typing it up already
[09:40] <kiko> rock and roll suicide
[09:44] <mdke> another css problem for you at bug #28824
[09:44] <Ubugtu> Malone bug 28824: "css broken on the people page" Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: Unconfirmed http://launchpad.net/bugs/28824
[09:51] <Kamion> Can anyone help me (a) update the svn URL for https://launchpad.net/products/net-retriever/+series/main (and two other related products), (b) ensure that the resulting baz branch continues smoothly on (mergeably) despite the changed svn URL?
[09:51] <Kamion> (I can't do (a) because I don't have permissions; and (b) might Just Work but I don't know.)
[09:53] <kiko> ddaa?
[09:53] <kiko> Kamion, I think you had better write to launchpad -- david needs to tell me if it's safe to do and if anything else needs to be done.
[09:53] <ddaa> yup?
[09:53] <kiko> ah, the man himself
[09:53] <kiko> see Kamion's query
[09:53] <ddaa> This stuff is usually safe to do.
[09:54] <ddaa> The product owner should have the privs to do that IIRC.
[09:54] <Kamion> I don't know who the product owner is; as far as I know it's not me, and it's not displayed by Launchpad on the URL above.
[09:55] <kiko> well
[09:55] <ddaa> It's "Registry administrators" i.e. nobody in particular
[09:55] <kiko> is that product yours?
[09:55] <ddaa> it's displayed on the product page
[09:55] <Kamion> I effectively maintain the package in Ubuntu, though, and am part of the upstream maintenance team
[09:55] <ddaa> Registry Administrators are actively looked for sane volunteers to adopt their products.
[09:55] <Kamion> Can I mail launchpad with a big list of products I volunteer for, then?
[09:56] <ddaa> Kamion: this is net-retriever. Net-retriever, this is Kamion.
[09:56] <Kamion> (ideally, we'd make /people/debian-boot a team, and I'd join it ...)
[09:56] <ddaa> Kamion: Absolutely, or you can ask any admin (about half the launchpad team) to give them to you on a per-need basis.
[09:56] <Kamion> thanks. In that case, if you could do the same for cdrom-retriever and floppy-retriever (per-need ...), I'd appreciate it
[09:57] <ddaa> kiko seems idle, I guess he would be happy to do that :)
[09:57] <ddaa> I'm running a race against the clock right now, so kiko, if you can do it, that would be really nice.
[09:57] <ddaa> otherwise I'd be happy to do it.
[09:57] <kiko> okay, gimme a few minutes
[09:59] <Kamion> Odd. https://launchpad.net/products/net-retriever/+series/main/+edit still gives me Forbidden; does it take a while to propagate somewhere?
[10:00] <ddaa> Mh...
[10:00] <ddaa> Try again?
[10:00] <Kamion> still
[10:00] <kiko> dude
[10:00] <kiko> not even I can access that page
[10:01] <ddaa> Mh
[10:01] <kiko> I think that page is buttsource only
[10:01] <ddaa> okay
[10:01] <ddaa> that seems weird
[10:01] <ddaa> Kamion: just tell me the URLs then
[10:01] <kiko> ddaa, a bug?
[10:02] <ddaa> Probably, to some extent
[10:02] <ddaa> Kamion: what you want to use is https://launchpad.net/products/net-retriever/+series/main/+source
[10:02] <Kamion> yeah, also forbidden
[10:03] <Kamion> ddaa: svn://svn.debian.org/d-i/trunk/packages/net-retriever
[10:03] <Kamion> ddaa: svn://svn.debian.org/d-i/trunk/packages/cdrom-retriever
[10:03] <Kamion> ddaa: svn://svn.debian.org/d-i/trunk/packages/floppy-retriever
[10:03] <ddaa> I guess that is related to some of my past requests.
[10:03] <ddaa> I wanted repo changes to be workflowed, to get through buttsource approval.
[10:04] <ddaa> also, I would like people not to rename series until importd is fixed to handle that...
[10:04] <ddaa> Long broken stuff that nobody ever came round to fixing... :(
[10:05] <ddaa> HOLY COW!
[10:05] <ddaa> not even I can modify it!
[10:05] <ddaa> kiko: can you file a bug about that, please
[10:06] <ddaa> I'll use database superpowers in the meantime
[10:06] <ddaa> "This source has been certified and is now unmodifiable"
[10:06] <ddaa> it says
[10:06] <carlos> kiko, ping
[10:06] <ddaa> idiot
[10:06] <carlos> kiko, I'm back
[10:06] <ddaa> I'm BUTTSOURCE
[10:06] <ddaa> I can modify any rcs detail I want, if I don't, who would????
[10:07] <kiko> carlos, I am so happy to hear that
[10:07] <kiko> ddaa, what bug do you want me to file? It may be more expedient for you to file it, I have a date with carlos right now..
[10:07] <ddaa> *grumble* *grumble*
[10:07] <carlos> kiko, you should call my land line, my mobile phone is still off
[10:07] <ddaa> okay, I will do the filing
[10:08] <ddaa> Kamion: I'm not forgetting you
[10:10] <Kamion> ddaa: heh, thanks
[10:11] <kiko> salgado, can you update our syncer to use temporary-rocketfuel?
[10:11] <Kamion> "This source has been certified to be unmodifiable for all time even by upstream" ;-)
[10:11] <ddaa> I made a request for workflowing...
[10:12] <ddaa> so people could not get approval with somehing sane, and then make it wrong, which is a real problem with Arch...
[10:12] <ddaa> With bzr it will not matter anymore.
[10:12] <buxy> and I configured it several days ago, what goes wrong ? what can I do ?
[10:12] <salgado> kiko, where's that tree?
[10:12] <ddaa> Well, it will still matter, but nothing that cannot be fixed.
[10:12] <Kamion> so a change in RCS details for Arch should make the source no-longer-certified?
[10:13] <Kamion> (sounds ok ...)
[10:13] <salgado> kiko, nm, found it
[10:13] <Kamion> until somebody comes along and rechecks it anyway
[10:13] <ddaa> Kamion: yes, that was the idea... But then I remember that importd is to broken to figure that something once certified is not anymore unless manually reloaded...
[10:13] <ddaa> okay... now it all makes sense...
[10:14] <ddaa> all because of Arch namespace...
[10:14] <Kamion> I think I'd better merge that package from Debian by hand in the meantime; I have a deadline for it
[10:14] <Kamion> thanks for the help, hopefully it'll get sorted eventually :)
[10:14] <ddaa> I'll babysit your imports
[10:14] <ddaa> It will be done within the hour.
[10:18] <Kamion> ddaa: oh, cool, thanks
[10:23] <ddaa> okay, the imports are running
[10:25] <lucas> hi
[10:27] <lucas> the username I usually use is 'lucas'. However, it's used on launchpad by somebody who was last active two months ago
[10:28] <lucas> I'm now an ubuntu-dev member, and find it increasingly inconvenient to have to use 'nussbaum' as login (my last name), since nobody know me under this name
[10:28] <lucas> is there something done about purging old LP accounts ?
[10:29] <Kamion> two months sounds rather short for any kind of purging, to me
[10:30] <lucas> what about forced renaming ? :-)
[10:30] <ddaa> lucas: if you can convince an admin to do it, it's possible
[10:30] <ddaa> it's not a decision I would take though
[10:31] <lucas> is there a predefined delay about such stuff ? like 3 months, or 6 months ?
[10:31] <ddaa> It has never come up before AFAIK
[10:31] <ddaa> so it do not think it has been discussed already
[10:32] <ddaa> And I just do not have the authority to change this sort of stuff.
[10:32] <lucas> ok
[10:32] <lucas> where should it be discussed ?
[10:32] <ddaa> I think the new launchpad-users mailing list would be the right place
[10:32] <lucas> ok
[10:35] <ddaa> Kamion: your archives are updated
[10:36] <ddaa> at your service if you have any other import-related trouble
[10:36] <ddaa> until I have some time to _think_ about what would be the right thing to do, just get RCS changes through to me.
[10:36] <ddaa> I will handle these requests at top priority
[10:41] <Kamion> ddaa: thanks!
[10:57] <sd-tux> carlos: hi, it's still about rosetta .po import :) my(Alexander) translations are still in Published=False state :( i want to upload more translations this days... can you try to find a solution plz 
[10:58] <carlos> sd-tux, the Published=False is not a problem (I think I will remove that column to prevent this kind of confussions...)
[10:58] <carlos> sd-tux, anyway, I will take a look to the queue now
[10:58] <carlos> jordi, hi, around?
[10:59] <Kamion> ddaa: thanks, updated net-retriever archive confirmed sensible and merged
[11:03] <sd-tux> carlos: https://launchpad.net/distros/ubuntu/breezy/+lang/ka ... evolution-2.4 is not updated... did you "publish" my evolution-2.4 translation ?
[11:03] <carlos> sd-tux, if it's imported, you should get an email
[11:04] <sd-tux> carlos: right, i didn't get any mails from rosetta ..
[11:04] <carlos> sd-tux, then it's not yet imported
[11:08] <sd-tux> carlos: hmmm.. what can i do now  :) ?
[11:09] <carlos> sd-tux, I'm handling your requests, they should be imported tonight
[11:09] <kiko> rock and roll carlos 
[11:09] <kiko> carlos, send in your activity reports dude
[11:10] <sd-tux> carlos: OK, thank you...
[11:11] <carlos> kiko, I promise you I will do it between tonight and tomorrow
[11:11] <kiko> cool
[11:11] <kiko> just don't forget
[11:12] <carlos> kiko, Don't worry, I will try to come back to the daily reports
[11:20] <lifeless> kiko: FYI there will be a delay getting built back to order
[11:20] <lifeless> kiko: there is a unicode bug I need to address
[11:25] <kiko> okay
[11:25] <kiko> that's fine
[11:25] <carlos> sd-tux, I think all your files are now ready to be imported, you should get emails soon
[11:36] <carlos> see you