/srv/irclogs.ubuntu.com/2008/06/26/#launchpad-meeting.txt

Tanyahalo01:37
=== salgado is now known as salgado-brb
=== salgado-brb is now known as salgado-afk
=== salgado-afk is now known as salgado
=== salgado is now known as salgado-lunch
=== cprov is now known as cprov-afk
=== mrevell is now known as mrevell-bbl
=== salgado-lunch is now known as salgado
Rinchenhello MootBot18:58
=== mrevell-bbl is now known as mrevell
Rinchennice of you to join us18:58
thumperRinchen: now what would be good was if MootBot could reply nicely18:59
=== Rinchen changed the topic of #launchpad-meeting to: Launchpad Meeting Grounds | Channel logs: http://irclogs.ubuntu.com/ | Meeting Logs: http://www.novarata.net/mootbot/meetinglogs/
Rinchenmtg logs have changed18:59
* Rinchen sighs.18:59
Rinchen#startmeeting19:00
RinchenWelcome to this week's Launchpad development meeting. For the next 45 minutes or so, we'll be coordinating Launchpad development.19:00
MootBotMeeting started at 13:00. The chair is Rinchen.19:00
MootBotCommands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]19:00
Rinchen[TOPIC] Roll Call19:00
MootBotNew Topic:  Roll Call19:00
mrevellme19:00
rockstarme19:00
Rinchenme!19:00
leonardrme19:00
al-maisanme19:00
marsme19:00
adeuringme19:00
schwukme19:00
bacme19:00
sinzuime19:00
barryme19:00
cprov-afkme19:00
flacosteme19:00
BjornTme19:00
mptme19:00
bigjoolsme19:00
statikme19:00
Rinchenallenap and gmb send their apologies today19:00
intellectronicame19:00
Rinchenkiko-fud, SteveA_19:00
kiko-fudme19:00
=== kiko-fud is now known as kiko
kikoor course!19:01
Rinchenmatsubara, ?19:01
matsubarame19:01
abentleyme19:01
salgadome19:01
Rinchenreleases team is here19:01
flacostefoundations is here19:01
Rinchenbugs appears to be here19:01
Rinchenthumper, ?19:01
Rinchenabentley, ?19:02
kikotim's here, yeah19:02
rockstarabentley me'd19:02
Rinchenah, good thanks19:02
RinchenEdwinGrubbs, ?19:02
abentleyYeah, me19:02
EdwinGrubbsme19:02
Rinchen[TOPIC] Agenda19:02
MootBotNew Topic:  Agenda19:02
Rinchen * Next meeting19:02
Rinchen * Actions from last meeting19:02
Rinchen * Oops report (Matsubara)19:02
Rinchen * Critical Bugs (Rinchen)19:02
Rinchen * Bug tags19:02
Rinchen * Operations report (mthaddon/herb)19:02
Rinchen * DBA report (stub)19:02
Rinchen * Sysadmin requests (Rinchen)19:02
Rinchen * New packages required (salgado)19:02
Rinchen * A top user-affecting issue (mrevell)19:02
mthaddonme19:02
Rinchen * Doc Team report (mrevell)19:02
herbme19:02
Rinchen[TOPIC] Next meeting19:02
MootBotNew Topic:  Next meeting19:02
kikoI won't be here!19:03
Rinchensame time same place kiko?19:03
Rinchenok, so it's me and the rest of the team minus you and gmb19:03
Rinchen[AGREED] Next meeting same time, place on 3 July - kiko and gmb send apologies19:04
MootBotAGREED received:  Next meeting same time, place on 3 July - kiko and gmb send apologies19:04
Rinchen[TOPIC] Actions from last meeting19:04
MootBotNew Topic:  Actions from last meeting19:04
Rinchen[AGREED] None19:04
MootBotAGREED received:  None19:04
Rinchen[TOPIC] Oops report (Matsubara)19:04
MootBotNew Topic:  Oops report (Matsubara)19:04
matsubaraSorry, haven't prepared an oops report. I'll follow up with individuals.19:05
Rinchenboo hiss19:05
kiko!19:05
kikowell19:05
kikoI can make one up on the fly19:05
kikoI have one for bac and EdwinGrubbs: OOPS-907S11419:05
ubottuhttps://devpad.canonical.com/~jamesh/oops.cgi/907S11419:05
kikoI believe it's fixed, but is it really?19:05
=== thumper_laptop is now known as thumper
thumperme19:06
kikoI have a fix for the RDF timeouts in my tree, will try and finish tonight19:06
matsubarakiko: I think that one is the one bac fixed as one of his BAD items on the test plan19:06
kikovery good19:06
kikothumper,         6 /    0  BranchMergeProposal:+request-review19:06
kikothumper, six timeouts on staging is really high19:06
backiko: actually that OOPS occurred during integration testing with the salesforce proxy -- an error on the far side.19:06
kikobac, will it happen in production, though, and should it?19:07
matsubarabac, so that's not bug 242422?19:07
kikothumper, can you look into that and see if it can be improved?19:07
ubottumatsubara: Bug 242422 on http://launchpad.net/bugs/242422 is private19:07
thumperkiko: that was me seeing if the storm branch with __contains__ was there19:07
backiko:  it has been fixed in my integration work with surfous19:07
kikothumper, oh, okay -- it works today?19:07
kikobac, okay.19:07
bacmatsubara: no, not part of bug 24242219:07
thumperkiko: I haven't tested today, only been awake for a few minutes19:07
ubottubac: Bug 242422 on http://launchpad.net/bugs/242422 is private19:07
kikothumper, okay, if you could test I'm your fan19:07
kikowhen you have a sec19:08
kikoBjornT, does your branch with +packages improvements live anywhere where somebody else could pick it up?19:08
thumperkiko: yes, works today19:08
Rinchenkiko just gave us a new t-shirt design "I'm your fan"19:08
kikoawesome19:08
BjornTkiko: well, it's landed19:08
Rinchenanything else kiko on oopses?19:08
kikowhen appropriate!19:08
kikoBjornT, OOPS-907S177 :-/19:08
ubottuhttps://devpad.canonical.com/~jamesh/oops.cgi/907S17719:08
kikoBjornT, I thought there were additional fixes though19:08
BjornTkiko: someone could certainly pick it up. there is still a soyuz task open for the bug19:09
kikoBjornT, do you have any work done on it that you could share, or is that it?19:09
danilos(just in, jtv's network is down, but he smsed me to say that log(0) fix is awaiting review)19:09
kikobarry, will you have a fix for OOPS-907S44 by monday?19:09
ubottuhttps://devpad.canonical.com/~jamesh/oops.cgi/907S4419:09
kikothanks danilos19:09
kikodanilos, log(0) fix? :)19:09
barrykiko: i'm working on it!19:10
kikobarry, that's not what I asked :)19:10
barrykiko: i know :)19:10
kikoheh19:10
kikoadeuring, do you have time to look at this many queries in OOPS-907EA192 -- though I believe I know what that one is, hard to fix19:10
daniloskiko: that's what jtv said "oopses about log(0)" :)19:10
Rinchenlog(0) sounds like a productivity modifier :-)19:10
BjornTkiko: no, i didn't have time to get rid of the soyuz queries. the work in the view class is pretty much there. what is missing is methods for getting the information from the db.19:10
kikowow19:10
adeuringkiko: yes19:10
Rinchenjtv just sms'd me with the same report19:11
kikoBjornT, thanks. bigjools, if you have a second left over tomorrow or monday, it's similar work to your +queue improvements and is a top-query page for us today.19:11
SteveA_hi19:11
kikothumper, are there fixes already in place for OOPS-907EC70 and OOPS-907EB53 -- that's on edge which is kinda old.19:11
kikothat's it for me19:11
thumperlemmie check19:12
Rinchen[TOPIC] Critical Bugs (Rinchen)19:12
MootBotNew Topic:  Critical Bugs (Rinchen)19:12
RinchenYay. We have none pending!19:12
Rinchensecond week in a row iirc19:12
kikowell, wait till monday! :)19:12
RinchenGood job folks19:12
thumperkiko: yes, both landed19:12
kikothumper, you the man, **2. :)19:13
Rinchen[TOPIC] Bug tags19:13
MootBotNew Topic:  Bug tags19:13
Rinchenwe have one19:13
kikolet me guess19:13
Rinchenhttps://help.launchpad.net/TaggingLaunchpadBugs19:13
kikopackage diffs! :)19:13
Rinchencodereview by matsubara19:13
cprov-afkkiko: again ?19:13
kikoheh19:14
thumper+1 to codereview19:14
kiko+1 to codereview19:14
thumperwe are going to get a number of bugs :-)19:14
kikothough19:14
mpta number? like 2?19:14
kikois that confusable with stuff in picked up during our internal code review process?19:14
thumpermpt: have more than that already19:14
mptok :-)19:14
RinchenAs we expand the capabilities of LP Code Review, it might be worthwhile now to make this category slightly more generic.19:15
Rinchenand also to prevent any confusion with infrastructure code review19:15
mptkiko, that problem applies to many things about Launchpad already (cf. the "Launchpad itself" hack)19:15
intellectronicawhy not merge-proposal. is that not the name we actually use in launchpad?19:15
thumperintellectronica: because we'd rather think of the thing as a code review of a proposed merge :)19:15
matsubaraI think codereview is slightly more generic than merge-proposal19:16
kikoit's fine by me, either way -- didn't want to polemize!19:16
Rinchento be consistent with the other tags, it should be "code-review"19:16
statiki always learn new words in this meeting19:16
matsubaraI'm not fussed about the tag name too, I just want to group those bug reports and codereview was the first thing that came to my mind19:16
RinchenI can't think of a more general term for branch actions of this sort so I'm +1 on "code-review"19:17
Rinchenmatsubara, thumper can you live with the hyphen?19:17
matsubaradanilos: you don't have any comment?19:18
thumperhyphen is good19:18
matsubaraRinchen: yes19:18
Rinchenany opposition?19:18
kikogo go go19:18
Rinchen[AGREED] "code-review" tag approved. matsubara to update page and tag19:18
MootBotAGREED received:  "code-review" tag approved. matsubara to update page and tag19:18
Rinchen[TOPIC] Operations report (mthaddon/herb)19:18
MootBotNew Topic:  Operations report (mthaddon/herb)19:18
herbThursday (2008-06-19): The last of the Launchpad machines were updated to hardy.19:19
herbThursday (2008-06-19): Codebrowse was restarted a handful of times. It was suspected this was due to people browsing MySQL branches. Based on some of the emails we've seen today it seems that this is the case.19:19
herbWednesday (2008-06-25): Codebrowse was moved to it's new server.19:19
herbEdge has been running with storm since Tuesday (2008-06-24) and seems to be doing well.19:19
herbWe're release-critical as of the wee hours this morning, and we're scheduled for the new release late Monday / early Tuesday.19:19
herbThere are 2 cherry picks waiting for approval (r6410 and r6451) and 1 that's been approved (r6530). We don't know yet when r6530 will be rolled.19:19
herbThat's it from Tom and me unless there are questions.19:19
mthaddoner, unfortunately codebrowse had to be reverted from it's new server - mwhudson is working on it19:19
kikoherb, does it really need to be rolled out?19:19
flacosteherb: is one of those for the xmlrpc private server?19:20
kikomthaddon, thumper: will the new codebrowse code rock the boat?19:20
herbkiko: we're trying to determine if it is necessary.19:20
barryflacoste: 653019:20
mthaddonkiko, in what way?19:20
thumperkiko: what boat?19:20
kikomthaddon, thumper: in a good way.19:20
kikoflacoste, statik: I'd like to know if that's really necessary or not, given the monday rollout19:21
mthaddonkiko, if it worked, yes - like I said, we had to revert, but mwhudson is working on it19:21
kikomthaddon, oh, that was with the new code already?19:21
mthaddonkiko, yes19:21
statikkiko: i'm ok with waiting for monday19:21
flacostestatik, SteveA_: can the mailing list deployment wait until the roll-out?19:21
beunoFWIW, it seemed to be a problem with accesing branches over HTTP instead of locally, because it got moved to a seperate machine  :)19:21
beuno(Loggerhead that is)19:22
beunoand hi  :)19:22
Rinchencaching ftw ;-)19:22
mthaddonbeuno, right, that was the problem (although that's what we were trying to do with it, so it's a problem that it's a problem)19:22
RinchenAnything else for herb or mthaddon ?19:22
mthaddonRinchen, er, no19:22
SteveA_flacoste: monday?19:22
kikofuntastic19:23
kikomthaddon, flacoste: will you guys coordinate on the /@@ fix?19:23
kikoI am really excited about getting that rolled19:23
flacostekiko, mthaddon: yes19:23
kikoI think it will make a huge difference19:23
mthaddonsounds good on /@@/19:23
kikomthaddon, note that the /@@ stuff also needs the cache-control jimmied in19:23
flacosteSteveA_: yeah, kikos wants to hold off the cherry pick and wait for the regular roll-out19:23
SteveA_if it's just monday, then okay19:24
mptkiko, if you start caching /@@/ at all, you could get six times as much benefit by fetching all those resources from launchpad.net rather than from the same hostname as the rest of the page (e.g. code.launchpad.net)19:24
mthaddonSteveA_, monday evening my time :)19:24
SteveA_fine19:24
kikompt, that requires changing paths everywhere in our templates though.19:25
mpthmm19:25
flacostekiko: we could do it at the apache level19:25
mptmaybe it could redirect?19:25
flacosteyep19:25
flacoste30119:25
kikompt, flacoste: wouldn't that be slow as slow?19:25
mptbut then maybe a redirect is slower19:26
mptyeah19:26
* mpt shrugs19:26
flacostenot really19:26
kikofor me, a redirect is very slow19:26
flacosteonly the first time19:26
flacostewell19:26
statikfetching static resources from a separate server is faster because it bypasses the browser throttling of how many simultaneous connections to a server are made19:26
kikonot for the first time -- every time code.l.n/@@/foo is hit, we'd be 301ing him back to l.n/@@/foo19:26
statiki think redirecting would kill any performance gains19:26
flacostebrowser should cache the redirect19:26
kikoI dunno if any browser does that, and not persistently19:27
flacosteanyway, fixing the path in all of the templates is outside the current scope19:27
flacosteso, it's redirect-away or forget this19:27
mthaddonflacoste, how much work would that really be?19:27
flacostemthaddon: fixing all the templates?19:27
mthaddonyeah19:27
flacostewell19:28
flacostemaybe not that much19:28
statiki'll volunteer to help mass-editing some templates to get a nice speedup for our users19:28
kikoit's annoying though, putting a tal: expression for each /@@ in there.19:28
kikonot sure.19:28
mptyeah, that would annoy me19:28
kikoflacoste, statik: we could do that post-1.2.6 anyway.19:28
statiktotally19:28
kikothe most important thing is serving that stuff statically19:29
kikothe rest is icing.19:29
mptPerhaps make it part of the template interpretation layer19:29
flacosteok19:29
flacostempt: dream on :-)19:29
* mpt handwaves19:29
flacostetemplate interpretation layer = TAL engine = not-customizable19:29
flacosteor rather !extensible19:30
RinchenI think we're ready to move on19:30
mptnot in the TAL engine19:30
mptmaybe just before that19:30
mptanyway19:30
mptsorry for the digression19:30
RinchenI think that was...thank you herb and mthaddon19:30
Rinchen[TOPIC] DBA report (stub)19:30
MootBotNew Topic:  DBA report (stub)19:30
Rinchenstub is not here19:30
Rinchenso19:30
Rinchen[AGREED] Rinchen to email stub for dba report19:30
MootBotAGREED received:  Rinchen to email stub for dba report19:30
Rinchen[TOPIC] Sysadmin requests (Rinchen)19:30
MootBotNew Topic:  Sysadmin requests (Rinchen)19:30
RinchenIs anyone blocked on an RT or have any that are becoming urgent? I know about Thumper's19:31
intellectronicayeah, we need api.staging.l.n set up19:31
mthaddonintellectronica, in process19:31
Rinchenmatsubara, your shot factory request has been denied.  Speak to me after the meeting for alternatives.19:31
bigjoolsRinchen: the one I talked about in the week with you19:31
intellectronicamthaddon: fantastic19:31
matsubaraRinchen: ok.19:31
mthaddonintellectronica, btw, it'd be nice to have a bit more lead time on RTs like that if possible19:31
cprov-afkbigjools: https://rt.admin.canonical.com/Ticket/Display.html?id=3101619:32
SteveA_we should have a checklist for this stuff19:32
SteveA_ - changes to DNS19:32
intellectronicamthaddon: i realise that19:32
SteveA_ - access to new machines19:32
SteveA_ - extra memory / disk resources19:32
thumper - email processing changes19:32
SteveA_etc.19:32
Rinchenbigjools, cprov-afk - need a due date for that if you could please19:33
SteveA_and go through that when planning changes19:33
SteveA_to give IS advance notice19:33
=== cprov-afk is now known as cprov
bigjoolsRinchen: it's for 1.2.619:33
SteveA_and note that many DNS changes need to apply to both lpnet and edge and staging etc.19:33
statiki just want to say that our IS and OSA team rocks19:33
statiki don't care who knows it19:34
intellectronicathey totally do!19:34
mthaddontoo kind, too kind19:35
Rinchenbigjools, I've updated the ticket and pinged elmo19:35
bigjoolsthanks Rinchen19:35
RinchenSteveA_, actually I think those sections are in the roll-out template19:36
=== SteveA_ is now known as SteveA
Rinchenbut not to that level of detail19:36
Rinchen[TOPIC] New packages required (salgado)19:36
MootBotNew Topic:  New packages required (salgado)19:36
kikoRinchen, would be useful to update that, though the problem I see is that the roll out of APIs spanned more than 4 months, so..19:36
salgadoany new packages this week?19:36
kikoI have a request.19:36
kikosalgado, can we use P3As for lp-meta-deps?19:36
kikopretty please19:37
thumpersalgado: I think mwh said there are new deps for loggerhead19:37
thumpersalgado: simpletal and paste19:37
bigjoolsPimp Your P3A19:37
salgadokiko, can we sign packages there?19:37
Rinchenkiko, SteveA - I'll update the roll-out template with that info19:37
salgadooh, Private PPAs19:37
kikosalgado, not yet. but you can rsync and sign afterwards! :)19:37
cprovsalgado: you can, we can't sign repositories.19:37
cprovsalgado: mirror & signing is trivial, though19:38
salgadowe haven't done so because people have complained about that19:38
kikosalgado, help dogfood p3a!19:38
kikowe'll let you use them free of charge, I checked with mark19:38
kikospecial offer19:38
salgadothekorn, I need a bug report19:38
bigjoolscheap at half the price19:38
salgadothumper, ^19:38
thumpersalgado: I'll pass that on to mwh19:39
thumpersalgado: filed against what?19:39
salgadothumper, meta-lp-deps19:39
thumperk19:39
Rinchensalgado, I wanted to thank you for updating these packages.19:39
SteveAthanks Rinchen19:39
salgadoRinchen, you're welcome19:39
salgadokiko, I can give it a try.  where should I look for docs about it?19:39
statikooh, i want to mirror and sign my own PPA19:40
kikosalgado, have a quick call with julian, it's all pretty easy19:40
kikothanks19:41
cprovstatik: someone posted a nice script sometime ago in lp-user ML (debmirror-based)19:41
statikcprov: thanks, will look for it19:41
Rinchenanything further for salgado?19:41
Rinchen[TOPIC] A top user-affecting issue (mrevell)19:42
MootBotNew Topic:  A top user-affecting issue (mrevell)19:42
mrevellHi19:42
mrevellAgain the theme this week has been people who have subscribed to all Ubuntu bugs, only to find they don't know how to unsubscribe. We also get quite a few requests for help with Ubuntu to feedback|help@launchpad.net19:42
mrevellI've been wondering if an auto-reply would enable people to help themselves, before I'm able to reply to them.19:42
mrevellI've popped a suggested auto-reply email on the team list and would appreciate your comments.19:42
mrevellOther than, nothing outstanding as a user-affecting issue this week. Thanks Rinchen.19:42
Rinchenthanks mrevell19:43
Rinchenanything for mrevell ?19:43
intellectronicamrevell: i think that these are two separate issues19:43
intellectronicaif users subscribe to ubuntu by mistake, it may be that the ui is not clear enough19:43
mrevellintellectronica: Ubuntu help requests and bug unsubscrive requests are, yes.19:43
mptThere are (multiple?) bug reports about this19:44
SteveAI believe we have plans for a UI to show you all subscriptions to mail, in one straightfoward page19:44
intellectronicaindeed, and i even think we have a pretty good idea of how it should be solved19:44
SteveAwhether these are mailing lists or subscriptions that send mail19:44
SteveAand this page will be included in the footer of all emails19:44
Rinchenthe issue is actually a bit compounded by gmail which treats the "you are receiving this bug report because" message as a quote and collapses it so you don't see it unless you know enough to expand the section19:44
SteveAwell, a link to hte page19:44
cprovstatik: I lied, the script is in our 'bug #1' -> https://bugs.edge.launchpad.net/soyuz/+bug/12510319:44
ubottuLaunchpad bug 125103 in soyuz "ppa archives are not signed" [High,Confirmed]19:45
ubottuLaunchpad bug 1 in ubuntu "Microsoft has a majority market share" [Critical,Confirmed] https://launchpad.net/bugs/119:45
intellectronicaRinchen: maybe we should change the formatting?19:45
Rinchenintellectronica, it's crossed my mind before. I was waiting to see if there were other reports of this before I suggested it.19:45
Rinchenwe know that many LP users have gmail accounts19:45
Rinchendon't we barry :-)19:46
mrevellSure, we plan to show all subscriptions but right now we don't. In the the time between now and our providing that, lots of people will remain confused as to how to unsubscribe. As soon as we implement that, or de-confuse the UI, I think an autoreply may help people find the answer they need. Lots of support contact addresses do the same sort of thing.19:46
intellectronicaRinchen: let's file a bug, if there isn't one yet, so that we see if there's much activity19:46
barryRinchen: indeed19:46
intellectronicai use gmail, but i always considered the folding of that footer a feature!19:46
mptbug 110953, bug 89938, bug 21733719:46
ubottuLaunchpad bug 110953 in launchpad "Can't easily see everything I'm subscribed to" [Medium,Confirmed] https://launchpad.net/bugs/11095319:46
ubottuLaunchpad bug 89938 in malone "Very difficult to unsubscribe from a package" [Medium,Confirmed] https://launchpad.net/bugs/8993819:46
ubottuLaunchpad bug 217337 in malone "Bug supervisors shouldn't be compulsorily subscribed to all bug reports" [Undecided,New] https://launchpad.net/bugs/21733719:46
Rinchenintellectronica, you are me to file the bug?19:47
Rinchens/are/or19:47
intellectronicaRinchen: i'd love to be you. for now i'll setlle for filing bugs ;)19:47
Rinchenlol19:47
Rinchen[ACTION] Rinchen to file bug on gmail footer auto-collapse as possible reason for inability of users to know how to unsubscribe from bug reports19:48
MootBotACTION received:  Rinchen to file bug on gmail footer auto-collapse as possible reason for inability of users to know how to unsubscribe from bug reports19:48
Rinchen[TOPIC] Doc Team report (mrevell)19:48
MootBotNew Topic:  Doc Team report (mrevell)19:48
* Rinchen nudges mrevell.19:49
mrevellThis week I've continued work on the user guide. Many thanks to jtv for his feedback on the translations section. No news on the doc team, I'm afraid, although I expect they will offer great help in reviewing the user guide ove the coming weeks19:49
mrevellsorry Rinchen, was hunting for the tomboy note19:49
RinchenWe also have had a member of the doc team file a bunch of bugs on the updated help wiki19:49
Rinchenthat was helpful19:49
* sinzui -> afk to get automobile19:50
mrevellYes. And it's worth noting, if you haven't seen already, that help wiki is now running MoinMoin 1.619:50
kikocool19:50
RinchenAnything further for mrevell?19:50
mrevellso if you notice anything odd, please let me or Rinchen know.19:50
Rinchenok, sorry for being a bit over today19:50
Rinchenblame kiko ;-)19:50
RinchenThank you all for attending this week's Launchpad Developer Meeting. See the channel topic for the location of the logs.19:50
Rinchen#endmeeting19:50
MootBotMeeting finished at 13:51.19:50
mrevellthanks rinch19:51
SteveAthanks!19:51
mrevellthanks Rinchen19:51
intellectronicathanks Rinchen19:51
al-maisanthank19:51
rockstarthanks19:51
al-maisans19:51
mptthanks Rinchen19:51
kikowoo, I take the blame19:51
Rinchenyou're the default :-019:57
=== kiko is now known as kiko-afk
=== salgado is now known as salgado-afk
=== kiko-afk is now known as kiko

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!