/srv/irclogs.ubuntu.com/2006/10/03/#launchpad-meeting.txt

=== ddaa [n=ddaa@nor75-18-82-241-238-155.fbx.proxad.net] has joined #launchpad-meeting
=== _thumper_ getting coffee back in 3 min
=== _thumper_ back early
ddaaHello12:00
_thumper_morning12:00
ddaalooks like some of the .au tribe is out12:01
_thumper_yeah12:01
=== mpool [n=mbp@ppp112-44.static.internode.on.net] has joined #launchpad-meeting
_thumper_speak of the devil12:01
mpoolhi12:01
mpoolwelcome!12:01
ddaaokay, only lifeless missing can start12:01
ddaaMEETING STARTS12:01
mpoollifeless is excused, he worked yesterday and said he might miss this12:01
ddaa== Agenda ==12:01
ddaaNext meeting 2006-10-09, 10:00-10:45 UTC.12:02
ddaaSteveA, _thumper_ and ddaa will be on a sprint in London.12:02
ddaa * roll call12:02
ddaa * production status12:02
ddaa * release finder12:02
ddaa * Python import12:02
ddaa * strategic plan12:02
ddaa * bzr-lp features12:02
ddaa * interesting bzr list threads12:02
ddaa * advertising12:02
ddaa * 1.0 targets12:02
ddaa * highlighted bugs12:02
ddaa * any other business12:02
ddaaIf you wish to change the time of the meeting or add/remove agenda items, say "bzzzt!".12:02
ddaaIf we are short on time, the "any other business" item will be automatically, dropped. So if you ''want'' to discuss something more, speak up now.12:02
ddaaI want to go back working on cscvs, so I plan to finish this meeting on time. No digressions today, please.12:02
ddaa== Roll call ==12:02
ddaaNo excuse was given.12:02
ddaampool said: mpool: lifeless is excused, he worked yesterday and said he might miss this12:02
spivI'm here.12:02
mpooljohn's arriving early tomorrow so i'd like to keep this short if possible12:02
ddaaAnyway lifeless' attendance at this meeting is optional12:02
mpoolhere12:02
_thumper_here12:02
jameshhere12:02
SteveAhi12:02
ddaaokay, everybody here12:03
ddaaeverybody happy with agenda12:03
ddaaroll it!12:03
ddaa== Prodution status ==12:03
ddaaNothing new to report on production.12:03
ddaaWell, importd rollout when okay, with only a couple of band-aid fixes12:03
ddaawhich mere merged as trivial end of last week12:04
ddaa== Product release finder ==12:04
ddaa * jamesh: report on PRF progress.12:04
SteveAwhat was the nature of the band-aid fies?12:04
=== lifeless [n=robertc@ppp245-86.static.internode.on.net] has joined #launchpad-meeting
jameshI've got a branch up for review that fixes the remaining problems that need to be fixed before rollout12:05
ddaaSteveA: needed to restore some importd file which is actually imported by buildbot, and needed for what we do12:05
jameshIt is in my "jamesh/launchpad/bug-50569" branch, along with some other ProductSeries cleanups12:05
jameshso hopefully we can turn it on after next rollout12:05
ddaaSteveA: and needed to tweak a method so the new source tree upload/download system does no cause a failure when the botslave data has been cleared out.12:06
ddaajamesh: ack good... I'm itching to ask who's going to be responsible for this code once it is landed, but I think it would be off-topic for the meeting, so I'm not asking12:07
ddaa== Python import ==12:07
ddaa * ddaa: status of Python blocker cscvs bug12:07
ddaa * jamesh: bzr-0.11 fixage for launchpad and cscvs12:07
jameshlifeless was merging this today.12:07
ddaaI'm hacking on cscvs like the end of the world is coming, so I can actually have the new logic for svn changeset working by the end of the week12:08
jameshI am not sure how that is going12:08
lifeless*is* metging this12:08
jameshlifeless: okay.  So barring any test failures we should have new bzr to play with12:09
ddaalifeless: thank you for the love12:09
ddaaaction: jamesh fix unexpected test failures ;)12:09
ddaaof course, there won't be any12:09
ddaa== strategic plan ==12:10
ddaa * mpool owns that agenda item12:10
jameshare you saying they would be expected test failures?12:10
mpoolddaa: thanks, i have comments from everyone, not all are integrated yet12:10
mpoolbut it's basically ticked off, you can remove the item12:10
ddaampool: ack12:10
ddaajamesh: ya trying to cscvsfuse me12:10
ddaa== bzr-lp features ==12:11
ddaa * mpool: report on bzr-lp features12:11
ddaawhat's hot, what's next, what's cold, what's out without love?12:11
mpoolwell let's hear from everyone working on bzr-lp features, *quickly*?12:12
mpooljamesh? spiv?12:12
SteveAI'm interested in the latest on the smartserver and supermirror application of it12:13
SteveAand getting the smartserver running on devpad for launchpad developers to use12:13
ddaaSteveA: later this meeting12:13
spivI'm working on smart server over HTTP -- the basic code in bzr has been written.12:13
SteveAand whether anything's happening about overlay repositories (sometime later)12:13
jameshI merged the .bzr dir stuff into rocketfuel today, so tomorrow "bzr branch https://staging.launchpad.net/products/bzr" should do something sensible12:14
spivI'm now looking at writing a WSGI application for it, so it can actually be deployed.12:14
ddaamoving on12:15
ddaa== Interesting bzr list threads ==12:15
ddaaDo you guys have keywords for outstanding bzr threads from last week?12:15
ddaaI'm hopelessly behind bzr mail again, so pointers would be greatly appreciated12:15
mpoolddaa, remind me why this topic is in the meeting?12:15
SteveAso, david would like to be able to keep up with bzr goings-on12:16
SteveAbut has not the time to keep up with the mailing list12:17
ddaait is a suggestion from SteveA to use you guys collectively as a replacement for the not-yet-here bzr community guy and help folks like me who need to stay up to date with the bzr ML but cannot get around to.12:17
SteveAvarious attendees at this meeting do keep up with the mailing list12:17
SteveAso, the idea is that david can be given pointers to topics that are pertinent to him that arise12:17
mpoolnice idea, but it doesn't seem to catch a lot of content12:17
ddaa*nod*12:17
SteveAdoes anyone do a bzr traffic kind of thing?12:17
mpoolnot at the moment12:17
ddaaNope12:18
SteveAthat was james b's job at one time12:18
mpoolit would be nice, but, again, it takes time12:18
mpoolrather than doing it on irc i think we should either12:18
mpool(A) actually do a traffic summary or12:18
ddaaWe need a trafficker in the bazaar, we!12:18
mpool(B) just forward interesting threads to ddaa or some internal list12:18
mpool(B) is cheaper12:18
SteveAif something is relevant to david, it's probably relevant to launchpad.  so I'd encourage bzr people to point out such threads on the launchpad list12:19
SteveAor even launchpad-users12:19
SteveAand get some cross launchpad-bzr interest going12:19
mpoolOK12:19
mpooli like when you've [steve]  done that on related topics in the past12:19
ddaalaunchpad-users sounds a good idea it first12:19
mpoolok, depending on whether it's a user or developer question12:20
ddaauntil people start complaining, then we can move it to launchpad@12:20
SteveAok, that's an action point for people who read the bzr list12:20
mpoollp-users seems mainly concerned with "send me cdsss!!"12:20
SteveAwho here reads the bzr list?12:20
=== spiv puts his hand up
mpoolme, lifeless, spiv, maybe jamesh?12:20
spivFSVO "read" :)12:20
=== _thumper_ gets it and reads some
SteveAa current target for the launchpad team is to increase useful traffic on launchpad-users12:20
mpoolok, so we all know something about what lp devs need to know 12:20
mpoolif you see something, forward it12:20
SteveAbetter than forwarding12:21
jameshI read it occasionally.  I certainly don't read all messages12:21
mpoolSteveA: or to lp if it's something about internals that might affect lp users12:21
SteveAsend a proper brief mail12:21
SteveAwith a summary of what it is and why it is important to launchpad12:21
SteveAthis doesn't take long12:21
mpoolright, if you do forward at least put on a summary12:21
mpoolOK, great12:21
SteveAand vastly increases the usefulness of the pointer to readers12:21
mpoolresolved12:21
ddaalet's see how it goes next week, moving on12:21
mpoollet's leave this item on the agenda (new title) as a reminder to get in the habit12:21
mpoolmove on12:21
SteveA(particularly to people who subscribe to the launchpad-users list who are not here today)12:22
ddaa== Advertising ==12:22
ddaa * spiv: post draft to LP mailing list for comments.12:22
spivD'oh.12:22
spivI haven't done that.  I just re-read it, though, and I think I'll Just Post It(TM).12:22
spiv(i.e. to my blog)12:22
ddaaas you want, but feel free to ask for review12:23
lifelessI dont read all the messages12:23
lifelessits a busy list12:23
ddaaflascoste gave very useful comments to me in the past12:23
lifelessI certainly read all the threads enough to decide to finish the thread or not12:23
ddaamoving on12:23
ddaa== 1.0 targets ==12:23
ddaasupermirror-smart-server: spiv: still looking on track for october 8th? What did you work on recently, what is still todo?12:23
ddaaimportd-bzr-native: ddaa: progress of Arch removal and dists cleanup. Next: slice out HCT to remove pybaz dependency.12:23
ddaabzr-roundtrip-svn: not for 1.012:23
ddaa * mpool: read up/tick off svn roundtripping discussion12:23
ddaampool: since I have no read the ML recently, I have no idea if anything relevent happened recently12:24
ddaaspiv: in summary, worked on http transport, next wsgi something to integrate into launchpad?12:25
mpoolddaa, sorry, it hasn't got to the top of my list yet12:25
spivddaa: For anonymous smart serving over HTTP, it looks likely I'll have the code ready or close to, but I'm not sure that I'll be able to get it deployed in that timeframe.12:25
spivddaa: Right.12:25
ddaaspiv: talk with SteveA and/or kiko about getting a delay12:25
lifelessddaa: yeah, the http transport needed a large refactor of the code base, more than expected12:26
lifelessbut thats over with12:26
ddaanice to hear12:26
ddaakiko said something about telling him about spec that will need some extra time to deliver12:27
ddaaso you should get that straightened out with him, I think, please keep me in CC12:27
=== ddaa needs to be more precise
spivddaa: Ok.12:28
SteveAand please keep the spec metadata in launchpad up to date12:28
SteveAas that is what we use in conf calls with mark12:28
ddaaOkay, I think I got my fuzzy12:28
ddaamy fuzzy point across12:28
SteveA  importd-bzr-transition, supermirror-smart-server,   bzr-roundtrip-svn12:29
SteveAon https://features.launchpad.net/products/launchpad-bazaar/12:29
SteveAdavid and andrew, please ensure the metadata is up to date12:29
ddaa== Highlighted bugs ==12:29
ddaa * Related to +source12:29
ddaa   * bug 2649: CVS branch details should not be editable or displayed. (https://launchpad.net/products/launchpad-bazaar/+bug/2649)12:29
ddaa   * bug 46240: posting $series/+source yields a confusing warning (https://launchpad.net/products/launchpad-bazaar/+bug/46240)12:29
ddaa   * bug 50569: the product series page does not allow entering source or ftp details for projects without SVN or VCS (https://launchpad.net/products/launchpad-bazaar/+bug/50569)12:29
ddaa * bug: 48813: Efficiently mirroring sftp hosted branches with minimal latency (https://launchpad.net/products/launchpad-bazaar/+bug/48813)12:29
ddaa * bug: 58889: Merged and abandoned branch should not appear in main branch listings (https://launchpad.net/products/launchpad-bazaar/+bug/58889)12:29
ddaaBonus bug: sabdfl asked for the ability to link a spec to a branch.12:29
spivSteveA: Ok.12:29
ddaabug 46240 fixcommitted last week12:29
ddaabug 50569 in-progress by jamesh12:30
ddaaalong with a ton a refactoring and cleanups for +source12:30
jameshbecause of the refactoring, bug 2649 should probably wait for my branch to land12:30
lifelessddaa: cscvs failure12:30
lifelessERROR: testCommitWithAutomaticLogFiller (cscvs.tests.test_bzr.TestBzrWorkingTreeCommit)12:30
lifelessERROR: testCommitWithAutomaticLog (cscvs.tests.test_bzr.TestBzrWorkingTreeCommit)12:31
lifelessjamesh: ^12:31
ddaa*sigh*12:31
ddaacontext: ^^ that is status of bzr-0.11 landing on rocketfuel12:31
lifelessthats with vanilla 0.11, bzrtools removed, lp 0.11 compat branc merged and cscvs compat branch merged12:31
jameshlifeless: weird.  I'll try with 0.11 final12:32
ddaaSteveA: I think the sabdfl bonus bug will be a good think to get _thumper_ started on next week12:32
ddaagets to touch all the webapp stack, and not really complicated or risky12:33
SteveAok12:33
_thumper_sounds good12:33
ddaaQuite unusually, we are ahead of time.12:34
mpoolkeep it up!12:34
ddaa== Any other business? ==12:34
ddaaSo, I'll wait for 10 minutes there ;)12:34
lifelessjamesh: details :  12:35
lifelessAssertionError: {'converted-by': 'launchpad.net', 'cscvs-id': 'MAIN.1', 'branch-nick': 'bzr_branch'} != {'converted-by': 'launchpad.net', 'cscvs-id': 'MAIN.1'}12:35
lifelessAssertionError: {'converted-by': 'launchpad.net', 'cscvs-defaultfiller': '1.1.1.1 1.1 file_name', 'cscvs-id': 'MAIN.1', 'branch-nick': 'bzr_branch'} != {'converted-by': 'launchpad.net', 'cscvs-defaultfiller': '1.1.1.1 1.1 file_name', 'cscvs-id': 'MAIN.1'}12:35
jameshyeah.  I see it12:35
_thumper_I did hit two problems with utilities/launchpad-database-setup12:35
mpoolspiv: jameinel will be at my house tomorrow12:35
SteveAso, I asked earlier12:35
spivmpool: Cool.  I'd be happy to join you.12:35
ddaaSo, I'm done for the meeting.12:35
ddaa<insert random rant here about cscvs>12:35
SteveAabout whether anything will happen about overlay repositories or similar thing12:36
spivRight.12:36
lifelessthere has been no serious discussion about it on list12:36
SteveAI'd also like to know about supermirror/smartserver stuff in more detail than "delayed"12:36
SteveAlike, delayed by how much12:36
lifelessunless its a mark-mandated feature or some such, I'd say its way off. Myself, I'm not convinced its even a feature rather than a trap.12:37
mpoollifeless: take it you're refering to overlay repos, not the smart server?12:37
lifelessright12:37
lifeless:)12:37
mpool:-)12:37
mpoolphew12:37
ddaampool: thanks for the clarification :)12:38
SteveAit feels silly to me to be storing and transporting revisions to the server we use many times more than necessary.  it's a problem I think it important to be fixed for other projects that work as launchpad does, assuming there are/will be any12:38
=== ddaa thinks overlay repos sounds like useful
SteveAthe problem can be fixed in many different ways, I'm sure12:38
mpoolSteveA: i think the motivation for it is definitely good12:38
mpooli think i suggested (or at least thought :-) that you should make a braindump spec12:38
SteveAwell12:39
mpoolthat will give it at least some lease of life12:39
SteveAI can describe the problem12:39
jameshlifeless: weird.  I thought I'd done a complete run of the test suite.  I'm uploading the fix now12:39
SteveAbut it appears, from lifeless's comments, that I'm not qualified to propose a solution12:39
mpoolwell12:39
mpoolit's a valid identification of a problem12:39
ddaaactually...12:39
lifelessthis is a wicked problem space12:39
mpooli don't think we should run into the design of the solution in this meeting :)12:39
mpoolbut we can talk about it this week with john12:39
lifelesswe tend to evaluate many proposals before finding a good balance12:40
ddaathinking of it, I think it's close to congruent to the smart server12:40
lifelesshaving a clear statement of the problem is very useful to guide those evaluations12:40
mpoolddaa: it's definitely related12:40
SteveAmpool: a good description of the problem is on the mailing list12:40
mpoolSteveA: so i guess you're mainly concerned at this point that it not just be dropped/ignored?12:40
SteveAand john did answer the email12:40
SteveAdo you want me to do something else with it?12:40
mpoolSteveA: yes, that's what i'm looking at12:40
ddaaunless convinced otherwise, I'll believe that SteveA wants a smart server with ACL12:40
mpooli read it too12:40
mpoolno, i'll paste it into the spec tracker12:41
mpoolddaa: what does this have to do with acls?12:41
spivddaa: that's one possible solution, but I don't think it's the only one or necessarily the obviously correctly one.12:41
ddaaI mean, access control12:41
mpoolSteveA:  the thread "Overlay Repositories"?12:41
ddaaokay, I'm just talking out of my ass12:41
mpoolSteveA: is that an ok outcome for you, for hte moment?12:41
SteveAmore or less I'm concerned with it not being dropped.  I want to start getting people to look at bzr for their VCS system, and issues like this make people feel like something is clumsily designed, even when that's not the case12:42
SteveAbecause it takes an insider to see the design trade-offs or the unfinished work12:42
SteveAI'm happy with that outcome.  thanks12:42
ddaaWell...12:43
spivSteveA: Also, you wanted to discuss launchpad developers testing the smart server on devpad?12:43
SteveAspiv: thanks.  yes, that seemed like a good way to get some testing out of the system.12:43
spivWe should get bzr 0.11 installed on devpad.12:44
ddaaspiv: yeah, gimme smart server, I give 50 dinars!12:44
SteveAmpool: also, did you ask etienne about packaging bzr-pqm ?12:44
SteveAor rather, whatever the plugin for pam-submit is called12:44
mpoolsorry, no12:44
jameshbzr-pqm is the right name12:45
spiv(it's possible to get people using it without installing it system-wide, but it'll be unnecessarily complex)12:45
ddaaMeeting nominally closed. Please go on.12:45
SteveAthanks ddaa, for a swift meeting12:45
SteveAspiv: when bzr 0.11 is packaged, we can get it installed on devpad12:45
spivThere's a package at http://bazaar-vcs.org/releases/debs/ -- is that sufficiently good?12:46
SteveAif it is for dapper, yes12:46
SteveAdo we also need to get launchpad developers to use 0.11 ?12:47
SteveAif so, "scratchy" should be updated12:47
spivThe package from there is working on my dapper system.12:47
spivOnce 0.11 is installed on devpad, I'll be very happy to mail the launchpad list with details of how to test it and what to expect.12:48
SteveAwell12:48
SteveAwe have a way to get launchpad developers to use it12:48
mpoolSteveA: apparently 'scratchy' split into different directories to accomodate edgy12:48
SteveAthat is, we put it in that "scratchy" apt repository12:48
SteveAI don't want to give launchpad developers a change to their sources.list without good cause12:49
ddaaSteveA: I'd love if we could get some traction on branch-notification-email too. It's more complicated, but not terribly so, so maybe we could put that on the sprint too?12:49
SteveAand I think the "scratchy" thing should be kept up to date with what we want launchpad developers to be using12:49
SteveAddaa: sure12:49
ddaaespecially since having you handy to point to the right infrastructure bits needed12:49
SteveAddaa: we'll see how that goes.  I added it to the agenda12:50
mpoolSteveA: we should probably just make a different repository for lp developers on edgy12:50
lifelessI thought scratchy was mant to be empty when the current release was in the distro ?12:50
ddaaSteveA: where's the agenda? Here's what I propose: hiding merged/abandoned branthes by default, branch-spec listing, branch-notification email.12:50
SteveAmpool: that's fine.  My concern is if we set up a repository, tell launchpad people to use it, and then forget about it12:50
lifelessmpool: thats what the subdirs in scratchy are12:51
SteveAit needs to be something that is maintained, so that we're not changing people's config12:51
lifelessmpool: scratchy-for-edgy, scratchy-for-dapper12:51
SteveAand so that we don't roll back versions for people12:51
lifelessanyhow, you should go sleep, as I. early morning tomorrow.12:52
SteveAddaa: the draft agenda is in a tomboy note on my laptop.12:52
ddaas/branch-spec listing/branch-spec linking/12:52
SteveAddaa: bring other things with you to the meeting and we'll set a firm agenda as the first thing in london12:52
mpoolSteveA: would it be too intrusive to get them to update their apt-sources once to the new url?12:53
jameshlifeless: I've fixed the cscvs test failure, so the merge should go through fine next time12:53
=== spiv -> dinner
lifelessjamesh: I'll try again12:53
lifelessjamesh: tomorrow, theres a merge pending12:53
lifelessnight all12:54
jameshthanks.12:54
SteveAmpool: if we have a new URL, I want it to be one that we can put all .debs that launchpad developers need in.12:54
SteveAthat includes the launchpad-dependencies things, and any other special debs they need12:54
SteveAas well as bzr tests12:54
lifelesswell, thats really something that should be hosted by the support team12:55
ddaajamesh: just to be sure12:55
lifelessrather than being on the bzr website TBH12:55
lifelessso I propose it be given as a request to Jeff Bailey12:55
ddaayou fixed the code not to set the branch-nick, right?12:55
SteveAit should be hosted by the sysadmin team12:55
SteveAand maintained by the support team12:55
SteveAwith input from the bzr team about what bzr version goes in there12:56
SteveAthen used by the launchpad team12:56
SteveAteam12:56
mpoolOK12:56
mpooli'll send a mail12:56
mpoolis there any equivalent of x-debbugs-cc for RT?12:56
SteveAwhat does that mean?12:57
mpoolany way to subscribe you to the ticket12:58
SteveAthere is in the web interface12:59
SteveAI don't know about via email12:59
SteveAboth the support people and the sysadmins use RT12:59
SteveAso they may know12:59
SteveAand this is kind of a support question :)12:59
=== poolie [n=mbp@ppp112-44.static.internode.on.net] has joined #launchpad-meeting

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