ReviewerMeeting2010mmdd

summary

• stuff
• more stuff

logs

ameu

[15:00] <bac> #startmeeting
[15:00] <MootBot> Meeting started at 09:00. The chair is bac.
[15:00] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[15:00] <bac> hi who is here for today's reviewer meeting?
[15:00] <jelmer> me
[15:00] <mars> me
[15:00] <sinzui> me
[15:00] <abentley2> me
[15:00] <adeuring> me
[15:00] <henninge> me
[15:00] <henninge> bac: danilo sends his apologies
[15:00] <leonardr> me
[15:00] <henninge> I think ;)
[15:01] <bac> gary, francis, julian are also away
[15:01] <mars> Foundations is here.  Record time :)
[15:01] <BjornT> me (although a bit busy with other things)
[15:01] <noodles775> me
[15:01] <bac> deryck too.  is gmb around today?  allenap?
[15:01] <gmb> me
[15:02] <gmb> bac, allenap is now a Landscaper for 6 moths.
[15:02] <adeuring> bac: allenap is on "LP holidays"
[15:02] <gmb> Er, months.
[15:02] <bac> ah, right
[15:02] <mars> gmb, ?  Did I miss that in the staffing mail?
[15:02] <bac> salgado: ping
[15:03] <mars> bac, salgado is off to other projects as well
[15:03] <bac> right, right
[15:03] <gmb> mars, Possibly; I wasn't paying attention. But he's definitely on rotation to LS.
[15:03] <bac> [topic] agenda
[15:03] <MootBot> New Topic:  agenda
[15:03] <bac> * Roll call
[15:03] <bac>  * Agenda
[15:03] <bac>  * Outstanding actions
[15:03] <bac>  * Mentoring update
[15:03] <bac>  * New topics
[15:03] <bac>    * ec2 failures - bac
[15:03] <bac>    * UDS next week.  No meeting or volunteer substitute chair?
[15:03] <bac>  * Peanut gallery
[15:03] <bac> [topic] outstanding actions
[15:03] <MootBot> New Topic:  outstanding actions
[15:03] <bac> * bac is a big loser
[15:03] <EdwinGrubbs> me
[15:04] <bac> i have two items on my plate that i haven't gotten to.  i *promise* to do them tomorrow so i can go to UDS with a clear conscience
[15:04] <intellectronica> me
[15:04] <bac> [topic] rockstar to start ML discussion about the lack of UI reviews
[15:04] <MootBot> New Topic:  rockstar to start ML discussion about the lack of UI reviews
[15:05] <bac> rockstar attends the AMEU meeting so he's not here now.  i don't think he did this, though.
[15:05] <bac> he brought it up in the meeting last week and had good statistics on the number of UI reviews that have been performed lately and it was very low.
[15:05] <bac> i'll remind him to take it to the list and we can chat there
[15:05] <noodles775> Yeah, I haven't seen any in a few weeks now (and few before)
[15:06] <intellectronica> bac: we've discussed this and we think the stats are wrong
[15:06] <bac> intellectronica: really?  that's good to hear.
[15:07] <mars> intellectronica, bac, true, but that discussion can probably be saved for the list
[15:07] <bac> sinzui: what is your impression?  have you been doing many?
[15:07] <intellectronica> because of ec2 land people don't commit with the quite tag
[15:07] <sinzui> I was doing many in week 1 and 2
[15:07] <sinzui> 1 last week
[15:07] <gmb> bac, intellectronica and sinzui did 3 or for for me last cycle, for sure.
[15:07] <bac> intellectronica: why is that?  i thought ec2 land did a good job of getting the ui tag?
[15:07]  * intellectronica shuts up
[15:07] <gmb> s/last/this
[15:07] <sinzui> I think this may be indicative that UI should land in weeks 0, 1, and 2
[15:08] <gmb> bac, Only if you tag the review with 'ui' in the mp.
[15:08] <gmb> Otherwise I think it just puts all the names in the [r=] block.
[15:08] <bac> gmb: ah, so we need to ensure the ui reviewers are tagging correctly or the MP author does it right.
[15:08] <gmb> Yes
[15:08] <bac> that's good information to pass on.
[15:09] <bac> so we have a failure in tools/procedure not lack of reviewing.  hurrah.
[15:09] <bac> moving on
[15:09] <bac> [topic] * ec2 failures - bac
[15:09] <MootBot> New Topic:  * ec2 failures - bac
[15:10] <bac> mars and i chatted this morning and we wanted to hear from people how often they are seeing ec2 jobs just disappear and the effect it is having.
[15:10] <bac> personally i see it a lot and it is very disruptive
[15:10] <jelmer> bac: I'm often seeing branches disappear, and it's quite annoying.
[15:10] <bac> for important branches i've been running them locally
[15:10] <mars> So far I know salgado and matsubara have both experienced a few test suite hangs in the past few days
[15:10] <bac> also, my ec2 bill last month was 4 times normal
[15:11] <gmb> bac, I've seen it happen twice this week already, it's getting annoyingly common.
[15:11] <gmb> I don't see many branches hanging; they just vanish.
[15:11] <bac> so how can we attack this problem to figure out what is going on?
[15:11] <mars> gmb, the instance dies?
[15:11] <jelmer> I haven't seen any hanging branches either (or do they get killed automatically?)
[15:11] <mars> jelmer, ^ ?
[15:11] <salgado> the ec2 instance shuts itself down after 8h
[15:12] <gmb> salgado, Ah, so they might well be hangs then.
[15:12] <noodles775> I've killed one hanging ec2 land after the log had not been updated for 2hrs.
[15:12] <mars> salgado, thanks, that is good to know
[15:12] <salgado> if it's left hung for long enough, it will shut down and you won't get any notifications
[15:12] <noodles775> (last week)
[15:12] <jelmer> mars: yeah, it just disappears silently
[15:12] <bac> did we have this problem before jml's subunit changes?
[15:12] <mars> jelmer, think it is inside our outside that 8 hour window?
[15:13] <jelmer> mars: outside probably
[15:13] <mars> ok
[15:13] <salgado> bac, I think we had it before.  do you know when the subunit changes landed?
[15:13] <mars> bac, the changes and fixes landed the 14th of April, your fix landed on the 21st
[15:14] <jelmer> I've definitely seen it before his changes as well.
[15:14] <bac> jelmer: ok, thanks.
[15:14] <mars> jelmer, salgado, that is also good to know.  Thanks.
[15:15] <mars> For everyone's benefit, I am tracking progress on this with Bug 570380
[15:15] <ubottu> Launchpad bug 570380 in launchpad-foundations "ec2test sometimes hangs on the first windmill test" [High,Triaged] https://launchpad.net/bugs/570380
[15:15] <bac> thanks mars
[15:15] <mars> I have some solutions being worked on for this, I will keep everyone updated as things progress
[15:16] <bac> i guess i'd encourage everyone to not just suffer in silence but (kindly) let mars know when it happens and try to gather some data
[15:16] <bac> [topic] * UDS next week.  No meeting or volunteer substitute chair?
[15:16] <MootBot> New Topic:  * UDS next week.  No meeting or volunteer substitute chair?
[15:16] <bac> i'm at UDS next week and a few others as well.
[15:16] <bac> shall we just skip the meeting?
[15:16] <bac> or would someone like to volunteer to lead it?
[15:17] <bac> i think the group silence is a vote to skip.
[15:17] <bac> [action] bac cancels next week's reviewers meeting
[15:17] <MootBot> ACTION received:  bac cancels next week's reviewers meeting
[15:17] <bac> [topic] mentoring update
[15:17] <MootBot> New Topic:  mentoring update
[15:18] <bac> noodles775, jelmer: everything still going well?
[15:18] <noodles775> Yep... although more non-soyuz reviews for jelmer would still be great :)
[15:18] <jelmer> I reviewed only two branches last week, mostly because I was distracted by other things. It's all going well otherwise though imo.
[15:19] <jelmer> and yeah, more non-soyuz branches please :-)
[15:19] <bac> [topic] peanut gallery
[15:19] <MootBot> New Topic:  peanut gallery
[15:19] <maxb> I have a peanut item: https://dev.launchpad.net/ReviewingCodeImports claims that community members can be members of ~vcs-imports, however, when it was brought up on the mailing list (subj: Code Import Reviews), the thread fizzled out with me producing a list of places where ~vcs-imports perhaps granted too-wide permissions. Someone should decide whether community members really _can_ be in ~vcs-imports or not, and either change the wi
[15:19] <maxb> ki page or work out how to make it happen.
[15:19] <bac> hi maxb
[15:20] <maxb> I believe this might dovetail with the promised community reviewers email :-)
[15:20] <bac> thanks for bringing that up max.
[15:20] <bac> maxb: touche
[15:20] <wgrant> Promised, legendary, or mythical?
[15:20] <wgrant> It seems to be in the realms of legend at the moment :P
[15:21] <bac> maxb: i'll bring it up in the ameu meeting this afternoon
[15:21] <maxb> thanks
[15:21] <sinzui> Looks like poolie and members of the Launchpad-code team can add you gentlemen
[15:21] <bac> any other topics?
[15:21] <mars> wgrant, it's not mythical, I've seen it!  I swear!
[15:21] <bac> it's right there in my drafts folder, wgrant.  really it is.
[15:22] <wgrant> Oh!
[15:22] <bac> going once
[15:22] <bac> twice
[15:22] <bac> thanks for coming everyone.
[15:22] <bac> * UDS next week.  No meeting or volunteer substitute chair?
[15:22] <bac> er,
[15:22] <mars> thanks bac
[15:22] <bac> #endmeeting
[15:22] <MootBot> Meeting finished at 09:22.

asiapac

[22:43] <bac> thumper, rockstar: ping
[22:44] <thumper> hi
[22:44] <rockstar> bac, we're still on the standup...
[22:44] <bac> ok
[22:44] <bac> how long until  you're done?
[22:44] <rockstar> bac, depends on chatty thumper
[22:45] <thumper> :-|
[22:45] <thumper> done now
[22:45] <bac> ok, then?  do i hear a chorus of 'me'?
[22:46] <rockstar> me
[22:46]  * rockstar choruses
[22:46] <bac> so in the ameu meeting we determined i was slack for not getting to my tasks
[22:46] <rockstar> I am also slack for not sending an email in my drafts folder about UI reviews.
[22:46] <thumper> what tasks?
[22:46] <bac> we also talked about rockstar's concerns about the UI reviews
[22:46] <rockstar> bac, yay.
[22:47] <bac> thumper: the same stuff that i've been ignoring:  community review policy and what constitutes testable documentation
[22:47] <thumper> :)
[22:47] <rockstar> bac, I has an announcement...
[22:47] <bac> rockstar: so the concensus is it isn't as bad as your data showed.  just that UI reviewers are marking the MP properly so ec2 land doesn't construct the correct PQM message
[22:48] <rockstar> bac, ah, that's what I was being told.
[22:48] <bac> we also talked some about the problem with ec2 losing branches with no email
[22:48] <rockstar> However, as a ui reviewer, I still don't feel as overwhelmed as I should be...
[22:48] <bac> it is really hurting us in lost time.
[22:49] <thumper> bac: that'll be the hang followed by the belt and braces shutdown
[22:49] <bac> thumper: what do you mean?
[22:49] <thumper> bac: perhaps the shutdown code should also email just before
[22:49] <rockstar> thumper, you mean "suspenders" ?
[22:49] <rockstar> :)
[22:49] <thumper> when the ec2 instances start up
[22:49] <thumper> we have some code that says
[22:49] <thumper> shutdown in 8 hours
[22:49] <thumper> or 6
[22:49] <thumper> whatever
[22:50] <thumper> a test hangs
[22:50] <thumper> machine gets shut down
[22:50] <thumper> no email
[22:50] <rockstar> thumper, although this doesn't always happen for me.
[22:50] <rockstar> Sometimes the machine is still running 10 hours later.
[22:50] <bac> thumper: hmm, i didn't know that.  i'll suggest it to mars
[22:50] <thumper> so perhaps before shutdown it should send suicide email
[22:50] <bac> thumper: indeed
[22:51] <rockstar> The subject should be "Goodbye cruel world"
[22:51] <bac> we also had this item from maxb:
[22:51] <bac> I have a peanut item: https://dev.launchpad.net/ReviewingCodeImports claims that community members can be members of ~vcs-imports, however, when it was brought up on the mailing list (subj: Code Import Reviews), the thread fizzled out with me producing a list of places where ~vcs-imports perhaps granted too-wide permissions. Someone should decide whether community members really _can_ be in ~vcs-imports or not, and either change the wi
[22:51] <bac> ki page or work out how to make it happen.
[22:52] <thumper> a bug should be filed about permissions that they shouldn't have
[22:52] <thumper> and we'll fix it
[22:52] <thumper> when the bugs are fixed
[22:52] <thumper> we could open it up again
[22:52] <bac> so 1) file bug on extra permissions, 2) fix bug, 3) open vcs-imports to community peeps?
[22:53] <thumper> aye
[22:53] <bac> excellent
[22:53] <bac> i'll find max's email and either file the bug or ask him to do it.
[22:53] <bac> [action] bac to see that a bug is filed about vcs-import permissions being to generous
[22:54] <thumper> cool
[22:54] <bac> also, i'm going to be at UDS next week so i'm cancelling this meeting.
[22:54] <bac> either of you going?
[22:54] <thumper> ack
[22:54] <thumper> nope
[22:54] <mars> thumper, good idea about the shutdown mail, I've noted it.
[22:54] <rockstar> nope
[22:54] <bac> i've heard to expect cold, rain, and extreme boringness of venue
[22:54] <rockstar> bac, I have an announcement.
[22:55] <bac> yes rockstar
[22:55]  * thumper has something to
[22:55] <rockstar> bac, noodles sent me an email today telling me to go ahead and announce that I'm a graduated UI reviewer.
[22:55] <bac> oh congratulations!
[22:56] <bac> so who are the full-fledged reviewers?  just you and noodles?
[22:56] <thumper> what about sinzui?
[22:56] <bac> i thought sinzui was still a UI-mentat
[22:57] <thumper> I've heard that noodles thinks he is good enough but sinzui is resisting
[22:57] <sinzui> I am
[22:57] <thumper> sinzui: stop resisting
[22:57] <thumper> resistance is futile
[22:57] <bac> uh oh, we woke him
[22:57] <rockstar> sinzui is gradiated.
[22:57]  * thumper sooths sinzui back to sleep
[22:57] <bac> congrats sinzui too
[22:57] <bac> thumper what was your item?
[22:57] <thumper> ok
[22:58] <sinzui> I am was mentoring intellectronica, but then he abandoned me.
[22:58] <thumper> more of a reminder for people to read my emails to launchpad-dev
[22:58] <thumper> there are two new testing things:
[22:58] <thumper> a better base breadcrumb test case, which uses test_traverse
[22:58] <bac> thumper: i read those.  i meant to bring them up today in ameu but forgot.
[22:58] <thumper> and self.oopses in the base TestCase class
[22:58] <thumper> I wrote up details for launchpad-dev
[22:58] <thumper> so details are there
[22:59] <sinzui> No one wants to be a UI reviewer. They do not like the cool sun glasses and hair cuts you get when you join the club
[22:59] <bac> and i'll certainly forget again in two weeks time
[22:59] <bac> thanks for those nice new tools thumper
[22:59] <thumper> np
[22:59] <rockstar> sinzui, I like my haircut!  :)
[22:59] <thumper> rockstar: isn't yours more like a lack of haircut?
[22:59] <thumper> :)
[23:00] <bac> which haircut are you sporting now paul?
[23:00] <rockstar> thumper, point
[23:00] <rockstar> bac, the unkempt "I haven't showered in 12 days" haircut.
[23:00] <bac> uh, nice
[23:00] <rockstar> It's called "hat"
[23:00] <bac> well i see we have devolved
[23:00] <bac> thanks for chatting.  see  you again in two weeks