18:02:02 <Oxf13> #startmeeting Fedora Release Engineering Meeting
18:02:02 <zodbot> Meeting started Mon Dec 14 18:02:02 2009 UTC.  The chair is Oxf13. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:02:02 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:02:10 <Oxf13> #meetingname feodra-releng
18:02:10 <zodbot> The meeting name has been set to 'feodra-releng'
18:02:19 <Oxf13> #topic roll call
18:02:57 <Oxf13> ping: notting jwb rdieter_work wwoods lmacken spot dgilmore poelcat warren
18:02:59 * poelcat here
18:03:30 <rdieter_work> yo
18:03:32 * dgilmore is here but is trying to bring builders back up
18:03:34 * notting is here
18:04:07 <Oxf13> ok, lets get started
18:04:50 <Oxf13> #topic old business (F13 Schedule)
18:05:06 <Oxf13> Last meeting we had talked about moving F13 schedule to not conflict with Ubuntu
18:05:20 <Oxf13> we were going to seek more data and report back, but the end result was that we moved it
18:05:33 <Oxf13> poelcat: is there any open business related to the F13 schedule?
18:07:50 <poelcat> Oxf13: any feedback to latest version i posted
18:08:06 <Oxf13> poelcat: did anything change from what we went over at FUDCON?
18:08:08 <poelcat> and we need to decide what terms we are going to use for the "freezes" that aren't "freezes" :)
18:08:41 <poelcat> Oxf13: only the things we discussed
18:08:53 <poelcat> changing or removing
18:09:06 <Oxf13> http://poelstra.fedorapeople.org/schedules/f-13-draft/f-13-releng-tasks.html
18:09:10 <Oxf13> that's the link by the way
18:10:44 <poelcat> i'm a little concerned that if we take "freeze" out of the milestone
18:10:50 <poelcat> people will take it less seriously
18:11:09 <poelcat> not that they do that much anyway :)
18:11:11 <Oxf13> yeah
18:11:26 <notting> well, it's still a code freeze. there's nothing to say we have to take your code :)
18:11:26 <poelcat> for exampele #40 used to be "Beta/Final Freeze"
18:11:28 <Oxf13> so, the other strategy we could use is to call them freeze (blocker fixes only)
18:11:58 <Oxf13> I think that's still accurate
18:12:30 <Oxf13> and would give it the oomph that poelcat is looking for
18:12:40 <Oxf13> (yes, I know this reverses what I've said previously)
18:13:20 <poelcat> how about if I start a thread on releng list w/ the ones i think we should clarify?
18:13:27 <Oxf13> so 18 could be "Start Alpha Freeze (blocker fixes only)"
18:14:08 <poelcat> wouldn't we still take !blocker fixes to non-crit path packages?
18:14:16 <Oxf13> a new line 41 coudl be "Start Beta Freeze (blocker fixes only)"
18:14:17 <poelcat> or maybe i'm not understanding
18:14:33 <Oxf13> poelcat: that is a good question
18:15:22 <Oxf13> if you wanted to make the line longer "Start Alpha Critical Path Freeze"
18:15:38 <Oxf13> and repeat for Beta/Final
18:17:09 <Oxf13> but that's just adjusting labels on the dates, I think we're pretty solid on the dates enough to stamp this schedule approved
18:17:22 <poelcat> yes
18:17:50 <poelcat> i wanted to get the terms right so I can include as part of a campaign to send out better/more reminders to devel
18:17:53 <poelcat> about what is coming
18:17:58 <poelcat> schedule wise
18:18:11 <poelcat> that's all, we can move on
18:18:15 <Oxf13> #proposal Accept Fedora 13 schedule, lables can be adjusted as we go
18:18:25 <Oxf13> +1
18:18:50 <Oxf13> shoot, I have to step out for a minute, a door got locked with nobody inside of it.
18:18:59 <notting> +1
18:19:29 <rdieter_work> +1
18:21:10 <dgilmore> +1
18:21:57 <poelcat> +1
18:24:47 <poelcat> while were waiting for Oxf13 to return
18:24:50 <poelcat> did everyone see http://poelstra.fedorapeople.org/schedules/f-13-draft/f-13-releng-tasks.html
18:24:56 <poelcat> oops
18:24:58 <poelcat> wrong link
18:25:23 <poelcat> https://fedoraproject.org/wiki/Release_Engineering_Release_Tickets
18:25:36 <poelcat> i think i captured all the feedback
18:26:24 <poelcat> it looks a lot like the schedule, but each line represents tickets that we will file in Trac
18:26:35 <poelcat> thoughts, comments, questions, concerns?
18:26:59 <Oxf13> I had some response, didn't get a chance to make it this morning
18:27:05 <Oxf13> but I think it's a good enough list for now
18:27:32 <poelcat> Oxf13: i'll work with you 1x1 to figure out how to take it to the next step
18:27:41 <poelcat> which was how to can auto-file them using this page
18:28:02 <Oxf13> k
18:29:09 <Oxf13> #agreed Fedora 13 schedule accepted.
18:29:21 <Oxf13> poelcat: since releng accepts the schedule, do we have to do anythign with FESCo or just post it?
18:30:06 <notting> fesco already approved it
18:30:14 <notting> it was on the agenda last friday
18:30:20 <notting> erm, friday before last
18:30:21 <Oxf13> oh haha
18:33:40 <Oxf13> ok
18:34:21 <Oxf13> #topic old business (LXDE Spin)
18:34:36 <Oxf13> As discussed at our last meeting, we re-spun LXDE using only a config change
18:34:44 <Oxf13> those new isos have been posted and as far as I know working great
18:36:05 <Oxf13> so that shoudl close out that issue.  Anything before w emove on?
18:36:50 <Oxf13> great.
18:36:59 <Oxf13> #topic No Frozen Rawhide
18:37:23 <Oxf13> I talked to lmacken on the bus ride up to FUDCon and he felt he could get the features we need for nfr into bodhi before the feature freeze
18:37:36 <Oxf13> he also thought he could get them into current bodhi rather than new bodhi
18:38:29 <Oxf13> so we'll be pinging you weekly for updates lmacken
18:38:56 <Oxf13> the second major road block is compose time, but we need the datacenter dust to settle before we test that
18:39:36 <Oxf13> anything else on this topic before we move on?
18:39:56 <Oxf13> #info lmacken feels he can get bodhi work done for n-v-r before feature freeze
18:40:08 <wwoods> any further discussion of when/how installer images will be pushed?
18:40:16 <Oxf13> #info compose performance work to be started after data center move is done
18:40:58 <Oxf13> #topic dist-git
18:41:13 <wwoods> I think the broad outline was something like: anaconda team will signal rel-eng somehow when a given anaconda build is good to use for building new images?
18:41:36 <Oxf13> wwoods: wrong channel?
18:41:38 <Oxf13> http://jkeating.fedorapeople.org/presentations/fudcon-die-cvs-die-2009.odp
18:41:45 <wwoods> no, this relates to no-frozen-rawhide
18:41:54 <Oxf13> I gave a talk at FUDCon about moving our package source control into git, the slides were linked to above
18:42:04 <wwoods> but whatever, some other time/place
18:42:24 <Oxf13> I've made some pretty good progress, I wrote up a script to use parsecvs to convert all of Fedora back to F7 and EL-4 into git with branches
18:42:45 <Oxf13> my next step is to publish this set of repos for anonymous cloning, and then hook in some way to do authenticated pushing
18:42:58 <Oxf13> while we play with tools to manipulate the sources
18:43:14 <Oxf13> as we play with this, I'm going to start drafting a Proposal wiki page that we can eventually shoot off to FESCo
18:43:22 <Oxf13> but that'll depend on what we find while playing with the conversion
18:43:35 <Oxf13> I'll definitely want some help, so if you're interested, ping me.
18:43:48 <Oxf13> Anything on this topic?
18:46:00 <Oxf13> alright, I'll be making regular reports about the progress here in these meetings.
18:46:20 <Oxf13> #info initial attempt at conversion complete, will be providing access to people for testing soon
18:46:38 <Oxf13> #info work to continue on writing tools to manipulate the repos
18:46:45 <Oxf13> #info find Oxf13 if you wish to help
18:46:54 <Oxf13> #topic PHX->PHX2 move
18:47:02 <Oxf13> Infrastructure had a fun weekend, that isn't over yet
18:47:51 <Oxf13> due to the move, we did not get a rawhide today
18:48:09 <Oxf13> there was some config changes to make, and now it looks like we're waiting on kojipkgs. to be brought up
18:48:24 <Oxf13> I'm going to try to get a rawhide out at some point.
18:48:35 <Oxf13> We also didn't have a signing server, but I just got that brought online
18:48:46 <Oxf13> and last I checked bodhi was having issues
18:48:52 <notting> Oxf13: well, given that no one could build, not sure there's a rush on rawhide :)
18:48:56 <dgilmore> we only have 2 builders
18:49:01 <Oxf13> and dgilmore is working to get the builders brought online
18:49:11 <Oxf13> so really, we're still pretty offline, but things are progressing.
18:50:21 <Oxf13> notting: no rush per se, but we want to make sure that task path is operational
18:50:45 <Oxf13> #info numerous releng services still offline, work continues to bring them up
18:50:53 <Oxf13> #info no rawhide yet, will attempt to get one out
18:51:08 <Oxf13> anything else on this topic?
18:51:53 <Oxf13> #topic open flooor
18:52:04 <Oxf13> any other topics worth discussing?
18:54:08 <wwoods> is there a plan/tool for building rawhide install images when the anaconda team (or QA or whoever) requests them?
18:56:08 <wwoods> ("not yet, but we'll think about it" is an acceptable answer, just trying to be clear about where things stand)
18:56:24 <notting> pjones and Oxf13 were discussing how to compose install images on #anaconda
18:57:05 <Oxf13> wwoods: run pungi on the rawhide tree, publish the images "somewhere"
18:57:27 <wwoods> so let's call that "we're discussing that with the anaconda team and will have a plan/SOP someday soonish" for the record
18:57:54 <pjones> wwoods: right now I've got a cronjob running every 3 hours (each an hour after we're syncing up the local mirror here) looking for new anaconda packages
18:58:04 <pjones> if it sees new packages, it builds a tree
18:58:41 <wwoods> nice. we can rig that (or something like it) up to autoqa for automated testing
18:59:04 <wwoods> sounds like everything's under control then. cool. thanks!
18:59:39 <Oxf13> hrm.
18:59:45 <Oxf13> so the problem I have is this
19:00:24 <Oxf13> if we're taking away images, but then anaconda team has to rig up some way to constantly have images, and QA team is going to rig something up to constantly have images, have we actaully made life better?