ReleaseEngineering/Meetings/2008-sep-22

Fedora Release Engineering Meeting :: Monday 2008-09-22

F10 Beta Delayed

• Still fighting installer issues
• Test results page is empty http://fedoraproject.org/wiki/QA/TestResults/Fedora10Install/Beta
• Proposing new dates to FESCo
  • Push all dates forward one week
  • Beta release on 2008-09-30
  • GA date of 2008-11-25

IRC Transcript

f13	ping: spot jeremy lmacken warren notting poelcat rdieter wwoods	10:00
rdieter	yo	10:00
* spot is here		10:00
* lmacken		10:00
* jeremy is around-ish		10:00
* warren is on drugs.		10:00
* wwoods here		10:01
f13	notting txtd me and said he's running late due to errands.	10:01
-!- f13 changed the topic of #fedora-meeting to: Fedora rel-eng - F10 Beta		10:02
f13	so... the beta that just won't be.	10:02
warren	how's X doing?	10:02
rdieter	boo	10:02
f13	we're still fighting installer issues, so beta is going to slip	10:02
f13	there just isn't enough time to do a compose today of everything and have it fully staged by Thursday	10:02
f13	warren: better?	10:03
f13	right now we're looking at more networking issues in anaconda, after fixing some language selection issues, twice.	10:03
warren	f13: is this pushing us against thanksgiving?	10:03
f13	we tried a couple builds over the weekend but it's just not in place.	10:03
f13	warren: we'll have to talk about that. We can either take a week in post-beta, or slip a week or two after turkey day	10:04
warren	going past thanksgiving at least makes it happen before christmas...	10:04
warren	taking away beta time seems unwise IMHO	10:04
warren	but we can discuss that later	10:04
* poelcat here		10:05
f13	so the thing is right now, we're still got a GA date of the 18th	10:05
f13	which is the week before thanksgiving	10:06
f13	even if we slipped a week and did GA on the 25th, that's two days before, /and/ we would have had to be fully done with development nearly a week before that anyway	10:06
wwoods	so slipping a week closer to thanksgiving is probably not disastrous	10:09
f13	that's my opinion	10:09
f13	and if we fail later, we just slip right over thanksgiving	10:09
warren	The week of thanksgiving is pretty much a loss	10:10
warren	people wont be around	10:10
warren	people will be traveling	10:10
notting	people *with lives* won't be around.	10:10
warren	people will be setting up their staycations in their patio	10:10
spoleeba	notting, i have no live... i will be travelling	10:11
warren	f13: ok, so we need to slip due to installer issues, anything need to be decided?	10:14
f13	I propose that we slip to releasing Tuesday the 29th	10:15
f13	push the remainder of the schdule dates out a week as well	10:15
notting	how broken is broken?	10:15
notting	'no trees now, so we're kind of screwed'?	10:15
f13	notting: media installs traceback	10:15
f13	notting: networkinstalls don't start unless you force network bring up in stage1	10:16
f13	(and currently traceback on something)	10:16
f13	so we have trees, they're just useless for installing	10:16
f13	they're essentially "little" things, but it's one of those clear the pathways to find more bugs kind of deals.	10:17
f13	that and I think our testing grid is rather... blank at the moment, and I'm a bit curious why that is	10:17
f13	wwoods: where is that testing matrix again?	10:18
f13	http://fedoraproject.org/wiki/QA/TestResults/Fedora10Install/Beta	10:18
f13	utterly blank.	10:19
wwoods	we keep respinning the tree and I keep invalidating my results.	10:19
f13	ok, but if we have failures, don't those stay failures until we have a tree that works?	10:20
f13	(maybe we need a way of marking this?)	10:20
warren	if network installs fail	10:21
warren	and media installs fail	10:21
warren	that sounds like fail?	10:21
poelcat	shouldn't we be filing bugs?	10:21
f13	warren: was there something useful to the discussion there?	10:21
warren	f13: no.	10:22
poelcat	and watching blocker bugs	10:22
f13	poelcat: yes, we should, I asked the original reporter of the network traceback if he filed a bug, however I think clumens already fixed it.	10:22
f13	poelcat: I've been entirely too busy making trees show up to do any testing this time around.	10:22
* poelcat was just thinking of from the perspective of tracking failures		10:23
f13	poelcat: sure, there are known bugs that haven't been filed, I'm going to try and sort those out here shortly	10:24
f13	and we really need to attack the testing matrix instead of running around like chickens with our heads cut off trying to bandaid failures.	10:25
f13	so I think we need to float this proposal of a week slip to FESCo just to get their stamp on it and go to work.	10:26
f13	As far as getting bits to test, mmcgrath hooked me up with alt.fedoraproject.org	10:28
f13	http://alt.fedoraproject.org/pub/alt/stage/	10:28
f13	which has rsync as well	10:28
f13	I've been staging live images and spins as I produce them.	10:28
f13	so if you can't get to things on reducto (and there aren't live images there so you can't) use alt.fedoraproject.org	10:28
f13	the live images there are a few days old, didn't put up the latest due to busted anaconda.	10:28
f13	butthat's where they'll show up for testing.	10:29
f13	anything further on Beta?	10:30
-!- f13 changed the topic of #fedora-meeting to: Fedora rel-eng - Open Floor		10:32
f13	anything else?	10:32
wwoods	nope, I think you pretty much hit it on the head	10:32
notting	f13: you sending the proposal fesco-wards? probably should do e-mail	10:32
f13	notting: yeah, I'll send the mail.	10:34
f13	I'm going to float it around internally for schedule conflicts as well.	10:35
f13	alright, calling the meeting.	10:35

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