ReleaseEngineering/Meetings/2008-nov-17

From FedoraProject

< ReleaseEngineering | Meetings
Revision as of 01:17, 20 November 2008 by Poelstra (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Fedora Release Engineering Meeting :: Monday 2008-11-17

Summary

  • Dicsussion about upcoming GA and blocker bugs

IRC Transcript

notting f13: jeremy: warren: spot: rdieter: wwoods: ping? (i probably forgot someone...) 13:09
* jeremy is here-ish 13:09
* rdieter waves 13:09
* warren here 13:09
spot f13 is ill today i think 13:09
notting ah, ok 13:10
notting well, that's a bit of a monkey wrench 13:11
wwoods so we're out f13 and ajax, mezcalero's been on holiday for a while 13:11
* jeremy is starting to feel better! :) 13:11
notting wwoods: there are no PA bugs on the blocker 13:11
wwoods well hey! that's a good sign 13:11
notting unless i missed something 13:11
wwoods no, but there's plenty of bug reports and angry users 13:11
wwoods I've just been putting my fingers in my ears 13:11
jeremy we definitely need an anaconda build for turning off betanag... clumens and I talked about branching and doing it today 13:12
wwoods yeah, today's the day 13:12
notting please do 13:12
notting looking at blockers 13:12
notting we have a plymouth one in MODIFIED, and i think halfline has a new plymouth build for this morning's bug of the day as well 13:12
notting halfline: going to request a tag for that? 13:12
halfline notting: well i'm waiting on feedback on a bug 13:13
wwoods I only bring up pulseaudio because I've been trying to listen to music to block out this irregular beeping somewhere nearby and it's been stuttering and generally making me stabulous 13:13
halfline bug 471498 has a bit of a mystery failure in it 13:13
buggbot Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=471498 high, high, ---, rstrode@redhat.com, MODIFIED, plymouth hangs in startup 13:13
* spot wonders why we have two blocker bugs with PATCHes in them that are sitting for more than a week 13:13
cebbert wwoods: erase it 13:13
halfline that i can't reproduce on the machines I try, but the reporter can reproduce on every machine he tries 13:13
notting halfline: has the reporter tested the fix? 13:13
halfline notting: i don't think so, that's what i'm waiting on 13:14
notting ok 13:14
wwoods cebbert: yeah, that's a workaround.. 13:14
halfline once i get feedback i'll request a tag 13:14
notting we have two dmraid bugs, and two grub timeout bugs 13:14
* notting looks at pjones 13:14
* spot wonders why the patches in 468461 and 448247 aren't applied 13:15
wwoods the latter (mkfs.vfat) we were waiting on a review from either the fedora package maintainer or the upstream maintainer 13:15
wwoods the last comment in that bug is from upstream 13:16
wwoods ("Got ping from notting, will have a look at this tomorrow and release 3.0.1.") 13:16
notting spot: re: 468461, the fact that it's *completely* untested probably has a lot to do with it ;) 13:16
notting we also have a cryptsetup bug that seems like keyboard error, so i'm not sure we'll hold for that 13:17
notting and X bugs for both radeon & intel 13:18
wwoods yeah I think the cryptsetup bug should drop to blocker or get closed as WORKSFORME 13:19
notting i moved it to target 13:19
wwoods reporter can't reproduce (he was given a package with debugging stuff enabled but no new code) 13:19
wwoods good call 13:19
notting but still... not sounding particularly hopeful on having something final today 13:20
notting jeremy: how do live images look? ok space wise? 13:20
wwoods there's more radeon badness than what's on Blocker, I just don't have the hardware/time/expertise to distill it all down to one or two actual bugs 13:20
jeremy notting: they're okay as far as I know... I did one earlier that was 680 13:21
notting got a link to latest images? 13:21
warren how bad are the "I need nomodeset" bugs? 13:21
jeremy notting: I just did one locally, didn't really put it up anywhere 13:21
wwoods it's hard to say, since people will continue reporting them long after they're fixed (because they're testing with Preview or hitting other bugs) 13:21
wwoods and I don't have any radeon test hardware so I have no first-hand knowledge of any of it 13:22
* notting wonders how many non-radeon users are currently passing 'nomodeset' in the hopes that it will fix their problems 13:22
wwoods notting: *lots* 13:22
rdieter ouchie 13:23
wwoods most of the intel bugs we see, the reporter says something like "I tried adding nomodeset but it didn't work" 13:23
* spot isn't having problems with the only radeon he has access to... 13:23
notting oh 13:24
notting jeremy: pjones: there's a report of things being fubar in either the installer or post-boot initrds due to glibc-2.9 13:24
halfline clumens, jrb, and others are currently seeing a massive slowdown with xorg-x11-drv-ati 13:24
wwoods sometimes even on nvidia bugs 13:24
pjones notting: uh-oh 13:25
wwoods although nvidia bugs are fairly rare since that's mostly unchanged since f9 13:25
notting pjones: ld-linux.so.2 being a symlink to the wrong place, or something :/ 13:25
halfline spot: jrb's problem goes away on a fresh reboot 13:25
jeremy notting: I updated my boxes earlier; workstation rebooted fine, haven't rebooted my laptop yet 13:25
pjones notting: which arch? 13:25
halfline spot: so it could be a problem related to keeping it booted and suspending/resuming for a long time 13:25
spot hmm. i'm not sure that suspend/resume works at all right now 13:26
notting pjones: don't know, it was on the list :/ 13:26
notting pjones: anyways, while you're here :P 13:26
notting comments on the grub and dmraid bugs? 13:26
wwoods yeah I've heard further suspend/resume complaints 13:26
wwoods beyond just intel+compiz+suspend = hang 13:26
pjones happen to have a link to the list? 13:27
notting found it 13:27
notting https://www.redhat.com/archives/fedora-devel-list/2008-November/msg01129.html 13:27
halfline spot, wwoods: well we ran sysprof on his machine while it was having issues and we found out it was spending a significant amount of time in one function looping over a list of buffer objects 13:27
halfline so one theory was that the list was getting really long after a lot of use 13:28
halfline anyway airlied will be around in a couple of hours and he should be able to shed some light on that issue 13:29
wwoods realistically, if we want to release on-time, we need a final package list appx. today 13:32
wwoods and any major bugs found (or unfixed) should cause us to slip the schedule. 13:32
notting we have a new kernel (113) in koji that fixes some gfx issues 13:32
wwoods well. maybe that's not "realistic", since the *realistic* approach is that we're going to release anyway 13:33
wwoods but now seems like a good time to start the traditional "slip/don't slip" argument 13:33
spot well, strictly speaking, the only radeon blocker is 467821 13:34
halfline hmm, marina still has the "need to move mouse to get graphics to update" issue with -109 13:34
notting would like testing of radeon with the new kernel 13:34
spot (not to put our head in the sand or anything) 13:34
halfline does anyone remember the bug number for that one? 13:35
wwoods strictly speaking there's probably other radeon blockers that are unfiled or misfiled 13:35
spot wwoods: well, please look for them? 13:36
spot i'd rather not start discussing slips for bugs that might exist 13:36
notting can i get another +1 on tagging the new kernel? 13:36
spot notting: i'm testing it now 13:37
warren which kernrel? 13:38
notting 113 13:38
halfline ah the intel bug is bug 471162 13:39
buggbot Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=471162 medium, medium, ---, airlied@redhat.com, NEW, kernel-2.6.27.5-100.fc10.x86_64 causes intel 945GME to loose irq 13:39
wwoods jeremy: you've got an open bug about radeon crashing (bug 468773) - any progress since your last comment? 13:40
buggbot Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=468773 medium, medium, ---, airlied@redhat.com, ASSIGNED, Non-working radeon driver on macbook pro with nomodeset 13:40
* jeremy wonders which macbook that was with 13:41
jeremy it was definitely when booting through efi which has some other problems in other cases 13:42
notting efi is a 'you get all the pieces' feature in f10, yes? 13:42
spot 113 boots fine on my radeon test box (desktop effects don't work, but they've never worked) 13:42
warren spot: what chipset? 13:42
jeremy notting: I believe that is the current status. the bug was filed before we downgraded to that point 13:43
spot Radeon HD 3200 13:43
wwoods bug 462157 has a really unhelpful title but it's a hang on radeon r500 13:45
buggbot Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=462157 urgent, urgent, ---, airlied@redhat.com, ASSIGNED, X hang up 13:45
warren Bug #471162 I have this chipset and I have never seen this on 109. It does sound nasty though. 13:46
buggbot Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=471162 medium, medium, ---, airlied@redhat.com, NEW, kernel-2.6.27.5-100.fc10.x86_64 causes intel 945GME to loose irq 13:46
spot warren: ditto, never seen that issue 13:47
notting i've not seen that symptom, but i've seen X spontaneously go out to lunch on i945, requiring reboot 13:47
wwoods notting: recently? that was supposedly fixed last week, and f13's test machine stayed alive for 3 days without doing that 13:47
notting wwoods: over the weekend 13:48
wwoods so we closed it (or maybe dropped it from Blocker) 13:48
wwoods well hell 13:48
wwoods oh wait, that's bug 464866, which is still open and on the blocker 13:48
buggbot Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=464866 medium, medium, ---, ajax@redhat.com, ASSIGNED, Xorg lockup with i945: "EQ overflowing. The server is probably stuck in an infinite loop." 13:48
warren brb, rebooting into 113 13:49
halfline wwoods: marina is seeing that issue right now... 13:50
halfline (like i said 10 minutes ago) 13:50
notting wwoods: feel free to blocker-ize 462157 if necessary 13:51
notting so, before making a decision on slip vs. no slip, i'd like to characterize which of the X issues are still prevalent with the latest kernel, and of those, whether or not slipping is realistically going to lead to a timely fix 13:52
warren Hmm, desktop effects are working, but it is considerably slower than F9. 13:53
notting also, some sort of comment on 468649, 467904, 468461, 468526 as to whether they're slip-worthy, or we can expect a fix 13:54
halfline notting: wrt to "timely fix" airlied is the best person to ask. krh is working on f11 work atm, ssp is working on non-X stuff... 13:55
notting halfline: right, which makes it not a discussion for right now 13:55
halfline yea 13:55
notting so, we're in a bit of a holding pattern. all up for a re-convened blocker meeting at 4PM once we have a bit more data (and access to picking X people's brains?) 14:00
wwoods sounds reasonable. 14:00
notting spot: warren: ok to move/tag/sign 113? 14:00
notting (we also have a few things in the tag queue waiting for +1/-1... get them while they're hot) 14:00
warren notting: seems OK on two of my systems here 14:01
notting jeremy: also, if we can get a final-ish anaconda by then, just to knock it off the list 14:02
jeremy notting: I'll throw something at clumens 14:02
notting ok then. thx all, back in two hours on #fedora-blocker for further updates. 14:04
notting anyone got anything else? 14:04
wwoods ulcers 14:05
* wwoods grabs -113 and reboots 14:06
notting plz check with your doctor or pharmacist. the TV tells me there are many things for that. 14:06

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