17:00:50 #startmeeting F16-blocker-review 17:00:50 Meeting started Fri Sep 23 17:00:50 2011 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:50 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:01:02 #meetingname F16-beta-blocker-review 17:01:02 The meeting name has been set to 'f16-beta-blocker-review' 17:01:07 #topic rollcall 17:01:17 OK, who's ready for some blocker bug review fun? 17:01:34 * brunowolff is for at least a while 17:01:35 * jsmith is here 17:01:39 * jdulaney is here for a bit 17:01:44 Class at two 17:01:57 * tflink is hoping for a shorter meeting today, it's a shorter list than usual 17:02:16 brunowolff, jsmith, jdulaney: welcome 17:04:26 adamw: you around? 17:04:56 any volunteers to do secretary duty? 17:05:18 * jdulaney won't be here long enough 17:05:48 ok, let's get this party started 17:05:54 * nirik is lurking around. ping if I can help 17:05:57 #topic why are we here? 17:06:00 Woo, party time! 17:06:10 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 17:06:19 The list I'll be working off of: 17:06:30 s/I/We 17:06:36 #link https://fedoraproject.org/wiki/Current_Release_Blockers 17:07:02 I think that everyone here is familiar with what we're doing, but just in case ... 17:07:06 #info Our purpose in this meeting is to review proposed blocker and nice-to-have bugs and decide whether to accept them, and to monitor the progress of fixing existing accepted blocker and nice-to-have bugs. 17:07:32 any objections to starting with the proposed blockers? 17:07:59 Nope 17:08:03 #topic https://bugzilla.redhat.com/show_bug.cgi?id=736793 17:08:04 WORKSFORME 17:08:04 Bug 736793: unspecified, unspecified, ---, rstrode, ASSIGNED, gdm 3.1.90-1 No way to reboot or shutdown system from gdm 17:08:07 #link https://bugzilla.redhat.com/show_bug.cgi?id=736793 17:08:08 Bug 736793: unspecified, unspecified, ---, rstrode, ASSIGNED, gdm 3.1.90-1 No way to reboot or shutdown system from gdm 17:08:17 #info gdm 3.1.90-1 No way to reboot or shutdown system from gdm 17:08:39 Tough one 17:08:50 this is somewhat related to: 17:08:59 http://bugzilla.redhat.com/show_bug.cgi?id=739253 17:09:00 Bug 739253: unspecified, unspecified, ---, rstrode, MODIFIED, unable to shut down from gdm greeter 17:09:26 This could fit into "All release-blocking desktops' offered mechanisms (if any) for shutting down, logging out and rebooting must work" 17:09:43 but my question ends up being, is it really offered if it isn't there? 17:09:47 +1 17:10:00 There is a fix, but do we want to go that far? 17:10:06 AFAIK, the option to shutdown from gdm greeter is back in 3.1.92 17:10:26 Indeed 17:10:28 and I think that ties into the question surrounding 739253 17:10:33 * jdulaney has tested and verified 17:10:52 the fix IS in 3.1.92 but do we really want to pull in that whole update for beta? 17:11:15 * jdulaney wonders how hard it would be to get just the fix? 17:11:28 I think that I'm -.5 on blocker for this 17:11:34 But, it takes more than just that one package update to get the fix working 17:11:41 tflink: How much risk do you think there is in pulling in the 3.1.92 package? 17:12:06 jsmith: honestly, I'm not sure. I haven't looked into what all changed for .92 17:12:35 That we don't know why it is working in some situations and not others makes me nervous. 17:13:08 See adamw's comment at the end: https://bugzilla.redhat.com/show_bug.cgi?id=739253 17:13:09 Bug 739253: unspecified, unspecified, ---, rstrode, MODIFIED, unable to shut down from gdm greeter 17:13:09 it sounds like this happened because gnome-shell is now responsible for the greeter screen 17:13:54 Does that mean it works differently depending on whether or not gnome shell can run on your hardware? 17:14:14 it sounds like, yeah 17:14:37 It shouldn't do that 17:14:53 * tflink wonders if the power options from gdm greeter show up in fallback mode only 17:14:55 The gnome team is making some odd decisions of late 17:15:17 jdulaney: as long as it looks the same on fallback and shell, I don't think it's a problem 17:15:24 and it sounds like that is true for .92 17:16:11 with adam's comment on 739253, I'm leaning more towards -1 blocker on this 17:16:29 any other votes? I'm seeing +1 -~1 so far 17:16:40 * jsmith is undecided 17:16:58 * nirik is -1 blocker. 17:17:01 I don't like not being able to shutdown from GDM 17:17:03 If we reject this as a blocker, I think the criteria should be reworded to account for why it was OK to let it go in this case. 17:17:11 Indeed 17:17:18 It does hit the criteria 17:17:43 it does hit the criteria, but the fix is either big or messy enough to justify letting this one slide, I think 17:18:02 * jdulaney stands at +1 blocker 17:18:11 Then it should be classified as NTH. 17:18:36 How hard would it be to do a test compose with the big messy fix? 17:18:45 not very hard 17:18:46 I'd like to hear what the GNOME folks think, personally 17:18:56 there's a workaround tho thats pretty easy, no? 17:19:04 yeah, switch to a VT and reboot 17:19:09 Try a TC with the messy fix in the meantime? 17:19:30 I think it meets the criteria as stated for a blocker, though personally I don't think we need to block for it because you can login to do a graceful shutdown. 17:19:31 unforutnately, the messy fix isn't available 17:19:39 tflink: or just login and reboot/shutdown from there. 17:19:50 it would be a lot of work to backport the messy fix 17:19:54 I can confirm that the fix works, and I haven't had any major bugs with it 17:19:58 most likely not worth the effort 17:20:12 also, it will be fixed for people who update post beta. 17:20:29 Hence, why not go ahead to .92 with a TC? 17:20:30 it sounds like we are mostly -1 blocker 17:20:35 I have been using the .92 updates as they fix a problem I am having with the main menu icon in fallback mode. 17:20:41 * jdulaney was always a rebel 17:20:44 It seems to be OK in normal use. 17:20:45 jdulaney: becuse we would have to retest all of the desktop stuff 17:20:52 Ah 17:20:53 jdulaney: and re-test the entire set of desktop tests? 17:21:04 True, that 17:21:05 I suppose if folks think we have time... 17:21:22 No, we've slipped twice already 17:21:44 like jsmith said, it would be nice to have some input from the gnome folks on this 17:22:08 and adamw, since it sounds like he's more familiar with what's going on with these bugs 17:22:11 I think that perhaps the criteria could be reworded to say there needs to be a working way to shut gracefully using a gui interface. 17:22:14 well, from the desktop channel, they said backporting the fix was not really possible, and the fix requires new gnome-shell. 17:22:51 so we're down to shipping with the bug or re-testing the whole desktop 17:23:00 Tough call 17:23:23 I don't want to ship that bug 17:23:35 But, do we really have time to test everything? 17:23:36 shipping with that known bug sounds problematic. 17:23:42 So if you can shutdown from either gdm or gnome (for desktop spin) things are good enough for beta. 17:23:56 brunowolff: +1 17:23:59 brunowolff: +1 17:24:05 could .92 wind up in Final? 17:24:16 * nirik is -1 blocker. document bug in release notes. 17:24:23 I would expect at least .92 will be in final. 17:24:38 yeah, gnome 3.2 is supposed to be released by final 17:24:38 jdulaney: I expect final 3.2 to be in final. 17:25:22 In that case, I'll say -1 17:25:40 -1 to beta, but I will say +1 to this blocking Final 17:25:45 proposed #agreed - 736793 - RejectedBlocker - This bug does hit the release criteria but the fix involves pulling in a new version of shell and other packages - too big to retest this late. This will be fixed by final 17:25:58 also, we should note to revise the critera here. 17:26:02 (later) 17:26:21 I'd like to include that we think the criteria is too strict for beta. 17:26:28 Keep in mind, this really hits most criteria on the revision 17:26:37 In some ways 17:26:46 But, that needs to go to the list 17:27:07 #info agreed that the beta release critera are a bit too strict, should be modified such that this isn't a bug 17:27:21 isn't a blocker 17:28:03 #info #info the release criteria could be reworded to say there needs to be a working way to shut gracefully using a gui interface. So if you can shutdown from either gdm or gnome (for desktop spin) things are good enough for beta. 17:28:09 ooh, duble info 17:28:12 double 17:28:15 #undo 17:28:15 Removing item from minutes: 17:28:23 #info the release criteria could be reworded to say there needs to be a working way to shut gracefully using a gui interface. So if you can shutdown from either gdm or gnome (for desktop spin) things are good enough for beta. 17:28:42 so, ack/nak/patch on the proposal? 17:28:49 ack 17:28:49 ack 17:28:52 +1 17:29:00 #agreed - 736793 - RejectedBlocker - This bug does hit the release criteria but the fix involves pulling in a new version of shell and other packages - too big to retest this late. This will be fixed by final 17:29:06 release/ga should have more stringent requirements than beta. 17:29:23 on the bright side, we only had the one proposed blocker :) 17:29:35 Yay! 17:29:45 Magical Unicorns! 17:29:46 let's do the lone proposed NTH before moving on to the accepted blockers 17:29:56 #topic https://bugzilla.redhat.com/show_bug.cgi?id=737532 17:29:57 Bug 737532: unspecified, unspecified, ---, anaconda-maint-list, MODIFIED, FormatDestroyError: error wiping old signatures from /dev/sda-1: 1 17:30:00 #link https://bugzilla.redhat.com/show_bug.cgi?id=737532 17:30:01 Bug 737532: unspecified, unspecified, ---, anaconda-maint-list, MODIFIED, FormatDestroyError: error wiping old signatures from /dev/sda-1: 1 17:30:09 #info FormatDestroyError: error wiping old signatures from /dev/sda-1: 1 17:30:28 we already have +2 NTH from the comments, so not much is needed here 17:30:58 this is a bug that came out of the fix for another NTH (http://bugzilla.redhat.com/show_bug.cgi?id=737916) 17:31:01 Bug 737916: unspecified, unspecified, ---, dlehman, MODIFIED, FormatSetupError: invalid device specification 17:31:19 +1 for NTH 17:31:29 proposed #agreed - 737532 - AcceptedNTH - This is a fix for a bug that came out of another NTH, would be nice to have fixed for beta 17:31:33 ack/nak/patch? 17:31:38 ack 17:31:46 #agreed - 737532 - AcceptedNTH - This is a fix for a bug that came out of another NTH, would be nice to have fixed for beta 17:32:36 alrighty, let's go over the non-VERIFIED beta blockers 17:32:44 +1 17:32:52 #topic Unable to make system bootable due to bootloader choice 17:33:01 #link https://bugzilla.redhat.com/show_bug.cgi?id=738964 17:33:03 Bug 738964: unspecified, unspecified, ---, anaconda-maint-list, NEW, Unable to make system bootable due to bootloader choice 17:33:36 this one is still in active testing, hopefully the .6 image in https://bugzilla.redhat.com/show_bug.cgi?id=738964#c65 will be the last one 17:33:37 Bug 738964: unspecified, unspecified, ---, anaconda-maint-list, NEW, Unable to make system bootable due to bootloader choice 17:34:11 #info needs testing with http://dlehman.fedorapeople.org/updates-738964.6.img 17:34:21 but I don't think that much else needs to be said about this one 17:34:35 a new anaconda build is one of the bigger things keeping us from TC3/RC2 atm 17:35:01 we have a small pile of patches on the mailing list to get built in today. it'll happen. 17:35:16 #info there is an alternative update available that doesn't use quite so big of a hammer - http://dlehman.fedorapeople.org/updates-738964.6.img 17:35:46 clumens: thanks for the update 17:36:01 Apologies, but I must take my leave 17:36:33 #info there are patches on anaconda-devel@ that should be built in to today, no problems expected in getting a new build 17:36:44 jdulaney: thanks for being here 17:37:01 unless there are other concerns/updates on this bug ... 17:37:05 shore 17:37:34 #topic (739746) dhcp / bind mismatch on f15 to f16 upgrade: no network 17:37:40 #link https://bugzilla.redhat.com/show_bug.cgi?id=739746 17:37:41 Bug 739746: unspecified, unspecified, ---, atkac, MODIFIED, dhcp / bind mismatch on f15 to f16 upgrade: no network 17:37:56 #info builds are available, need more testing and karma 17:38:19 I don't think that there is much else to say about this one - it doesn't look like getting it finished will be a problem at this point 17:38:43 #topic (739253) - unable to shut down from gdm greeter 17:38:49 #link https://bugzilla.redhat.com/show_bug.cgi?id=739253 17:38:51 Bug 739253: unspecified, unspecified, ---, rstrode, MODIFIED, unable to shut down from gdm greeter 17:38:55 sorry folks, was in the shower 17:39:06 adamw: about time you showed up :) 17:39:36 I think that this one falls into the same category as the related 736793 17:40:04 in that we have a choice between shipping with this known issue or pulling in the whole 3.1.92 gnome update 17:40:19 yup. 17:40:44 since 736793 was rejected, I'm thinking move to final on this one 17:41:01 unless a fix to remove the power options from gdm greeter would be easy 17:41:18 any other thoughts? 17:41:24 * nirik is +1 to move to final. 17:41:36 ack 17:41:40 well 17:41:49 we can make it nth and pull in 3.1.92 to fix this 17:42:04 aiui right now, pulling gdm 3.1.92 fixes the pm options when they show up (i.e., fallback mode) 17:42:17 This one is a bit worse than the otehr one, since people may think all shutdown buttons are broken rather than just the gdm button. 17:42:22 but i may be missing stuff. 17:42:30 proposed #agreed - 739253 - This bug is a blocker but the only available fix involves pulling in gnome-shell 3.1.92 and that seems to be a bit large to be re-testing this late 17:42:42 nack 17:43:01 I suppose that a resolution would have been nice on that 17:43:02 the specific bug in 739253 is fixable with gdm 3.1.92, i think. 17:43:04 #chair adamw 17:43:04 Current chairs: adamw tflink 17:43:15 I still think we would be OK. But if the related bug is masking this one now, it probably is moot. 17:43:20 can we pull in gdm without all the rest? 17:43:23 yes. 17:43:33 the only problem is that it doesn't fix the Shell case, so the bug report has gotten confused. 17:43:47 but I *think* you only see PM options in fallback mode, with gdm 3.1.90 they didn't work, with gdm 3.1.92 they do. 17:43:51 ah, so pulling in new gdm will fix for everything using gdm other than shell 17:43:53 it could stand a bit more data and testing, though. 17:43:58 and for shell, the PM options would just disappear 17:44:02 that's my current read on it, yes. 17:44:12 I'd be OK with that fix 17:44:17 i don't think what gdm version we have makes any difference to whether you see PM options in the shell or not/ 17:44:36 unless there are unresolved dependencies from pulling in gdm only 17:45:10 don't think so, i built a live image that way with no problems. 17:45:41 #info this particular bug could be fixed by pulling in gdm-3.1.92 only without the rest of that update 17:46:03 #info gnome-shell still wouldn't have PM options @ gdm greeter, but what is there would work 17:46:51 i'd like to be more sure about the above, but i think that's the case. 17:46:54 proposed #agreed - 739253 - pulling in gdm-3.1.92 alone sounds like a reasonable solution to fix this particular bug even though it won't touch 736793 17:47:14 ack 17:47:15 proposed #agreed - 739253 - pulling in gdm-3.1.92 alone sounds like a reasonable solution to fix this particular bug even though it won't touch 736793. This needs to be tested more before pulling it in to beta release, though 17:47:25 any other ack/nak/patch? 17:47:31 but I thought it requires more of the newer stack? 17:48:02 adamw: new gdm doesnt pull in more of 3.1.92/ 17:48:03 ? 17:48:16 yeah, I would have thought so, too 17:48:46 #info pulling in gdm-3.1.92 needs more testing to verify that it can be pulled in alone 17:49:02 tflink: no, not afaict. 17:49:14 proposed #agreed - 739253 - Assuming that it can be pulled in by itself, pulling in gdm-3.1.92 alone sounds like a reasonable solution to fix this particular bug even though it won't touch 736793. This needs to be tested more before pulling it in to beta release, though 17:49:15 huh. ok 17:49:21 it only has three gnome deps, and only one of them is versioned - gnome-settings-daemon >= 2.21.92 17:50:07 ack I guess. 17:50:09 +1 17:50:14 nirik: are you OK with proposed? 17:50:14 ack 17:50:29 sure 17:50:40 #agreed - 739253 - Assuming that it can be pulled in by itself, pulling in gdm-3.1.92 alone sounds like a reasonable solution to fix this particular bug even though it won't touch 736793. This needs to be tested more before pulling it in to beta release, though 17:51:00 another one that should be quick 17:51:11 #topic (738735) repoclosure failure for 16-Beta.RC1 DVD 17:51:17 #link https://bugzilla.redhat.com/show_bug.cgi?id=738735 17:51:18 Bug 738735: unspecified, unspecified, ---, rpandit, NEW, repoclosure failure for 16-Beta.RC1 DVD 17:52:04 so the build that we think is causing problems was just pushed to stable 17:52:05 not much to say about this one, we just need digikam update karma'ed and pushed 17:52:27 #link https://admin.fedoraproject.org/updates/FEDORA-2011-13116 17:52:42 #info digikam was recently pushed to stable 17:52:51 #info this bug should be fixed @ next compose 17:53:01 anything else? 17:53:21 * tflink assumes not 17:53:25 nope, just confirm it's fixed in RC2 and close 17:53:54 #agreed - 738735 - Need to confirm fixed in RC2 and close 17:54:04 #topic (737731) Bootloader is left in F15 configuration when preupgrading to F16 17:54:14 #link https://bugzilla.redhat.com/show_bug.cgi?id=737731 17:54:16 Bug 737731: high, unspecified, ---, hughsient, ASSIGNED, Bootloader is left in F15 configuration when preupgrading to F16 17:54:31 not sure if there has been much progress on this lately 17:54:46 #info fix for this will be in F15's preupgrade 17:55:21 it doesn't block image compose, is the good thing 17:55:26 i can never seem to find hughsie, is the bad thing 17:55:43 #info since fix for 737731 is in F15, it isn't blocking compose 17:55:49 this should be relatively easy to patch, except a variable that'd probably be needed for the necessary logic doesn't currently seem to exist in firstboot (see my comment) 17:56:36 #info proposed solution in https://bugzilla.redhat.com/show_bug.cgi?id=737731#c13 17:56:37 Bug 737731: high, unspecified, ---, hughsient, ASSIGNED, Bootloader is left in F15 configuration when preupgrading to F16 17:57:01 it sounds like we "just" need to find the right person to fix this 17:58:30 any other thoughts? 17:58:40 that's about it. 17:58:57 alrighty, on to what I think is the last of the accepted blockers 17:59:09 #topic (735866) boot hangs with udevadm settle - timeout of 120 seconds 17:59:15 #link https://bugzilla.redhat.com/show_bug.cgi?id=735866 17:59:16 Bug 735866: unspecified, unspecified, ---, harald, ASSIGNED, boot hangs with udevadm settle - timeout of 120 seconds 18:00:10 it sounds like we still haven't quite pinned this down yet 18:00:29 but on the bright side, it also sounds like there aren't too many users hitting it, either 18:00:38 yeah, i'm getting sour on this one 18:00:48 it doesn't feel like anyone cares about it much, as no-one who's hitting it is providing any info 18:00:50 and if they do hit it, a reboot will often fix the problem 18:01:08 yeah, i think we can just say 'workaround == reboot' 18:01:31 * tflink contemplates between -1 blocker or move to final 18:01:47 maybe NTH? 18:02:16 -1 blocker from me 18:02:37 jsmith: beta and final, I assume? 18:02:37 If folks are able to reproduce it and can provide info, I don't mind it being on the final list 18:03:00 mmf, i think it's either beta or nothin' 18:03:02 * tflink is thinking NTH and re-propose if it becomes a bigger issue 18:03:06 yeah, i'm with tflink 18:03:10 That works 18:03:18 we should at least get more data with the beta release 18:04:11 proposed #agreed - 735866 - RejectedBlocker AcceptedNTH - This has turned out to be a little less common and severe than we originally thought. If it turns out to be a bigger issue, re-propose as blocker 18:04:26 ack/nak/patch? 18:04:29 +1 18:04:32 ack 18:04:34 ack 18:04:41 #agreed - 735866 - RejectedBlocker AcceptedNTH - This has turned out to be a little less common and severe than we originally thought. If it turns out to be a bigger issue, re-propose as blocker 18:04:54 ok, I think that's it 18:05:00 is there anything that I missed? 18:05:10 not off the top of my head 18:05:21 #topic open floor 18:05:27 note there are two updates images to test for the bootloader bug now 18:05:31 we should really get on that as a priority 18:05:41 yep, I noted that when the bug came out 18:05:56 if everyone can test their reproducers with both updates images and report back we should have good data to go ahead with the anaconda build 18:05:58 but I think that an email to test@ explaining why there are 2 images and what we're looking for in testing would help 18:06:01 yup 18:06:12 adamw: you want to send that email out or should I? 18:07:14 FYI, I didn't get the digikam update signed right, so thats why branched didn't show up today. 18:07:24 it should show up tomorrow with the ones pushed to stable. 18:07:26 #action tflink or adamw send out request for testing BOTH of the updates for 738964, explaining why there are 2 updates and what we're looking for in testing and data for feedback 18:07:37 nirik: ok, thanks for the update 18:08:25 #info digikam update for 738735 had issues with pushing for stable, it has been fixed and should show up with the next push to stable 18:08:45 unless there are any other topics ... #endmeeting in 2 minutes 18:09:31 #info next blocker meeting (if needed) will be 2011-09-30 @ 17:00 UTC 18:09:46 * tflink hopes that it won't be needed 18:10:22 OK, thanks for coming everyone! 18:10:39 Time for some testing and getting TC3/RC2 out today! 18:10:51 * tflink will send out minutes shortly 18:10:51 * jsmith just finished a successful installation with the .6.img anaconda update, fwiw 18:10:53 #endmeeting