17:02:30 #startmeeting f19alpha-blocker-review-1 17:02:30 Meeting started Wed Mar 13 17:02:30 2013 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:02:30 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:02:30 #meetingname f19alpha-blocker-review-1 17:02:30 The meeting name has been set to 'f19alpha-blocker-review-1' 17:02:30 #topic Roll Call 17:02:43 * kparal is here for 20 minutes 17:02:44 * satellit_e listening 17:02:51 Who's ready for some blocker review fun? 17:03:55 * jreznik is here for that few months of blocker bugs review meetings fun 17:04:21 hopefully it'll be less painful than 18 was 17:05:12 yep 17:06:21 * tflink waits another minute or so for others to join 17:08:43 I kind of wish we had a few more people but I guess let's get started 17:08:58 #topic Introduction 17:09:06 Why are we here? 17:09:08 #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. 17:09:16 #info We'll be following the process outlined at: 17:09:16 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 17:09:22 #info The bugs up for review today are available at: 17:09:22 #link http://qa.fedoraproject.org/blockerbugs/current 17:09:28 #info The criteria for release blocking bugs can be found at: 17:09:28 #link https://fedoraproject.org/wiki/Fedora_19_Alpha_Release_Criteria 17:09:53 #info Up for review today, we have: 17:09:59 #info 2 Proposed Blockers 17:09:59 #info 0 Accepted Blockers 17:09:59 #info 1 Proposed Freeze Exceptions 17:09:59 #info 0 Accepted Freeze Exceptions 17:10:23 unless there are objections, I'll get started with the proposed blockers 17:10:48 #topic (919743) gdm unusable on old video card 17:10:48 #link https://bugzilla.redhat.com/show_bug.cgi?id=919743 17:10:48 #info Proposed Blocker, gdm, NEW 17:11:41 hrm, it would be nice to know how many cards are affected by this 17:11:41 I see the fail whale even in VMs. so I don't think this is related to a particular hw 17:11:53 ah, nvm then 17:12:21 is firstboot affected? 17:12:38 so we need to fix the general issue first and then have a look at older hw 17:12:45 I never saw any firstboot on F19 17:13:45 isn't there some big change for firstboot planned for F19? 17:13:58 tflink: initial-setup is the new name 17:14:09 using a gnome tool for gnome installs and firstboot for everything else? 17:14:14 hrm, 17:14:18 last time I talked to marsik, it should be ready now with working systemd stuff around 17:14:29 hrm, the criteria need to be updated then 17:14:32 tflink: no info about the new gnome one 17:14:38 msivak said it will be ready after a new anaconda build is ready 17:14:41 which currently isn't 17:14:46 ah 17:14:59 ok, we can deal with that when it changes, then 17:15:05 but firstboot is now called initial setup (not talking about gnome initial experience) 17:15:07 thoughts on blocker? 17:15:28 (so quite confusing) 17:15:31 #info for F19, firstboot is now called initial setup 17:16:07 kparal: do you know if there is a bug for the VM issue you mentioned? 17:17:53 * nirik is sorta here. ping if I can help. 17:18:13 unfortunately I don't know 17:18:42 but punt this bug now, it seems very possible this will turn out to be a non-issue 17:19:13 how so? 17:19:32 because I think this is really not related to hardware, but generally broken 17:20:30 if it's generally broken, it's actually worst 17:20:40 it's easier to fix :) 17:21:01 I'm not sure I follow you on how that leads to a punt but it wouldn't hurt to have more information 17:21:15 well, more likely to be fixed, anyways 17:21:44 depends how you look at it. I think we can punt this bug and find the general fail whale bug and put a blocker status on that one 17:21:45 hrm, the topic wasn't changed 17:22:10 but other approaches are possible 17:22:15 I'm sorry, I need to leave now 17:22:30 kparal: ok - sorry about the time mixup 17:22:46 np 17:23:36 proposed #agreed 919743 - We suspect that this is a more general problem not limited to older graphics hardware - will revisit at the next meeting after more triage work has been done 17:23:41 the second bug is probably connected to this general breakage too 17:23:45 ack/nak/patch? 17:23:50 ack 17:26:27 sigh, not sure we have enough people for quorum anymore 17:26:34 * nirik looks up 17:26:36 we were pushing it with 3 of us 17:26:46 ack 17:26:52 ack 17:27:03 #agreed 919743 - We suspect that this is a more general problem not limited to older graphics hardware - will revisit at the next meeting after more triage work has been done 17:27:20 #topic (909473) Fail whale page appears after logging in 17:27:20 #link https://bugzilla.redhat.com/show_bug.cgi?id=909473 17:27:20 #info Proposed Blocker, gnome-shell, NEW 17:28:20 * nirik waits for bugzilla 17:28:33 sorry i'm late 17:29:01 adamw: no worries 17:29:14 hi adamw! 17:29:17 * nirik thinks the report needs more info... gdm messages output, X output, etc. 17:29:40 and seems like it's a similar generic issue as with the previous one 17:29:41 also, if kdm works, is this a gnome-shell bug? 17:29:43 yeah, I also suspect that this and 919743 are related 17:30:05 does kdm require llvmpipe when no hw 3d is present? 17:30:37 this could be related to the blacklisting I guess.. 17:31:18 kdm shouldn't 17:31:20 proposed #agreed 909473 - This needs more triage and logs before we make a decision on blocker status but we suspect that it could be related to 919743 - will revisit at the next review meeting 17:31:29 tflink: kdm does not need acceleration 17:31:53 adamw: kparal mentioned that he's been seeing something similar with VMs, so it might be llvmpipe 17:32:08 llvmpipe is completely blacklisted in kwin, upstream does not want it ever enabled :) 17:32:17 yep, could be llvmpipe 17:33:08 either way, I don't think we have quite enough information to work off of right now 17:33:16 same as the previous one 17:33:21 blacklisted cards don't fall back to llvmpipe, that's a longstanding issue. i don't think they broke at gdm before, though. 17:33:23 ack/nak/patch to the proposal? 17:33:43 proposal? 17:33:48 ah, I see 17:33:51 sorry, blind today 17:33:55 ack 17:33:58 ack 17:34:08 didn't fallback mode get removed for gnome 3.8? 17:34:20 or was that just discussion? 17:34:42 #agreed 909473 - This needs more triage and logs before we make a decision on blocker status but we suspect that it could be related to 919743 - will revisit at the next review meeting 17:34:57 and that would be all of the blockers for today :) 17:35:03 yeah...non-3D fallback got removed for classic... 17:35:05 that may be it. 17:35:34 yeah, can we just make 909473 a blocker? now i wake up i think i see what's going on 17:35:58 adamw: oh? 17:36:02 #undo 17:36:02 Removing item from minutes: 17:36:05 we need blacklisted cards to fallback to llvmpipe now. 17:36:15 or else gnome just isn't going to work, as there's no unaccelerated session for them. 17:36:34 it's clearly a blocker, only question is whether it's a dupe 17:36:46 are there enough people using blacklisted cards to justify taking it as a blocker? 17:36:57 if gdm needs llvmpipe, then it's dupe I'd say 17:37:36 jreznik: it's not gdm 17:37:38 it's gnome 17:37:49 well, may be gdm _too_, but gnome is the simplest case. 17:37:52 one is gnome, one was gdm 17:37:55 ah 17:38:02 more or less the same bug then 17:38:14 what was the other bug so i can see it?> 17:38:29 proposed #agreed 909473 - AcceptedBlocker - With gnome 3.8, the non-3d-accelerated option is being removed and blacklisted cards don't fallback to llvmpipe making them unusable with gnome 17:38:30 adamw: 919743 17:38:31 tflink: for alpha, hum, maybe not. we don't need info, though. 17:39:21 yeah, they're pretty much dupes. 17:39:35 whether they're blocker for alpha depends how much we care about blacklisted hardware. i think maybe beta. 17:39:54 more beta but earlier fixed, better 17:40:14 alpha fe, then? 17:40:36 * tflink is going to type nth several times before that change really sinks in 17:41:19 * adamw gets out the training whip 17:41:38 fe would be reasonable as it's unlikely to make stuff worse and is a config rather than code change 17:42:28 proposed #agreed 909473 - RejectedBlocker, AcceptedFreezeException - With gnome 3.8, the non-3d-accelerated option is being removed and blacklisted cards don't fallback to llvmpipe making them unusable with gnome. While not quite an alpha issue due to the limited amount of hw affected, a tested fix would be considered after F19 alpha freeze 17:44:03 ack 17:44:09 ack 17:44:21 i'll double check all that with ajax and bring it up again if i'm wrong 17:44:28 ok, thanks 17:44:30 #agreed 909473 - RejectedBlocker, AcceptedFreezeException - With gnome 3.8, the non-3d-accelerated option is being removed and blacklisted cards don't fallback to llvmpipe making them unusable with gnome. While not quite an alpha issue due to the limited amount of hw affected, a tested fix would be considered after F19 alpha freeze 17:44:37 did someone secretarialize already? or should i go back over the notes and do it? 17:45:03 adamw: I don't think so - I was going to do it after the meeting if it didn't get done 17:45:08 OK, 17:45:09 okay, i'll get it 17:45:23 OK, that's all of the proposed blockers for today 17:45:25 on 17:45:35 on to the proposed FE! 17:46:00 and remembering where the return key is so I don't keep hitting it by accident :-\ 17:46:12 #topic (894110) The "standard" comps group could use a bit of spring cleaning 17:46:15 #link https://bugzilla.redhat.com/show_bug.cgi?id=894110 17:46:17 #info Proposed Freeze Exceptions, comps, NEW 17:46:44 I agree with adamw here, -1 FE - the FE process is not a todo list 17:47:02 it's abuse of process, -1 17:47:41 proposed #agreed 894110 - RejectedFreezeException - This bug does not meet the requirements of a FE and should be tracked like a normal bug. 17:47:52 ack/nak/patch? 17:48:24 ack 17:50:07 #agreed 894110 - RejectedFreezeException - This bug does not meet the requirements of a FE and should be tracked like a normal bug. 17:50:09 ack 17:50:23 #topic Open Floor 17:50:43 * tflink wishes all blocker review meetings were this short 17:50:58 anything else that should be brought up? 17:51:25 I assume that we'll start seeing more blocker/fe proposals once we have more images to test 17:51:36 * nirik nods. 17:52:00 * tflink needs to get set up for building smoke images for F19 17:52:09 should be x86_64 images at least in branched today (whenever it manages to copy) 17:52:49 yep, smoke is what we want now 17:53:02 we would probably need that anaconda build kparal mentioned with inital setup 17:53:29 +1 for testable f19 .isos 17:53:35 yeah, it sounds like we're waiting on that and for the F19 sync (unless that's done now) 17:54:23 it's not. actually branched also got restarted... 17:54:29 do both rawhide and branched will land later today 17:54:35 s/do/so/ 17:54:45 ok, sounds good 17:55:12 nirik: are you talking about the boot.iso that's in the tree or the nightly builds? 17:55:22 the boot.iso in the tree. 17:55:31 ok, cool 17:56:26 if there's nothing else ... 17:56:48 * tflink sets the fuse for [0,inf) minutes 17:57:45 thanks for coming, everyone! 17:57:53 * tflink will send out minutes shortly 17:57:57 #endmeeting