20:17:29 #startmeeting Spins SIG Meeting 2011-04-04 20:17:29 Meeting started Mon Apr 4 20:17:29 2011 UTC. The chair is cwickert. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:17:29 Useful Commands: #action #agreed #halp #info #idea #link #topic. 20:17:52 #chairs brunowolff pbrobinson_ nirik` 20:17:56 #chair brunowolff pbrobinson_ nirik` 20:17:56 Current chairs: brunowolff cwickert nirik` pbrobinson_ 20:18:22 ok, not much on the agenda really 20:18:37 so let's revisit the action items from last week 20:18:46 * cwickert digs .... 20:19:24 ok, number 1: 20:19:25 cwickert to announce the %include change 20:19:27 I failed 20:19:38 I forgot to do it, I first want to do some testing 20:19:53 because I didn't take nested stuff into account 20:20:09 I'm currently very busy with my dayjob 20:20:10 I need to spend some time in the next couple of days to verify and fix any issues with the SoaS spin. Other than that I don't think I have any points 20:20:11 We have beta freeze coming up tomorrow, so it should either happen right away or wait a couple of weeks. 20:20:39 brunowolff: ok, I'll try to do it now then because I want to fix as many spins as possible 20:21:18 damn, soo much to do... 20:21:32 The freeze really starts at the end of the day tomorrow. 20:22:07 ok, I'll try my best 20:22:14 next action item: 20:22:16 The spin-kickstarts build I made will be eligible to be pushed to stable pretty late tomorrow, but I am hoping to get it in. 20:22:18 I can hopefully get some testing done by late tomorrow 20:22:25 brunowolff will try if the fix proposed by Lennart (ln -sf /dev/null /etc/systemd/system/hwclock-save.service) can go into /etc/rc.d/init.d/livesys 20:22:30 and all that will need to change is the .ks for soas 20:22:40 #topic Action items from previous meetings 20:22:49 There was one a couple of weeks ago, so it won't be too out of date even if it doesn't make it. 20:22:54 brunowolff: any update on hwclock? 20:23:11 I made the changes, but only did smoke testing. 20:23:38 It worked OK in a case where I didn't change the clock setting while running the live image. 20:23:45 did you commit them? 20:23:50 The reporter hasn't given me any feedback. 20:24:00 Yes, the changes were commited. 20:24:10 ah, just found it 20:24:53 #info more testing on hwclock needed 20:25:04 brunowolff: you have the bug # at hand? 20:25:48 297421 20:26:05 It's an ancient bug, that muted a bit. 20:26:15 mutated 20:26:28 .bug 297421 20:26:31 cwickert: Bug 297421 LiveCD handles timezones inconsistently - https://bugzilla.redhat.com/show_bug.cgi?id=297421 20:26:42 #link https://bugzilla.redhat.com/show_bug.cgi?id=297421 20:26:51 The not updating the hwclock stuff was already there, but was broken by the switch to systemd. 20:27:09 ok, lets all test this when we are testing our spins, shall we? 20:27:12 The other timezone related questions aren't really for us. 20:27:17 +1 20:27:42 don't mess with timezones, we are just having a horrible discussion about this in kolab 20:28:12 ok, next action item: cwickert to ask probinson if he wants prelink in fedora-live-mini.ks 20:28:25 done, thanks pbrobinson_ for the fast response 20:28:29 this issue is solved 20:28:31 next 20:28:38 nirik to add prelink to %post of fedora-live-base.ks 20:28:42 this also happened 20:28:47 brunowolff fixed it 20:29:06 ok, that's all for last week's action items 20:29:23 There was still some discussion about the -R option, but I don't see a rush to worry about that right now. 20:29:33 The basic support should still be there. 20:29:47 New item - https://fedorahosted.org/fedora-infrastructure/ticket/2699 20:29:51 what does -R do? 20:30:00 -R randomize. 20:30:19 It's a security feature that doesn't help when the image is fixed and public. 20:30:29 At least not enough to worry about. 20:31:03 2699 is about an issue preventing downloading of the games spin. 20:31:21 I suspect only spins actually over 2 GB are affected. 20:31:39 Answered on list I think, and dealt with I believe :-) 20:32:04 brunowolff: confirmed 20:32:07 thanks to cwickert and kevin 20:32:50 brunowolff: I don't think we need to worry about randomization then 20:33:24 We can add it later after more discussion. It isn't related to performance, which is the reason for the change. 20:33:36 #info there is a problem downloading the Games Spin from nightly composes at http://alt.fedoraproject.org/pub/alt/nightly-composes/ 20:33:47 #link https://fedorahosted.org/fedora-infrastructure/ticket/2699 20:34:12 ok, anything we can do about the 2 GB thing? 20:34:20 s/thing/problem 20:34:43 Kevin is travelling this week, so he may not get a lot of time to look at it. 20:34:55 sorry not aware of the problem enough to comment 20:35:09 I would expect that otherwise he would be in the best position to fix it. 20:35:21 It looks like a type issue. 20:35:22 * cwickert thinks that we cannot do much about it and Kevin is the only one who can 20:35:58 Games spin doesn't block the release and the problem should be fixed by beta, so I am not too worried about it. 20:36:04 but I know there's issues with some downloads of files over 2Gb in some cases (in general) 20:36:41 well, but this looks like a koji issue, not a download thing 20:36:52 I wanted it to do some testing of game specific problems to see if they were video card specific. 20:37:01 either kevin or dgilmore or other infra people 20:37:06 I'll just end up making a local copy to do this instead. 20:37:24 brunowolff: I suggest to just make the games spin just work (tm) so we don't need no testing :) 20:38:09 I was really interested in whether I broken vegastrike or if it the graphics guys did. 20:38:40 In the latter case it might work on some less than ancient video cards. 20:38:45 just blame the others... 20:39:12 ok, anything more we need to discuss now? 20:39:23 #topic Open Floor 20:40:08 nothing from me (that I can think of in my jetlagged state) 20:40:33 We should send out a reminder late tomorrow stating that we are in a freeze. 20:40:37 I'll just send around one more ping to all spin maintainers 20:40:50 yupp, that's what I was just about to do 20:40:52 A followup to the reminder I sent last week would probably work. 20:42:21 I hope so 20:42:34 anything more? 20:42:42 otherwise I will just on the kickstarts now 20:42:44 That's all I can think of. 20:42:58 ok, thanks evrybody for coming 20:43:01 crap ... I missed it 20:43:05 maxamillion: haha, you are late :) 20:43:08 sorry, I was afk dealing with $dayjob stuffs 20:43:11 you didn't miss much 20:43:20 ah, ok 20:43:23 I'll read the minutes 20:43:24 maxamillion: out Xfce spin is in good shape 20:43:29 yey! 20:43:29 s/out/our 20:43:30 We are technically still in open floor. 20:43:34 * maxamillion does a little dance 20:43:49 well, one thing that really sucks is that Xfce is bigger than GNOME now 20:44:11 this is because a whole lot of dependencies was added through gnome-shell 20:44:18 but we cannot do much about it 20:44:30 at least not at this state one day before beta freeze 20:44:34 rgr 20:44:44 I thought the agreement was that they weren't supposed to do that. 20:44:45 * pbrobinson_ suspects he'll find that tomorrow when he deals with the soas spin issues :( 20:45:00 it's really a pitty, we did our best 20:45:03 But right, now isn't the time to change things. But hopefully after beta. 20:45:08 pbrobinson_: yeah, have fun ;) 20:45:29 I filed some bugs in the depchain tracker bugs 20:45:36 but there was not much response 20:45:52 one of the problems is that many things now need gnome-control-center 20:45:59 There was a proposed solution that looked like it would get gnome-shell pulled in to updates without 20:46:11 e.g. gnome-packagekit, gnome-bluetooth... 20:46:22 forcing gnome-panel to drag in gnome-shell. 20:46:26 brunowolff: as maintainer of clutter and a lot of gnome 3 dependencies its a massive problem just getting the gnome desktop team to deal with minor things like pushing builds to rawhide first before f-15 branch, let alone something like deps 20:46:33 brunowolff: this is not the problem 20:46:46 I mean, I dislike it, too, but this is not *our* problem 20:47:11 one is the dependencies on gnome-control-center 20:47:21 which could be solved by making a -libs package 20:47:24 * pbrobinson_ feels like just blocking the ACLs on certain packages and having a standoff on fedora policy (but really doesn't have the strength for it at the moment) 20:47:43 and the other is that gnome-settings-deamon crashes if certain plugins are not there 20:47:57 for example gnome-screen-saver 20:48:11 which we don't need because we already have xscreensaver 20:49:03 #info a lot of dependencies were added to the GTK based spins through GNOME and GKT 3. we should look into the depchains once again 20:49:14 ok, not much more we can do here now, right? 20:49:23 or are there any more ideas? 20:49:23 agreed 20:50:07 * cwickert will end the meeting in 3 minuntes 20:50:15 I suspect we will need to ignore it this release but I think we need to get a commitment that it will be fixed the next round. The gnome3 team is under pressure to get the release out and make it look good, but if we give ground to help them this release they need to commit to fixing the issues in the next release 20:52:49 +1 20:52:54 ok, thanks for coming 20:52:56 #endmeeting