17:01:51 #startmeeting 17:01:51 Meeting started Mon Aug 15 17:01:51 2011 UTC. The chair is bmwiedemann. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:51 Useful Commands: #action #agreed #help #info #idea #link #topic. 17:02:10 Hello everyone 17:02:10 As always, today's agenda is on http://en.opensuse.org/openSUSE:Testing_meeting 17:02:10 You can still add to the agenda on the wiki while the meeting is in progress 17:02:35 #topic MS4 experience - if available 17:03:19 actually, there is currently no real MS4 available 17:04:30 I came as close as possible - I think. 17:04:35 so we could at most discuss the current state of factory and if/when that is suitable to snapshot for Open-Bugs-Day use 17:05:11 OK. 17:05:49 openQA results show problems with all NET installs segfaulting in yast2 during second stage autoconfiguration. not so nice. but DVDs work. 17:06:21 and the latest KDE/KWin still has some problems 17:07:26 hi 17:07:58 I installed KDE Live CD Build 0186 and zypper dup'd from factory up to today. It seems OK, but not much testing. 17:08:07 Holgi: Hi 17:08:35 I'm using a VB VM. 17:09:24 At a minimum, the NetworkManager applet works, even with wireless. 17:10:02 btw: I added code to openQA to test sound - dheidler made that concept+code and I finally fixed+merged it. 17:10:11 It has some policy problems and asks for a password as often as Windows 7, but if functions. 17:11:51 I think, the policykit defaults need some review. there were funny popups on entering the root PW to query background brightness 17:11:58 hi everybody 17:12:41 I like that bughunter nickname. 17:12:59 LWFinger: I know that problem from 3G connections with 11.3 17:13:16 LWFinger: get asked for PIN & root-pw 17:13:43 LWFinger: several times - is it like this? 17:13:45 I too, its me ;) @ LWFinger 17:14:07 Holgi: I don't use 3G stuff as I have no hardware. My testing is with normal 802.11a/b/g/n. 17:14:25 LWFinger: ok, but is it the same behavior? 17:15:24 No. It keeps asking for the root password as "policy prevents user action". 17:15:36 LWFinger: ok 17:16:00 LWFinger: my problem cause is also policy kit 17:16:34 I still need to put together the info for a proper bug report for my problem. Have you filed one? 17:17:30 LWFinger: no mine is for 11.3 17:17:46 Is it OK in 11.4? 17:18:06 sorry, I ment 11.4 17:18:50 For 802.11 stuff, the policies are OK in 11.4. It just broke in 12.1. 17:19:09 LWFinger: ok 17:22:19 so do we have any opinion if current factory is suitable for testing or what needs fixing urgently? 17:22:48 maybe we should delay Open-Bugs-Day to happen after a real milestone release? 17:22:48 I think it is OK. 17:23:59 bmwiedemann: what prevents coolo from releasing M4? 17:24:29 absense :D 17:24:40 Having a real release would make it easier for the non-techies as you said before the meeting started. 17:26:03 the alternative would be to target more techy people (who know how to unbreak their system) 17:26:14 I agree - we should have a release 17:27:24 Delaying for M5 would put our two OBD's within 3 weeks of one another. Is that wise? 17:27:58 no, not really 17:28:24 no 17:28:39 Should we combine them? They have quite different purposes. 17:29:48 bad thing is 17:30:09 next OBD is already announced, right? 17:30:33 Yes, but it could be delayed or cancelled. 17:31:03 of course, I could also setup a number of prepared VMs again, so that people have an easier time 17:31:09 what do the others think? 17:31:29 I would do it 17:31:45 bmwiedemann: using a factory build? 17:31:59 Holgi: yes. 17:32:05 ok 17:33:02 #topic Open-Bugs-Day 17:33:32 I would really like to see what 11.4 bugs are still present. But to get a consistent base, we do need to supply detailed instructions. 17:35:03 I can snapshot factory to my own server that has sufficient space+bandwidth. and maybe even have prepared VM-images for download so that different people can try the very same software 17:36:15 That sounds good. Would the prepared images include those for VirtualBox? 17:36:51 it can run .vmdk files from VMware, right? and KVM can do that, too 17:37:16 I don't know, but I can test that. 17:38:09 # 17:39:32 are there a lot of gnome/kde specific bugs or could I just prepare one image? 17:39:33 should we focus on s specific topic? 17:40:19 we can focus on kde now and next time on gnome? 17:40:23 bmwiedemann: I suspect so, and we should have both. 17:41:00 OK 17:42:52 if I read numbers correctly, there are around 1500 bugs for 11.4 with 1086 in NEW state, 160 NEEDINFO and 78 REOPENED 17:43:53 appears like a rather big number, so some focus could indeed be helpful 17:44:20 I have a total of 1324 ( incl the reopen) 17:44:20 agree 17:45:04 How do we decide which are important? 17:45:30 Severity + Priority 17:46:22 or State NEEDINFO? 17:46:23 That assumes those are set correctly. If we cut off at some level, how many are left? 17:46:59 around half of the bugs are marked Critical or Major 17:47:18 That helps a bit. 17:47:26 but priority is set to P5 on quite many 17:48:11 The poster frequently overstates the importance. It is affecting THEIR system. 17:48:38 then priority is not very useful 17:49:17 actually, Priority is intended to be set by the developer or the project leads (coolo,AJ) 17:49:37 If it was posted as Critical or Major, and it has not been downgraded, then we should look at it. 17:49:46 and P5 = None is just the default value 17:51:16 Actually, does it matter how many bugs are in the pool? The tool is only going to present a few at a time. 17:51:55 as long as it presents the more important ones earlier, it does not make much difference 17:52:21 only that it can feel better to solve 400 of 500... instead of 400 of 1400 17:52:44 doing state=NEEDINFO first and than reopen, new etc? 17:53:34 I would skip the ASSIGNED ones as someone could be working on it 17:53:47 agree 17:57:16 so I would pull e.g. 1325 bugs from https://bugzilla.novell.com/buglist.cgi?action=wrap&bug_file_loc=&bug_file_loc_type=allwordssubstr&bug_id=&bugidtype=include&chfieldfrom=&chfieldto=Now&chfieldvalue=&classification=openSUSE&deadlinefrom=&deadlineto=&email1=&email2=&emailassigned_to1=1&emailassigned_to2=1&emailcc2=1&emailqa_contact2=1&emailreporter2=1&emailtype1=substring&emailtype2=substring&field0-0-0=noop&keywords=&keywords 17:57:53 sorted by severity 17:58:29 that should work 17:58:59 still a lot 17:59:44 Holgi: if we find that a bug is still in Factory - what do we do with it? 18:00:11 just adding a comment would not make it searchable 18:00:21 bmwiedemann: I would change product to latest 18:00:30 bmwiedemann: but you can clone it 18:00:46 bmwiedemann: and just add comment: "still in version 12.1" 18:01:08 12.1-MS4? 18:01:18 bmwiedemann: yes 18:01:42 bmwiedemann: but I'm not sure about "official" way 18:02:16 Holgi: then who would? 18:02:22 bmwiedemann: most important: increase priority: description - also in new product 18:02:38 how to clone @Holgi? 18:03:02 bmwiedemann: who would know? dokomentation 18:03:29 there is a "Clone this bug" link on bottom right 18:04:32 this is very, very small ;) 18:05:23 I think cloning is better tahn putting some text into the original 18:05:42 Agreed. 18:05:59 in any case: increase prio 18:06:23 ok, should I put this on the wiki? 18:06:48 probably not since I don't know the official way 18:06:56 ok 18:09:27 bmwiedemann: As soon as possible, I would like to have a download location for a virtual image. For publicity, I will also need the URL for the snapshots. 18:09:42 OK 18:10:23 You should probably Email those as we may not want that info in the public record now. 18:11:24 OK 18:14:09 anything else for today? 18:15:36 #action bmwiedemann will prepare VM images for OBD and mail the location to LWFinger 18:16:16 #topic open discussion 18:16:49 I will prepare the revised announcement message for OBD and send to the testing ML for comments. 18:17:55 any plans for OSC? @Holgi 18:18:51 there will be a testing BoF 18:19:07 jradzuweit: I will be there 18:19:16 jradzuweit: but don't have a talk 18:19:28 I will be there, too 18:19:42 jradzuweit: so it's again the three of us? 18:19:59 yes 18:20:32 ok 18:20:55 bmwiedemann: submitted a talk about openqa 18:21:13 bmwiedemann submitted a talk about openqa 18:21:31 titled "one year of openQA" 18:22:04 when? 18:22:39 some input of you would be nice. Did it matter for you? what most? 18:23:04 I think, there is no schedule, yet. 18:23:17 non seen yet 18:23:27 bmwiedemann: I guess your best customer: release manager 18:24:17 sure. but openQA.opensuse.org is also for the community. 18:24:57 bmwiedemann: sure 18:25:17 bmwiedemann: for me it's always a good source to check if I should install a build or not 18:25:29 I do the same. 18:25:59 me too 18:26:29 It is also a quick check on what is failing to build. It may not say why, but it gives an overview. 18:29:21 nice :) 18:31:24 Has anybody info about this Beta thing? 18:31:39 means MS6=BETA? 18:31:57 Coolo has agreed to do it. 18:32:07 yes. coolo renamed it on his roadmap 18:32:25 :), so weget lots lots of new testers than? 18:32:39 and I updated the wiki accordingly 18:32:45 * iznogood thinks that was a pointless change, but we'll see if it changes the number of testers 18:33:17 The main problem will be documenting what happened. 18:33:44 I think we should be happy about our request was successful! 18:34:47 As I have never met coolo, I probably don't quite unerstand that comment, but I can guess. 18:34:48 yes 18:37:16 I guess, we are done for today then 18:37:35 I think so. Next meeting? 18:39:18 probably around MS5 (if we do not count Open-Bugs-Day) 18:39:58 on OSC? 18:40:08 would be 12.9. 18:40:31 Sep. 5 would be the likely date if MS5 is on time, but that is a holiday in the US. I might not be available. 18:41:16 we can still use different days than Mondays 18:41:23 The 12th is better for me. 18:41:44 Other days are OK, too. 18:42:33 I can arrange for other days too 18:44:45 so 2011-09-12 17:00 UTC? 18:44:51 I would stick with Mondays, everybody of us is available 18:45:07 agree 18:45:32 OK. 18:46:17 very well. then thank you for attending our meeting today 18:46:32 good night 18:46:34 I will add links to the logs on http://en.opensuse.org/openSUSE:Testing_meeting 18:46:41 #endmeeting