14:05:04 #startmeeting oVirt Weekly Meeting 14:05:04 Meeting started Wed Jul 31 14:05:04 2013 UTC. The chair is oschreib. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:05:04 Useful Commands: #action #agreed #help #info #idea #link #topic. 14:05:28 #topic agenda and roll call 14:05:34 #info 3.3 status update 14:05:53 #info infra team update 14:05:59 #info Other Topics 14:06:16 * lvernia here 14:06:17 * ofri here 14:06:20 * mgoldboi here 14:06:25 * mskrivanek here 14:06:29 here 14:06:42 * sahina here 14:08:05 #topic oVirt 3.3 status Update 14:08:12 #link http://www.ovirt.org/OVirt_3.3_release-management 14:08:19 #info 3.3 Tracker bug: 14:08:24 very nice participation on the test day 14:08:30 #link https://bugzilla.redhat.com/show_bug.cgi?id=918494 14:08:46 * jb_netapp here 14:09:05 #info test day happened last week, very nice participation 14:09:28 ok, some general stuff: 14:09:39 1. I'm going to create ovirt-engine-3.3 branch today 14:10:04 a separate email will be sent. please make sure to cherry pick important stuff. 14:10:44 2. RC build will happen next week OR afterword if we will find critical issues 14:11:11 how are we dealing with the 79 bugs with target release of 3.3? 14:11:43 itamar: lets go over team by team, add the critical bugs to the tracker, and postpone others. 14:11:44 go group by group for their comments on them (critical and should be backported to branch or can wait - and change their target release from 3.3)? 14:11:49 ok 14:12:25 #info ovirt-engine-3.3 branch will be created today 14:12:32 #info RC build will happen next week OR afterword if we will find critical issues 14:12:54 3. I expect all missing features to be removed from http://www.ovirt.org/OVirt_3.3_release-management 14:13:09 ok, now we can go team by team. 14:13:23 #info gluster status 14:13:34 #undo 14:13:34 Removing item from minutes: 14:13:46 #info open bug for 3.3 14:14:03 #link http://alturl.com/aq23x 14:14:15 now we can move to team stuff 14:14:23 #info gluster status 14:14:47 sahina: any updates about 3.3? 14:15:08 mskrivanek: gnome boxes? no waith... i read that from the ovirt web gui -> run once -> Boot option -> attach floppy..... 14:15:41 mskrivanek: ...but I do not see in the combo my driver... i can only read [sysprep] 14:15:49 oschreib, 988453 - we will be sending a patch to change default cluster to virt only 14:16:54 oschreib, Bz 988299 need to reproduce as there were no logs. hope to get to it this week 14:17:06 Ub_User: then you don't have any floppy in iso domain. upload some .vfd 14:17:31 sahina: can you please add those blockers to the 3.3 release tracker? (and make sure they're cherry picked into the branch I'll create later today) 14:17:33 oschreib, not really blockers for 3.3 14:17:38 hi, i apologize for being late, had connectivity issue. i'm representing infra for bugs discussions. 14:17:56 yzaslavs: ok, thanks 14:18:17 sahina: OK, I couldn't understand that bug sevirity 14:18:38 sahina: all features listed in http://www.ovirt.org/OVirt_3.3_release-management are in 3.3? 14:19:05 oschreib, the REST API is not in yet 14:19:24 sahina: so please remove it :) 14:19:35 sahina: no more new features in 3.3 ovirt 14:20:06 oschreib, it's posted upstream, waiting for final review. should i still remove it? 14:20:59 oschreib, i will remove the swift service feature for gluster. but gluster hooks management would like to get it in 14:22:10 sahina: well, we're supposed to build RC in one week, I'm afraid of big changes before RC 14:23:58 sahina: we can take that issue offline (or in mailing list) 14:24:02 oschreib, ok, 14:24:04 lets go forward 14:24:20 * danken is here, only now. 14:24:31 #info no blockers for gluster 14:24:41 #info integration status 14:24:53 mgoldboi: any blockers? 14:25:30 oschreib: one blocker 987939 on tracker already 14:25:37 sahina - can you please change the target version of the bugs from 3.3 to ---? 14:25:40 oschreib: release page is updated 14:26:08 oschreib: pending item to remove old installer for RC 14:26:22 oschreib: old = legacy ;) 14:26:36 sahina - for those not critical enough to be backported to ovirt 3.3 14:27:26 mgoldboi: about the blocker - estimated time for the fix? 14:28:02 itamar, yes. will do. 14:28:16 oschreib: should be ready for next week 14:28:45 * doron here stealth mode 14:29:35 mgoldboi: ok. 14:29:37 next 14:29:54 #info integration: single blocker, targeted for next week 14:29:58 #info infra 14:30:09 yzaslavs: any blockers for 3.3? 14:31:18 oschreib: no, last potential blocker was merged 10 minutes ago. 14:31:26 amuller: do you have new infomation about the odd case of supervdsm dying? 14:31:33 oschreib: that was https://bugzilla.redhat.com/show_bug.cgi?id=990524 14:31:36 this may be an infra issue 14:31:38 yzaslavs, so what are the 18 bugs with 3.3 target release? 14:32:03 itamar: i went over with bazulay on them, only the above was recognized as blocker. 14:32:22 yzaslavs - ok, please change target version of the rest to --- then 14:32:30 itamar: sure, no problem. 14:32:38 ybronhei: I do not understand it completely, but somehow deployment has restarted libvirt without restarting supervdsm later 14:32:54 which caused supervdsm to die on the first setupNetwork command. 14:33:44 danken1: if that issue is a blocker, please add it to the release tracker. 14:33:55 yzaslavs: anything else to report? 14:34:12 oschreib: I do not think we have a bug opened, yet. I'm asking ybronhei and amuller if they have news about it. 14:34:14 oschreib: nope 14:34:27 #info infra: single blocker merged, 18 bugs will move to next version 14:34:31 ok, next. 14:34:35 #info networking 14:34:46 danken1: any blockers for us? 14:34:57 so... 14:34:58 danken1: I see 13 bugs on the 3.3 list 14:35:08 I have only 11 14:35:37 btw, target release 3.4 is not defined in bugzilla ATM 14:35:54 anyway, ovirt-node registration flow does not work due to 988990 14:36:50 ok, I see it already blocks tracker. 14:37:04 oschreib: nowadays that we have ovirt-host-deploy working for ovirt-node, it may not be the end of the world 14:37:34 but it should be a project decision: are we willing to reslease 3.3 with registration flow broken. 14:37:54 danken1: well, I agree there are other registration options, but I think it's important bug IMO, which I'd like to get in 14:38:23 oschreib: it won't be in by tomorrow's deadline 14:38:23 oschreib: I'm looking at that bug right now 14:38:37 but yeah, I don't think it'll be done by tomorrow 14:38:41 danken1: any estimation? 14:39:00 Another issue that is borderline network is 988986 NEW jboggs@redhat.com libvirt network directory is not persisted 14:39:05 mskrivanek - indeed. please move to --- for now... 14:39:22 itamar: and release 3.3 without node registration? 14:40:00 oschreib: i was talking to mskrivanek... 14:40:09 not on registration bug 14:40:30 itamar: oh, missed that 14:40:44 oschreib: I suspect that it would take a week, assuming that dougsland and/or someone from ovirt-node helps 14:40:56 danken1, looking at it now 14:41:00 danken1: libvirt wasn't restarted 14:41:10 it's not only mine decision, but I think it worth waiting, 14:41:20 danken1: threw exception though , but no restart 14:41:23 ybronhei: are you sure? have you looked at meni's logs? 14:41:35 danken1: looked in rose08 14:41:41 We saw the ominous "libvirt version" text in libvirtd.log. 14:41:47 where? 14:41:50 yeah, that was it 14:41:55 danken1: and it's defined as MUST, so we will wait for it. 14:42:06 guys, lets move on 14:42:14 last restart was at 14:40 14:42:31 the issue took place at 15:30 14:42:35 #info networking: single blocker, might take a week to solve 14:42:44 ybronhei: I do not have the logs in front of me - please discuss this with amuller. 14:42:44 ybronhei: please do that discussion in private. 14:42:53 next 14:42:55 libvirt through exception, might be because the conf file was changed.. 14:43:10 #info SLA 14:43:18 bugs: nothing big. no blockers but bz#987926 need to make it in asap 14:43:21 doron: any 3.3 blocker? 14:43:40 ofri: I see 2 bugs for 3.3 on your list 14:43:45 oschreib: not really. we have a minor issue with vdsm which should be merged 14:43:47 yes 14:44:02 one of them is 987926 14:44:04 ofri: so please move the to --- version 14:44:17 the other is planned to be merged in few minutes 14:44:19 oschreib: what about the other network bugs with TR 3.3? 14:44:30 ofri: it only awaits verification 14:44:41 itamar: I asked danken to take a look on them. 14:44:47 (and then a backport to the ovirt-3.3 branch) 14:44:55 i know. msivak is working on it now 14:46:09 ofri: doron: so no blockers for 3.3? 14:46:39 doron. is 987926 a blocker? 14:46:59 sec 14:47:01 sounds like one? 14:47:06 itamar: I've just moved https://bugzilla.redhat.com/show_bug.cgi?id=988397 to ON_QA, I do not think we have anything to do there now 14:47:37 ofer it is. 14:47:39 itamar: so the only net blocker is the registration flow. 14:47:56 danken1 - but also add it to blocker tracker if fails testing? 14:48:23 itamar: ? /me fail to parse your question 14:49:10 ofri: verified 14:49:16 danken1 - to have this bug on our radar for 3.3 version, wouldn't it be more prudent to add to the version tracker bug (tracking blocker bugs), so if it fails testing, we'll see it? 14:49:46 Quick note - I just added https://bugzilla.redhat.com/show_bug.cgi?id=984028 as ovirt blocker 14:49:52 it *i*s blocking the tracker 14:50:21 danken1 - ok. 14:50:26 itamar: 988397 blocks the tracker 14:51:07 sahina - shouldn't bug 988299 be a blocker for gluster (well, and storage)? a vm can't launch from the new gluster storage domain 14:51:41 amuller: I agree that "Keep ovirt management as default in the main routing table always" is pretty bad 14:52:07 it makes the new multigw feature quite troublesome to use 14:52:10 danken1: it is pretty bad, but it is existing functionality 14:52:20 it was already broken before 3.3 14:52:24 apuimedo_: But the new feature exposes this bug 10000x 14:52:43 itamar, there are certain settings to be done on the gluster volume outside of ovirt. not sure if user did that before using as storage domain. 14:52:47 amuller: that is also true 14:53:00 guys, we have to go further. 14:53:11 itamar, i will add as blocker till we repro and figure out issue. 14:53:30 #info no sla blockers 14:53:32 next 14:53:37 #info UX 14:53:44 sahina - great. thanks 14:53:54 who's here for UX? ecohen is that you? 14:54:05 oschreib, yes - sorry for late response. 14:54:33 oschreib, one issue I would like to fix (default language selection), other than that - nothing seems like a blocker to me 14:54:35 ecohen: Do we have blockers for 3.3? I see 10 bugs for 2.2 14:54:46 oschreib, I will change their target release soon. 14:54:47 oschreib: wasn't 987926 an sla blocker? 14:55:07 oschreib, only https://bugzilla.redhat.com/show_bug.cgi?id=988008 is worth fixing. 14:55:20 itamar: it is verified and approved. it will be in 14:55:51 itamar: I thought ofri said there are no blockers, only a single issue that needs to get in. 14:56:00 ofri: taken. now please backport. 14:56:17 it requires a backport, so need to be tracked... 14:56:19 danken1: rogger that 14:56:57 ecohen: I'd add it to the release blockers, just to track it, if it's going to get in. 14:57:02 (itamar: though it's not soooo horrible anyway) 14:57:31 ok, ecohen, anything eles? 14:57:33 else* 14:57:40 oschreib, that's it, thanks 14:57:59 #info ux: single important issue should get it 14:58:01 next 14:58:06 #info storage 14:58:35 amureini: are you representing storage team here? 14:59:17 https://bugzilla.redhat.com/show_bug.cgi?id=987965 is somehow listed under storage 14:59:22 I thought it was virt 14:59:30 "Cannot run vm with spice console protocol " 14:59:44 you suggested storage 15:00:21 I understand we have no storage rep 15:00:40 danken1: thanks, I'll take care of the 6 storage bugs. 15:00:43 mskrivanek: there is a storage problem there, but I think that the crux is virt 15:01:07 #info no storage representative 15:01:10 next 15:01:16 #info virt 15:01:24 danken1: I can take a look, but I would doubt it doesn't work...I'm sure people succeeded creating a VM during test day 15:01:33 mskrivanek: any 3.3 blockers? I see 22 bugs for 3.3 15:01:55 not anymore:-) moved them to ---. Then there is this 987965 as danken1 mentioned 15:02:07 And https://bugzilla.redhat.com/show_bug.cgi?id=988065 15:02:18 yah, I'd be happy if you could take a look in it 15:02:19 the rest is not important and I have removed 3.3 15:02:33 mskrivanek: oj, thanks 15:02:48 mskrivanek: please make sure you cherry pick them. 15:02:58 danken1: if you're fine with the dirty fix from Assaf, can you merge it? 15:03:36 mskrivanek: any estimation regarding those blockers? 15:03:56 mskrivanek: I am not at all fine with it... hence I gave it -1. 15:04:00 danken1: I backported (cherry-pick) the ksmtuned patch - http://gerrit.ovirt.org/#/c/17537/ 15:04:16 I think we should collected the true alias for all devices 15:04:40 msivak: taken 15:05:08 oschreib: danken1: yeah, so the estimate depends more on Dan and Assaf to come to an agreement. Vinzenz is fixing Console in 3.3 so we won't see the bug in 3.3<->3.3 migrations, but that doesn't solve the issue for previous releases 15:05:33 mskrivanek: why? 15:05:36 #info virt: two blockers currently 15:05:53 I'll let you guys discuss it after the meeting :) 15:06:05 but we're really after the single hour we had. 15:06:07 ok, I think we're done with 3.3 issues. I'll build RC in the next few days, and will build more once patches will get in 3.3 branch. anything else to discuss? 15:06:42 I suggest discussing Infra/conventions/other in the next week, as we're passed our schedule. 15:06:49 oschreib: what about node bugs? 15:06:57 danken1: because it's still broken in 3.0. Console nor Balloon has an alias. And it's not going to be sent during migration so it will crash unless you patch 3.0/3.1/3.2. That's my understanding 15:06:58 i see 3 of them with TR 3.3 15:07:15 oh, missed node. 15:07:33 that's what happens when mburns isn't here 15:07:37 #info node 15:08:01 3 node bugs with target release 3.3 15:08:34 jboggs: fabiand: any update on 3.3 node bugs? 15:09:08 no update afaik, wasn't aware of the meeting :) 15:09:15 mburns and fabian are on pto 15:09:30 im working on 988986 15:09:57 there are 3 bugs for 3.3, can you take a look/talk with mburns, and change their TR if they're not blockers? 15:10:14 will do 15:10:22 jboggs: thanjs 15:10:24 jboggs: we could use your help with the failing registration flow 15:10:31 danken1, which bz? 15:10:48 988990 15:10:49 could you remind me 15:10:49 jboggs: https://bugzilla.redhat.com/show_bug.cgi?id=988990 15:11:11 how vdsm-reg collects the target Engine url from kernel args? 15:11:25 #info node: 3 bugs for 3.3, node team will take a look on them, currently only a single blocker. 15:11:29 jboggs: thanks 15:11:43 ok, I think we passed all the areas. 15:11:48 itamar: anything we missed? 15:11:54 jboggs: please move non blockers to TR --- 15:12:06 danken1: I was leaning for a try getOvirtBridgeParams 15:12:12 oschreib: i think not - next time just go by the sort order of the report ;) 15:12:12 if the bridge doesn't exist 15:12:26 create the bridge on top of the nic 15:12:31 (vdsm-reg) 15:12:50 apuimedo_: this would be unhelpful for Engine 3.3. 15:13:02 I'd rather skip the bridge building, if possible 15:13:09 itamar: yah, I got that mistake after the third team :) 15:13:16 i.e. unless a user explicitly asked to register to an old Engine 15:13:38 anything else guys, or I can close this meeting? 15:13:52 (you can keep on with tech discussions, I don't mind) 15:14:05 danken1: that would mean do a rest call to the engine to ask it about the version 15:14:11 wouldn't it? 15:14:22 oschreib, one thing - when u said remove feature from 3.3, does it mean also change feature compat level 15:14:24 apuimedo_: no, I'm opting for using the ovirt-node TUI 15:14:42 that they would add a checkbox 15:14:43 oschreib, or just remove from feature page for 3.3 release? 15:14:51 and pass it to vdsm-reg? 15:14:57 apuimedo_: but frankly, your idea is better (though more complex, I think) 15:15:04 sahina: remove it from 3.3 ovirt release page 15:15:34 apuimedo_: yes, that's what I was asking from douglas. 15:15:40 oschreib, k 15:15:56 endmeeting in 3 15:15:59 danken1: well, I'd hope that the version information would be the first thing that vdsm-reg retrieves 15:16:01 apuimedo_: but there should be a way to check that box via ovirt-node kernel args 15:16:02 tbh 15:16:09 hence my question to jboggs. 15:16:24 danken1: that would surprise me 15:16:42 danken1: what would you get from cmdline? 15:17:10 2 15:17:42 danken1: jboggs: on the kernel args I don't see anything relevant in the node 15:17:43 1 15:17:50 what information would you be looking for? 15:17:51 #endmeeting