14:04:19 <oschreib> #startmeeting oVirt Weekly Meeting
14:04:19 <ovirtbot> Meeting started Wed Aug  7 14:04:19 2013 UTC.  The chair is oschreib. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:04:19 <ovirtbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:04:28 <oschreib> #topic agenda and roll call
14:04:32 * dustins_ntap here
14:04:35 * jb_netapp here
14:04:42 * deepakcs here
14:04:46 * ofri here
14:04:48 <oschreib> #info 3.3 status update
14:05:06 * derez here
14:05:10 <oschreib> (Afterwards, if we will have some spare time)
14:05:16 <oschreib> #info infra team update
14:05:23 <oschreib> #info Conference and Workshop
14:05:27 <oschreib> #info Other Topics
14:05:40 <oschreib> (waiting a minute for people to come)
14:06:01 * fabiand is here
14:06:11 * evilissimo is here
14:06:15 <ybronhei> also here
14:06:16 * danken is here.
14:06:29 * evilissimo is here on behalf or the virt team
14:06:31 * sahina here
14:06:54 <oschreib> shell we start the fun part? :)
14:07:06 <oschreib> #topic 3.3 status update
14:07:13 <oschreib> #info ovirt 3.3 releae page and tracker:
14:07:18 <oschreib> #link http://www.ovirt.org/OVirt_3.3_release-management
14:07:24 <oschreib> #link https://bugzilla.redhat.com/show_bug.cgi?id=918494
14:08:03 * amureini here, sorry for the delay
14:08:26 <oschreib> current GA is 2013-08-14, RC release was a week ago and didn't happen (we will discuses those dates straight after blocker review)
14:08:53 <oschreib> I'll go over blocker bugs from the tracker, in order to understand where are we standing
14:09:08 <oschreib> please, don't go into too much technical details...
14:09:18 <oschreib> any comments, so far?
14:09:38 * jboggs is here as well
14:10:32 <oschreib> #info blocker review
14:11:10 <oschreib> #info: Bug 984586 - Cannot start a VM with USB Native - Exit message: internal error Could not format channel target type. (backend, MODIFIED)
14:11:27 <oschreib> who's here for infra?
14:12:42 <ybronhei> i am
14:12:59 <ybronhei> any question?
14:13:03 <oschreib> indeed
14:13:11 <oschreib> Bug 984586 - Cannot start a VM with USB Native - Exit message: internal error Could not format channel target type.
14:13:20 <ybronhei> its on infra? let me see
14:13:56 <oschreib> few issues: 1. it's RHEVM bug for some reason. It should be duplicated to ovirt. please make sure it happens.
14:14:36 <oschreib> 2. What's the bug status? is it cherry-picked to ovirt-engine-3.3 branch? (if not, any estimation on that?)
14:15:50 <ybronhei> on modified, so it should.
14:16:05 <ybronhei> ill verify it in a minute with martin
14:16:24 <oschreib> ybronhei: can you please take care of #1 as well?
14:16:25 <ybronhei> we'll clone to ovirt
14:16:29 <oschreib> thanks
14:16:33 <ybronhei> yes.
14:16:42 <oschreib> #info Bug should have been fixed.
14:16:44 <oschreib> next
14:16:45 <evilissimo> martin left already home I think
14:17:14 <oschreib> ybronhei: we must make sure this bug gets into 3.3 branch.
14:17:35 <ybronhei> oschreib: no worries, ill check it
14:17:56 <evilissimo> http://gerrit.ovirt.org/#/c/17708/ is in ovirt-engine-3.3 branch
14:18:15 <evilissimo> which is part of that bug
14:18:17 <evilissimo> it's merged
14:18:31 <oschreib> ybronhei: so please close that bug as CLOSED_NEXT_RELEASE (Or ON_QA if you have someone to test it)
14:18:53 <oschreib> #info Bug 988004 - [vdsm] OSError: [Errno 2] No such file or directory: '/sys/class/net/ovirtmgmt/brif' (vdsm ON_QA)
14:18:55 <ybronhei> oschreib: ok
14:19:06 <oschreib> danken: amuller_WFH: I see it's on_qa.
14:19:15 <oschreib> 1. do we have a new vdsm build with it?
14:19:33 <oschreib> 2. any reason for it to be ON_QA and not CLOSED_NEXTRELEASE?
14:20:33 * danken looking
14:21:34 <danken> oschreib: yes, it's in vdsm-4.12.0, and mpavlik should test it.
14:22:04 <danken> but let's not waste more time on this - I'll close it.
14:22:23 <oschreib> danken: thanks
14:22:40 <oschreib> #info bug already fixed
14:23:06 <oschreib> next
14:23:22 <oschreib> #info Bug 988065 - Migration fails - AttributeError: 'ConsoleDevice' object has no attribute 'alias' (vdsm, MODIFIED)
14:23:33 <oschreib> evilissimo: that's all yours
14:23:36 <evilissimo> it's merged in master and vdsm ovirt-3.3
14:23:43 <oschreib> any build with it?
14:23:46 <evilissimo> not sure though if it was build already
14:24:01 <oschreib> #info pushed, not sure about build yet
14:24:06 <evilissimo> danken: you know about it?
14:25:19 <evilissimo> well it has not been tagged
14:25:23 <evilissimo> so I assume no
14:25:40 <oschreib> ok, please make sure it's included in the next build :)
14:25:51 <oschreib> next (sorry for the rush, we have 10 more to go)
14:26:23 <oschreib> #info Bug 988299 - Impossible to start VM from Gluster Storage Domain (engine, NEW)
14:26:33 <oschreib> deepakcs: this one is yours
14:26:36 <deepakcs> 988299 - root caused to RPM dep issue for gluster and qemu pkgs.
14:26:37 <deepakcs> gluster pkg dep issue resolved after discussing on gluster-devel. vdsm.spec.in patch needs to be sent.
14:26:37 <deepakcs> qemu pkg dep issue needs some discussion. Sent mail to vdsm-devel, awaiting resolution.
14:27:07 <oschreib> so any time estimation for build?
14:27:16 <oschreib> fix, i meant
14:27:31 <deepakcs> oschreib, I have 1 basic Q.. what OS'es on VDSM host does ovirt 3.3. will support ?
14:27:47 <deepakcs> oschreib, it looks like for Gluster SD, we need qemu 1.3 which is available on F19 but not on centos 6.4!
14:27:53 <deepakcs> and the bug is on centos 6.4
14:28:08 <oschreib> deepakcs: well, that's a good Q
14:28:15 <oschreib> deepakcs: which I'm not sure about
14:28:18 <deepakcs> oschreib, the fix depends on that :)
14:28:33 <oschreib> but a thread on engine-devel will give you the best answer I hope
14:28:41 <deepakcs> oschreib, Who / Why does the oVirt 3.3. test day page says "install F19 or centos 6.4" ?
14:28:45 <oschreib> wait
14:28:47 <oschreib> my bad
14:28:52 <oschreib> I thought of something else.
14:29:11 <oschreib> we should support F19 and Centos 6.4 (hopefully, Fedora 18 as well)
14:29:43 <deepakcs> oschreib, whatever node OS we support its repo should have the qemu 1.3
14:29:56 <oschreib> deepakcs: I understand that
14:30:08 <deepakcs> oschreib, unless we have some ovirt or other repo which hosts qemu 1.3 until its officially availbl eon the distro repo
14:30:33 <deepakcs> oschreib, so I am not sure whats the right way to fix this. I cannot change vdsm.spec.in >= qemu 1.3. that will break many folks
14:30:38 <oschreib> well, I don't have such build, but we can host it if someone will have it.
14:31:01 <oschreib> danken: any thought about qemu req in vdsm spec?
14:31:03 <deepakcs> danken, ^^ any idea how to resolve such dep issues ?
14:31:05 <danken> evilissimo: you were right, there's no rc build with it
14:31:07 <tontsa> is there somewhere "clean" src.rpm of qemu1.3 with gluster support for el6?
14:31:41 <deepakcs> F19 already has qemu 1.4 so the src should be there available within fedora rite ?
14:31:45 <danken> deepakcs: I do not think you can
14:32:20 <tontsa> the F19 src.rpm has systemd/other stuff that is not very el6 friendly
14:32:51 <deepakcs> danken, so we just wait until centos repos pick qemu 1.3 ?
14:33:06 <oschreib> ok, sounds like this is a longer issue, deepakcs/danken, can you start a thread about 6.4 support?
14:33:17 <tontsa> hopefully rhel6.5 beta will have 1.3 included ;)
14:33:41 <deepakcs> oschreib, I already have danken pls see the mail with subject "How to handle qemu 1.3 dep for Gluster Storage Domain". thanks
14:34:01 <danken> I would not be able to join such a thread (/me going on vacation)
14:34:09 <danken> I think you would have to release-note it.
14:34:14 <deepakcs> danken, oh ok.. i was not aware of that
14:34:36 <danken> the only other alternative is to block ovirt-3.3 until that qemu is ready
14:34:40 <oschreib> danken: release note what? that vdsm doesn't work on centos 6.4?
14:34:46 <deepakcs> danken, you mean put qemu 1.3 as a pre-req ? I already have put that on the Gluster_Storage_Domain feature wikipage
14:35:08 <danken> oschreib: that it does not work with the new feature of gluster storage domains.
14:35:45 <oschreib> danken: ok, so we do have vdsm on centos 6.4?
14:36:14 <deepakcs> danken, that would be bad. Why isn't it possible to have a qemu 1.3 for F18, centos 6.4 hosted on some fedora/ovirt repo ?
14:36:22 <danken> danken: I did not try it, but I am not aware of any problem other than gluster.
14:36:36 <oschreib> ok
14:36:47 <danken> deepakcs: we could theoretically host newer qemu on ovirt.org
14:37:05 <danken> taking the benefit of mburns_vacation.
14:37:05 <oschreib> #info no qemu 1.3 for centos 6.4, we might want to release note that issue, followup in the mailing list
14:37:17 <oschreib> next :)
14:37:19 <danken> since he was adamantly against adding such dependencies...
14:37:42 <deepakcs> oschreib, does that mean no gluster SD in 3.3 ?
14:37:45 <danken> deepakcs: I wish http://fedorapeople.org/groups/virt/virt-preview/ had EL6
14:38:21 <oschreib> deepakcs: no, it means we need to go further, and we can decide on that in a mailing list thread
14:38:34 <deepakcs> oschreib, ok, i will watch the list
14:39:03 <itamar> deepakcs  ovirt has gluster SD in 3.3. you need a qemu supporting that. we need to figure out hot to get these for .el6 which is an open issue at this point. i hope dave neary will have some ideas when he's back from vacance.
14:39:03 <danken> deepakcs: it means that there should be a big fat release note - to use gluster sd, install qemu 1.3.
14:39:19 <oschreib> #info BZ#988397 - ovirt-node post-installation setup networks fails when NetworkManager is running (vdsm, ON_QA)
14:39:31 <oschreib> apuimedo: any updates?
14:39:47 <apuimedo> oschreib: no word from QA
14:40:22 <apuimedo> I made my own tests
14:40:30 <apuimedo> but QA must check, I guess
14:40:47 <oschreib> apuimedo: whos the qa for this one?
14:41:28 <itamar> apuimedo: who is "QA"? if you want to ask red hat QE to test an ovirt bug, you need to do so personally. ovirt bugs go to modified until RC build, then move to ON_QA at RC phase.
14:41:34 <apuimedo> Haim
14:42:09 <apuimedo> itamar: alright, I was not aware of this
14:42:11 <oschreib> #info bug is fixed
14:42:41 <oschreib> #info BZ#994523 - [vdsm][multiple gateways] Keep ovirt management as default in the main routing table always (vdsm, POST)
14:42:55 <oschreib> danken:  amuller_WFH: what's up with that bug?
14:42:58 <apuimedo> oschreib: it is still under review that one
14:43:13 <amuller_WFH> oschreib: Working on it
14:43:18 <apuimedo> it was put as post due to a malfunction in the ovirt upstream hooks, I'd say
14:43:31 <amuller_WFH> apuimedo: I put it as post
14:43:39 <oschreib> TTP? (Time to Push :)
14:43:44 <apuimedo> ah, I thought it was the hook
14:43:46 <amuller_WFH> oschreib: Not yet
14:43:59 <oschreib> time estimation, I meant
14:44:19 <amuller_WFH> oschreib: I was ready to ask Danken to merge it but I discovered a bug which I do not understand yet
14:44:29 <amuller_WFH> oschreib: optimistic: 1-2 hours
14:44:36 <apuimedo> amuller_WFH: I think it is something we can deal with today
14:44:44 <oschreib> #info should be pushed today
14:44:48 <oschreib> ok, thanks
14:45:23 <oschreib> #info BZ#990963 - vdsm must require selinux-policy-3.12.1-68.fc19 (vdsm, MODIFIED)
14:45:38 <apuimedo> oschreib: it is pushed
14:45:46 <apuimedo> and verified by Mark Wu of IBM
14:46:14 <apuimedo> I personally checked that the selinux packages have the policy Dan Walsh made for us
14:47:03 <danken> apuimedo: so let's CLOSE.
14:47:13 <oschreib> apuimedo: good work there :)
14:47:19 <apuimedo> thanks
14:47:23 <apuimedo> yes, let's close it
14:47:30 <oschreib> #info issue fixed
14:48:05 <oschreib> #info BZ#988990 - oVirt 3.3 - (vdsm-network): netinfo - ValueError: unknown bridge ens3 (vdsm, ON_QA)
14:48:17 <oschreib> apuimedo: you're very popular in the bugs today
14:48:31 <apuimedo> oschreib: I'm not sure being popular this way is good
14:48:33 <apuimedo> xD
14:48:44 <oschreib> I have to agree
14:48:50 <apuimedo> oschreib: I verified it with the graphical interface
14:48:57 <apuimedo> and it works
14:49:09 <outi_> does the selinux bug affect el6, too?
14:49:11 <apuimedo> if one works around the blocker I added today
14:49:18 <apuimedo> outi_: no, it does not
14:49:20 <outi_> ok
14:49:34 <apuimedo> oschreib: the one about firewalld
14:49:43 <oschreib> we will get to it
14:49:59 <oschreib> so, I understand we can close that one, right?
14:50:42 <apuimedo> yes
14:50:49 <oschreib> #info issue fixed
14:50:50 <oschreib> cool
14:51:13 <oschreib> #info Bug 988986 - libvirt network directory is not persisted (node, POST)
14:51:37 <apuimedo> well, that one I reported but I'm not the one to tell the status, yay!
14:51:49 <oschreib> jboggs: any updates on that one?
14:52:00 <jboggs> oschreib, just awaiting a build
14:52:13 <oschreib> time estimation for it to be close?
14:52:47 <jboggs> asap I'll see if I cna get a build within the next day or so
14:53:10 <apuimedo> jboggs: the build will include the latest vdsm, right? (thinking about vdsm-reg here)
14:53:53 <jboggs> right
14:53:56 <apuimedo> cool
14:53:56 <oschreib> #info fix is posted, should be closed in by tomorrow
14:54:09 <jboggs> apuimedo, which version just to make sure
14:54:36 <apuimedo> jboggs: well, it should be the tip of the ovirt-3.3 branch
14:54:53 <apuimedo> I guess fsimonce or danken will tag a version
14:54:58 <apuimedo> once the meeting is over
14:55:14 <jboggs> ack
14:55:15 <apuimedo> (maybe they'll wait for multiple gateways fix)
14:55:48 <oschreib> #info: BZ#988849 - edit-node does not handle firewall rules (node,  NEW)
14:55:59 <oschreib> jboggs: yours once again
14:55:59 <danken> apuimedo: we'd release a micro version of vdsm with ovirt-3.3, I suppose
14:56:08 <danken> 4.12.1
14:56:13 <jboggs> working on that one right now hope to have it done today
14:56:17 <danken> but only after the last blocker is merged.
14:56:27 <apuimedo> exactly
14:56:58 <danken> jboggs: I think ovirt-node should just take the latest vdsm-reg it has
14:57:07 <apuimedo> jboggs: is the issue on the el6 spin and on f19 or only on the latter?
14:57:09 <danken> and avoid another dependency
14:57:34 <oschreib> #info fix should be published today
14:57:57 <oschreib> ok, last one
14:58:08 <oschreib> #info Bug 987939 - engine-setup -> engine-cleanup -> engine-setup -> fails (installer, NEW)
14:58:13 <jboggs> f19 / firewalld
14:58:22 <jboggs> err sorry all
14:58:31 <oschreib> alonbl: are you here in any case? what's the status of ^^
14:58:52 <alonbl> oschreib: I wrote you already
14:58:58 <alonbl> oschreib: I will reach that when I have time
14:59:16 <oschreib> #info no updates yet,
14:59:26 <oschreib> alonbl: well, it blockers upstream release.
14:59:37 <oschreib> ok, that blocker review took some time
14:59:43 <alonbl> oschreib: so that all other issues we did not find
15:00:00 <oschreib> true.
15:01:06 <oschreib> I suggest trying to release RC at the 14.08, and GA at 28.08. any objections/thoughts?
15:02:20 <oschreib> anyone?
15:02:23 <jboggs> looks ok for our side
15:03:30 <oschreib> danken: itamar: ?
15:04:08 <danken> oschreib: I'm good.
15:04:24 <danken> hmmm,
15:04:45 <danken> do we really need full 2 weeks delay?
15:05:07 <oschreib> well, we need two weeks between RC and GA
15:05:23 <oschreib> and I hope to publish RC early next week.
15:05:35 <oschreib> I havent got all builds yet...
15:05:47 <itamar> and we can re-spin if needed on a critical issues, or fix in a 3.3.0-1/3.3.1, etc. as well
15:06:45 <danken> oschreib: you are right.
15:07:10 <oschreib> ok
15:07:10 <oschreib> so
15:07:12 <oschreib> #info RC at 2013-08-14, GA at 2013-08-28
15:07:21 <oschreib> and I think we're done.
15:07:34 <oschreib> #topic other topics
15:07:43 <oschreib> anything more? or can we close this one?
15:09:49 <oschreib> #endmeeting