14:01:36 <mburns> #startmeeting ovirt node weekly sync
14:01:36 <ovirtbot> Meeting started Thu Jan  5 14:01:36 2012 UTC.  The chair is mburns. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:01:36 <ovirtbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:01:45 * jboggs` here
14:01:48 <mburns> #topic Roll Call
14:02:04 <mburns> anyone else here?
14:02:17 <mburns> pmyers: ?
14:02:21 * pmyers in
14:02:36 <mburns> ok, key players are here at least
14:02:48 <mburns> #info mburns jboggs` pmyers are here
14:02:53 <mburns> #topic agenda
14:03:03 <mburns> Action item review
14:03:03 <mburns> 2.2.1 release
14:03:03 <mburns> oVirt first release status
14:03:03 <mburns> Meeting time
14:03:03 <mburns> Open discussion
14:03:16 <mburns> anyone have anything else?
14:03:28 <pmyers> nope
14:03:37 <mburns> #topic action item review
14:03:40 <mburns> #link http://ovirt.org/meetings/ovirt/2011/ovirt.2011-12-15-14.00.html
14:03:59 <mburns> jboggs: where are we with uefi?
14:04:08 <mburns> (usb patch was pushed today)
14:04:32 <jboggs> it works for usb, due to a design feature, sata /internal drives require more work, im drafting an email to explain to the list
14:04:58 <mburns> #info uefi works for usb, still working on sata/internal drives
14:05:05 <mburns> #action jboggs to send email with status to list
14:05:23 <pmyers> jboggs: who have you reached out to at this point for expert help on uefi in general?
14:05:24 <jboggs> gist is: livecd-iso-to-disk only supports grub1, so boot screen is empty, and can't include grub-efi due to conflicts chicken/egg
14:05:38 <jboggs> mgarret was a huge help and bcl form livecd-list
14:05:49 <pmyers> ok, cc them on your email thread that you're going to start
14:05:56 <pmyers> and let's drag them into ovirt.org mailing lists :
14:05:57 <pmyers> :)
14:06:22 <xTs_w> masayag: http://pastebin.com/rY74bFwr so, it doesn't seem to be a pebkac
14:06:28 <mburns> jboggs: i think pjones is a uefi expert
14:06:30 <xTs_w> trying for hours now to compile ovirt
14:06:38 <mburns> my personal action items:
14:06:58 <mburns> #info mburns talked to dougsland about vdsm patches -- should all be merged
14:07:13 <mburns> #action mburns to test vdsm/registration with build today
14:07:20 <jboggs> mburns, im tottaly at the 99% marker on uefi its just sitting a a blinking cursor now :)  and the only folks that have used uefi+grub2 are fedora+macs
14:07:33 <jboggs> so not any good examples
14:07:40 <mburns> jboggs: ok
14:08:01 <mburns> #info still facing real hardware issues, will follow up with smoke testing today
14:08:28 <jboggs> mburns, i can help with that just give me details
14:08:32 <ovirtbot> 14[[07Node Backlog14]]4 !10 02http://www.ovirt.org/w/index.php?diff=1536&oldid=1174&rcid=1580 5* 03Pmyers 5* (+520) 10
14:08:48 <mburns> #action mburns to parse rhel bugs for upstream inclusion and flag as "EasyFix"
14:09:10 <mburns> jboggs: might have you update the engine machine
14:09:27 <mburns> but honestly, i haven't looked at the real hardware issues since before the shutdown
14:09:37 <mburns> so i don't recall what was happening exactly
14:09:50 <mburns> #info all pending patches reviewed and pushed
14:10:04 <mburns> ok, that's it for old action items
14:10:22 <mburns> 2.2.1 release
14:10:29 <mburns> #topic 2.2.1 release
14:11:28 <mburns> #info 5 bugs currently opened and flagged for 2.2.1
14:11:48 <mburns> 3 are in post and merged (will move to modified later today)
14:12:05 <primalmotion> mburns: well, I'm still stuck on the login prompt, even after pulling the change you pushed
14:12:26 <mburns> other 2 are uefi and a simple change with /etc/hosts file
14:12:34 <mburns> primalmotion: ok, will look after this meeting
14:12:40 <primalmotion> thanks
14:13:04 <mburns> #info more bugs will be cloned from rhel 6 bug list to 2.2.1 release
14:13:07 <pmyers> mburns: 2.2.0 list still hase 1 bug in NEW and 5 in MODIFIED
14:13:26 <pmyers> mburns: so we need to close current release the modified ones and move the NEW one to 2.2.1 probably?
14:13:26 <mburns> pmyers: yes, modified i need to close
14:13:36 <pmyers> the NEW one is 753297
14:13:44 <mburns> new one is f16 based ovirt node iso image release
14:13:46 <mburns> which is done
14:13:48 <pmyers> ok
14:13:57 <mburns> so i'll close that one as well
14:14:02 <pmyers> so just bad bugzilla hygiene then :)
14:14:08 <mburns> yep
14:14:18 * mburns head slaps self
14:14:32 <pmyers> yeah where's hewbrocca's halibut when you need it :)
14:14:35 <mburns> pmyers: the NEW one, i just missed when parsing
14:14:46 <mburns> i look at all upstream bugs at the same time
14:14:51 <pmyers> yep
14:14:57 <mburns> and 1 2.2.0 bug in the list is easy to miss
14:15:09 <pmyers> btw the bz queries on http://www.ovirt.org/wiki/Node_Backlog are useful
14:15:15 <pmyers> i just added query for 2.2.1 bugs as well
14:15:23 <mburns> i'd like to get 2.2.1 wrapped early next week
14:15:38 <mburns> #link http://www.ovirt.org/wiki/Node_Backlog
14:16:06 <mburns> well, actually we have to get 2.2.1 wrapped early next week
14:16:11 <mburns> for initial ovirt release
14:16:21 <mburns> #topic first oVirt release
14:16:35 <mburns> #info Release scheduled for 31 Jan
14:16:47 <mburns> #info Test day scheduled for 16 Jan
14:17:11 <mburns> #info Builds need to be tagged/ready for 12 Jan
14:17:38 <mburns> #action mburns to build ovirt-node for 12 Jan
14:17:47 <mburns> #action mburns to build iso image for 13 Jan
14:18:41 <mburns> given how close these dates are, i might punt some of the easy rhel bugs that i'm going to clone later to 2.2.2 or 2.3.0
14:18:50 <mburns> but i'll decide that when i look
14:18:55 <pmyers> 2.2.2 is fine
14:19:05 <pmyers> we know we need to do another build for jan 31 anyhow
14:19:16 <mburns> yep
14:19:17 <pmyers> since we'll need to incorporate new vdsm with RHEVM removed from the oVirt Node plugin
14:19:22 <pmyers> dougsland: ^^
14:19:44 <mburns> well, we need new iso definitely, but not necessarily a new ovirt-node rpm
14:19:50 <pmyers> right
14:19:52 <mburns> we most likely will need it though
14:19:55 <pmyers> so how do we handle that?
14:20:02 <pmyers> if we keep version of ovirt-node the same
14:20:04 <pmyers> but new ISO
14:20:09 <pmyers> how do we version the node itself?
14:20:14 <mburns> ovirt-node rpm will be 2.2.1
14:20:18 <mburns> iso is 2.2.1-0
14:20:20 <pmyers> ah ok
14:20:21 <mburns> 2.2.1-1
14:20:21 <pmyers> cheers
14:20:54 <mburns> or something like that...
14:21:20 <mburns> ok, moving on if there's nothing else for the first release...
14:21:44 <mburns> #topic meeting time
14:21:56 <pmyers> mburns: one quick Q
14:22:01 <mburns> #undo
14:22:01 <ovirtbot> Removing item from minutes: <MeetBot.items.Topic object at 0x993052c>
14:22:02 <pmyers> when do you anticipate 2.2.1 build
14:22:07 <pmyers> what date
14:22:13 <pmyers> and what about 2.2.2, what target date do we want to set for that
14:22:28 <pmyers> and when was 2.2.0 formally released again?
14:22:42 <mburns> 2.2.1 targeted for 11 Jan
14:22:48 <mburns> iso by 12 Jan
14:23:10 <mburns> 2.2.2 will be for 31 Jan
14:23:18 <mburns> but probably a little earlier
14:23:26 <mburns> but we depend on vdsm for that as well
14:24:14 <mburns> #info 2.2.1 rpm build targeted for 11 Jan with iso build 12 Jan (dependent on vdsm being ready)
14:24:42 <mburns> #info 2.2.2 build targeted for 29 Jan with iso build around the same time (dependent on vdsm and timelines set by ovirt release manager)
14:24:46 <pmyers> what about original release date for 2.2.0?
14:25:16 <mburns> 7 Dec
14:25:24 <mburns> was when the build was done
14:25:36 <ovirtbot> 14[[07Node Backlog14]]4 !10 02http://www.ovirt.org/w/index.php?diff=1537&oldid=1536&rcid=1581 5* 03Pmyers 5* (+669) 10
14:25:48 <pmyers> ok take a look at http://www.ovirt.org/wiki/Node_Backlog and verify my dates
14:26:08 <mburns> ack
14:26:16 <ovirtbot> 14[[07Node Backlog14]]4 !10 02http://www.ovirt.org/w/index.php?diff=1538&oldid=1537&rcid=1582 5* 03Pmyers 5* (-70) 10
14:26:30 <mburns> #topic meeting time
14:27:10 <mburns> i have a standing call immediately prior to this meeting that tends to run over
14:27:19 <pmyers> tuseday at same time then?
14:27:31 <mburns> and they're talking about moving it half an hour later too, so looking for new time proposals
14:27:38 <jboggs> tues wfm
14:27:40 <mburns> pmyers: i thought you had a conflict tuesday at 9?
14:27:52 <pmyers> nope
14:28:09 <mburns> ok, then i'll move this to tuesday at 9
14:28:32 <pmyers> send out an updated invite to the node devel list to let folks know
14:28:48 <mburns> #action mburns to update meeting request to be tuesdays at 9AM ET
14:29:13 <mburns> #topic open discussion
14:29:29 <pmyers> we went through 2.2.0, 2.2.1 and you're going to create some 2.2.2 bugs
14:29:37 <mburns> there's interest in ovirt-node from the archipel team
14:29:39 <pmyers> but we should take a look at 2.3.0 and review to make sure that everything is still feasible
14:29:48 <mburns> primalmotion is looking at adapting to make it work
14:29:53 <pmyers> what's archipel?
14:30:05 <mburns> http://archipelproject.org/
14:30:32 <pmyers> so something that competes with ovirt and/or openstack?
14:31:02 <mburns> looks similar
14:31:09 <pmyers> (not raising objections or anything just trying to understand scope)
14:31:37 <mburns> if nothing else, from node perspective it should help us understand flexibility and extendability
14:31:42 <pmyers> agreed
14:31:43 <mburns> which will help for plugins
14:31:45 <pmyers> i think it's a good idea
14:31:59 <pmyers> however it does start begging the question of whether or not we need a more general project
14:32:07 <jboggs> interesting it uses xmpp for comm
14:32:22 <pmyers> i.e. oVirt Node is really a variant of 'Embedded Node' or something
14:32:22 <mburns> pmyers: as for 2.3.0 bugs, i think we should defer to tuesday for that discussion
14:32:27 <pmyers> mburns: ack
14:33:04 <pmyers> mburns: do you think it makes sense to generalize oVirt Node to be smth like Embedded Node (please find a better name) and then make oVirt Node a variant that takes plugins for it?
14:33:29 <mburns> pmyers: i think it's certainly something to investigate
14:33:38 <pmyers> or should we use oVirt Node as the base and in some cases plugins for other variants will remove core things
14:33:47 <mburns> but i think we need to solidify the plugin stuff before we spend anytime on it
14:33:49 <pmyers> like for example I doubt that project you mentioned uses vdsm so it would remove vdsm
14:33:52 <pmyers> agreed
14:34:06 <mburns> pmyers: i really don't like the idea of removing anything
14:34:11 <pmyers> nor do I
14:34:22 <pmyers> so that really means we should look at adding 'oVirt/vdsm' as a plugin
14:34:26 <mburns> so for now, archipel would need to disable what it doesn't need
14:34:47 <pmyers> there may be other use cases for oVirt Node even outside of virtualization enviornments
14:34:53 <mburns> but long term, with plugins, we can move vdsm/ovirt-engine stuff to be true plugins
14:34:59 <mburns> pmyers: true
14:35:06 <pmyers> I've discussed with folks from gluster community around using the oVirt Node structure for a storage brick/appliance
14:35:12 <pmyers> which doesn't even need libvirt/kvm at all
14:35:16 <pmyers> just the gluster stuff
14:35:18 <mburns> though i'm not sure how much we want to go down that path with removing things
14:35:30 <pmyers> right so in that case we'd need a JEOS base image
14:35:41 <pmyers> upon which we can add virt support, ovirt support, gluster support, etc
14:35:48 <mburns> eventually, we get to the point that the plugins are too big and the node is too small to be useful
14:36:07 <pmyers> ok, so then you would argue that those variants are different kickstarts
14:36:09 <pmyers> vs. plugins
14:36:29 <mburns> i think we should stick with a base virt node with libvirt/kvm
14:36:34 <pmyers> ok
14:36:46 <mburns> for the gluster use case, they can disable virt stuff
14:36:50 <pmyers> ok fair enough
14:36:57 <mburns> but i think we want to stick with virt as our base
14:37:03 <pmyers> for now, agreed
14:37:13 <mburns> yes, for now
14:37:21 <mburns> it's all moot until we have working plugins anyway
14:37:24 <pmyers> yep
14:37:51 <mburns> any other topics for open discussion?
14:37:52 <jboggs> gluster a good plugin example at this point, i can work with them on it
14:37:55 <jboggs> ?
14:38:11 <pmyers> not sure if it's a short term thing for them
14:38:16 <pmyers> it maybe be further out on their roadmap
14:38:17 <pmyers> but
14:38:24 <pmyers> you can definitiley work with them
14:38:36 <jboggs> do we have a contact already?
14:38:48 <mburns> jboggs: vdsm/engine might be the place to start
14:38:55 <mburns> see what it does to the node already
14:39:01 <jboggs> yeah can do that then
14:39:48 <mburns> ok, anything else?
14:39:56 * mburns starts countdown
14:40:00 <mburns> 3....
14:40:15 <mburns> 2...
14:40:33 <mburns> 1..
14:40:46 <mburns> OK, thanks everyone
14:40:48 <xTs_w> your seconds are pretty long ;)
14:40:49 <mburns> #endmeeting