20:01:29 <markwash> #startmeeting glance
20:01:29 <openstack> Meeting started Thu Jun 12 20:01:29 2014 UTC and is due to finish in 60 minutes.  The chair is markwash. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:01:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:01:32 <openstack> The meeting name has been set to 'glance'
20:01:57 <esheffield> o/
20:02:03 <markwash> First item on the agenda today is trying to pick a time for the mid cycle meetup
20:03:27 <markwash> are there any issues with the 24th and 25th of July for folks present?
20:03:34 <markwash> arnaud__: ^^
20:03:37 <arnaud__> :)
20:03:46 <arnaud__> a sec
20:04:27 <arnaud__> seems good
20:04:47 <jcook> I have vacation then
20:05:04 <jcook> ends the 27th
20:05:19 <nikhil___> markwash: a sec, we'r are doing a quick co-ordination amongst folks at the Rackspace as ashwini is not around today
20:05:40 <markwash> I'm out Aug 1 through 8
20:06:03 <arnaud__> and July 28 - Jul 30 Nova
20:06:56 <markwash> oh, the general location is "bay area, california"
20:07:06 <markwash> maybe in SF, maybe in Palo Alto
20:07:34 <nikhil___> markwash: post aug 8th might be issue for some folks here as the schools starts then
20:07:55 <arnaud__> yeah good point nikhil___
20:08:59 <nikhil___> some nova folks would want to attend and prefer to travel in the dates range of nova summit
20:08:59 <markwash> okay
20:09:07 <alaski> o/
20:09:13 <alaski> I'm one of those Nova folks :)
20:09:16 <nikhil___> (rather before or after it)
20:09:52 <markwash> so july 24th and 25th are right before the nova one
20:09:55 <markwash> so that's sort of a bonus
20:10:09 * markwash wishes TravT were around
20:10:42 <markwash> okay, apart from jcook I'm not hearing a no quite yet
20:10:49 <markwash> so I think we might have to settle on that
20:10:56 <markwash> I"ll send something to the list
20:11:29 <markwash> maybe there can be a weekend caravan up from SF to Portland
20:11:45 <markwash> I know a lot of great breweries on the way
20:11:47 <alaski> +1
20:11:55 <arnaud__> +1
20:12:25 <markwash> lets move on, I think an email is the best we can hope for out of this topic for right now
20:12:45 <arnaud__> markwash, I will update the etherpad (https://etherpad.openstack.org/p/glance-juno-mid-cycle-meeting)
20:12:58 <markwash> arnaud__: thanks!
20:13:27 <markwash> actually, I'm not sure we have that much in the way of agenda items today, after I give some updates about project and mission stuff
20:13:40 <markwash> so let me run through those project / tc meeting updates briefly
20:14:00 <markwash> First, the TC approved our overdue statement of our old mission.
20:14:12 <markwash> So yay! we have mission now, just in time to change it :-)
20:14:28 <markwash> We also had a discussion in the TC meeting about the new mission
20:14:30 <jokke_> \\o \o/ o// o/7
20:14:46 <markwash> review is still here https://review.openstack.org/#/c/98002/
20:15:00 <markwash> I think we've got at least one more revision coming but I'm waiting to gather a bit more feedback
20:15:12 <arnaud__> for those who missed it: http://www.openstack.org/blog/2014/06/openstack-technical-committee-update/
20:15:26 <markwash> some of that feedback is already there I see, so I'll try to make a pass through tomorrow
20:15:52 <markwash> The TC also chatted with me about our api consistency issue and actions
20:16:08 <markwash> and generally said "Go for it, we like your proposal better and think everything else should become consistent with your proposal"
20:16:19 <hemanth_> \0/
20:16:26 <markwash> so great job on that proposal folks (hemanth_, rosmaita, et al)
20:16:53 <markwash> there's a little snag about restiness in the ML but I think we've got consensus to move ahead
20:17:43 <markwash> Finally, we also reviewd glance's compatibiltiy with the general integration requirements
20:18:01 <markwash> the only urgent takeaway from that meeting was that we need to do some immediate better testing in tempest
20:18:16 <markwash> specifically, apparently there is an issue where we were not testing high-bit-set content in image uploads
20:19:02 <markwash> I don't have a bug link for you unfortunately
20:19:05 <jokke_> markwash: sorry, could you open that up a bit, or am I the only one out as a snow man?
20:19:06 <markwash> but we need to tackle that pronto
20:19:35 <markwash> jokke_: sorry I"ve been a bit of a firehose, which part do you want me to elaborate on?
20:20:13 <jokke_> So what's the issue with testing we are having and specially what did you mean with that high-bit-set?
20:20:48 <markwash> apparently our tempest tests are missing a fair amount of v2 coverage and, also, image uploads are not testing "binary" image data, only ascii
20:21:01 <jokke_> aha
20:21:12 <jokke_> thanks
20:21:28 <markwash> so ideally for me, we'd have a patch submitted to tempest to fill that gap by next tuesday
20:22:11 <markwash> any takers :-)
20:22:57 <iccha> else markwash will pick someone :p
20:23:05 <markwash> we could also settle for filing a bug or other ticket and assigning someone
20:23:16 <markwash> iccha: haha
20:23:25 <markwash> okay I'll loop back around on that one
20:23:36 <markwash> There were also recommendations that I think made a lot of sense
20:23:38 <jokke_> unless the timeline was so tight I'd be more than happy to participate on that, but I think this is bit short notice to start looking what tempest is :P
20:23:50 <markwash> 1) generally expand our tempest test coverage
20:24:04 <markwash> 2) try to even out our review cadence
20:24:50 <markwash> I think some of the review cadence stuff should be improved as we bring in the graffiti and catalog folks more into the team
20:25:14 <markwash> The last project update I have is, we released juno-1
20:25:55 <markwash> #topic open discussion
20:26:05 <markwash> a slow summer day it seems like :-)
20:26:17 <jokke_> What's the plan for Image API v1?
20:26:48 <jokke_> Is the deprecation schedule still in place (Deprecated at Juno, removed at K)?
20:26:49 <markwash> our plan was to move nova to v2
20:26:58 <markwash> to add tempest tests for v2
20:27:05 <markwash> and then announce the planned deprecation of v1
20:27:32 <markwash> jokke_: based on the lack of momentum it might be here a bit longer
20:27:59 <markwash> at this point, we could not remove it at K IIUC
20:28:17 <wayne__> Regarding graffiti, wanted to mention last week we (HP/Intel)submitted an rst doc covering the metadata-schema-catalog service proposal
20:28:22 <wayne__> (https://review.openstack.org/#/c/98554/3/specs/juno/metadata-schema-catalog.rst)
20:28:29 <jokke_> markwash: tbh I wouldn't mind ... It would be great to see how V2 performs in real life before throwing v1 out of the window
20:28:38 <wayne__> We welcome any comments.
20:28:47 <markwash> wayne__: thanks
20:29:18 <arnaud__> https://review.openstack.org/#/c/84887/ for those interested
20:29:29 <markwash> jokke_: we also had a hintergedanken of firming up tests around v1 and then porting it to the domain code, to make future maintenance easier
20:29:58 <wayne__> I should note the document is on the long side and to that issue we were wondering if there was any interest in an IRC or Google hangout meeting to facilitate/expedite a discussion around.
20:29:59 <markwash> arnaud__: cool thanks
20:31:11 <jokke_> One more v1  & v2 question. Do we have a plan to make rollback possible or do we stick on one way plan?
20:31:43 <markwash> jokke_: one way plan with loads of time for deprecation
20:31:46 <jokke_> Our testing has indicated that FE images created via v2 does not show up on v1 listing
20:31:52 <markwash> wayne__: that's a good idea
20:32:01 <TravT> Yes, we can host a google hangout or IRC discussion on the metadata catalog for anybody that is interested.
20:32:07 <markwash> jokke_: FE ?
20:32:13 <jokke_> markwash: For Example
20:32:46 <markwash> jokke_: that. . . sounds very strange
20:33:31 <jokke_> markwash: I haven't been able to verify it fully yet, but that was message I got ... v1 created stuff works without issues via v2 but not ther way around
20:33:37 <markwash> regarding TravT and wayne__ 's point: can we get a show of hands people here who feel like the need more info about Graffiti stuff ?
20:33:52 <markwash> I feel pretty well informed at this point but I would attend to help facilitate
20:34:24 <markwash> its okay if people feel like they've got the picture as much as they need already :-)
20:34:30 <jokke_> TravT: thanks for the video demo ... made the spec doc way more understandable (hit for everyone, watch it)
20:34:34 <jokke_> hint even
20:34:50 <TravT> jokke_ thanks.
20:34:54 <nikhil___> is there a link?
20:34:58 <TravT> Let me grab that link.
20:35:08 <jokke_> nikhil___: it's in the spec proposal .... a sec
20:35:12 <nikhil___> have seen the doc not the video
20:35:19 <nikhil___> kk
20:35:29 <TravT> link: https://www.youtube.com/watch?v=Dhrthnq1bnw
20:35:45 <nikhil___> #link https://www.youtube.com/watch?v=Dhrthnq1bnw
20:36:09 <nikhil___> s/seen/browsed/g
20:36:31 <nikhil___> Thanks TravT !
20:36:59 <TravT> Sure!
20:37:05 <markwash> so it sounds like maybe you guys have already got it covered for the folks present, TravT
20:37:26 <markwash> I'll #info your offer just in case there are more folks reviewing the meeting logs
20:37:48 <TravT> Ok.  If anybody wants to talk in person about it, probably best to email me at travis.tripp@hp.com and we'll set up a hangout.
20:38:20 <markwash> #info TravT et al available for a google hangout to further explain and answer questions about graffiti work
20:38:36 <markwash> TravT: once the spec is nailed down and approved, how far are you guys from proposing merges?
20:38:57 <markwash> s/merges/patches/
20:39:10 <TravT> Thanks markwash. We've got some stuff started in horizon, but can't go too far without moving further on the Glance side.
20:39:36 <markwash> TravT: it would be amazing if we could have a substantial amount proposed for juno-2 at least
20:39:53 <TravT> Definitely!
20:40:17 <markwash> TravT did you see our previous discussion about a midcycle meetup date of july 24-25th ?
20:40:50 <TravT> No, we barely just joined. We fully expect that the code reviews will provide more opportunity for refinement...
20:41:00 <TravT> We are interested in attending that.
20:41:11 <markwash> TravT: do you know if that date range is feasible for you?
20:42:00 <TravT> I think it is.. I really want wayne and lakshmi to be there too since they will be doing a lot of code.
20:42:19 <wayne__> Those dates work for me
20:42:48 <markwash> okay cool
20:42:51 <TravT> markwash: basically code wise, we had it all working in our standalone POC service
20:42:55 <markwash> right
20:43:09 <TravT> but we'll need to adapt to whatever standards the Glance team uses..
20:43:22 <TravT> For example, the POC was pecan / wsme
20:43:48 * markwash cringes at the idea of switching it to our non-pecan/wsme setup
20:44:26 <markwash> I would love to see if the pecan/wsme and eventlet-homespun stacks could stand side-by-side in our project without requiring another server process
20:44:49 <TravT> We could look into that.
20:44:49 <wayne__> we could try that on our side
20:45:09 <markwash> thanks that would be great
20:45:44 <TravT> Otherwise, we did change a few other things from POC, because believe it or not we learned a few things doing a POC. ;-)
20:46:02 <markwash> haha
20:46:14 <markwash> okay folks I think we've about covered what we have for today
20:46:21 <markwash> shall we close out?
20:46:39 <jokke_> sounds good, thanks
20:47:13 <markwash> thanks everybody
20:47:14 <TravT> sounds good.
20:47:16 <markwash> #endmeeting