14:01:14 <markwash> #startmeeting glance
14:01:15 <openstack> Meeting started Thu Sep 26 14:01:14 2013 UTC and is due to finish in 60 minutes.  The chair is markwash. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:01:16 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:01:19 <openstack> The meeting name has been set to 'glance'
14:01:20 <markwash> there we go
14:01:23 <zhiyan> \o
14:02:24 <markwash> First I wanna talk about cutting RC1 and then briefly take a look at proposed summit sessions
14:02:31 <markwash> #topic cutting rc1
14:03:08 <flwang> markwash: I'd like to add this bug in rc1 https://code.launchpad.net/bugs/1231255
14:04:23 <markwash> let's take a look at that in  a bit
14:05:01 <markwash> So, the way cutting an RC works, is we basically do so when we are comfortable that the RC1 bug list is empty/low enough in importance
14:05:21 <markwash> #link https://launchpad.net/glance/+milestone/havana-rc1
14:05:28 <markwash> this process is more flexible than I thought
14:05:45 <markwash> I had previously thought we were going to be forced to cut rc1 today come hell or high water
14:05:47 <iccha> is there an rc2 markwash ?
14:06:04 <markwash> iccha: only if strictly necessary
14:06:11 <iccha> ok gotcha
14:06:13 <rosmaita> we had high water in our dev space this morning
14:06:15 <markwash> meaning, only if we happen to find a release-critical bug after cutting rc1
14:06:33 <flwang> markwash: so any due date/deadline for rc1?
14:06:45 <markwash> well, I think the best thing for us to do is to cut it this week
14:06:48 <markwash> and I think that is possible
14:07:01 <flwang> markwash: got
14:07:06 <markwash> that gives us about 1-2 days to deal with remaining bugs on the list
14:07:20 <iccha> yeah the bug list looks pretty good
14:07:26 <markwash> in my opinion we can quickly handle the several "medium" or higher bugs on the list
14:07:54 <iccha> wish the bugs had in review status
14:07:59 <markwash> heh me too
14:08:17 <markwash> anyone know the status of https://bugs.launchpad.net/bugs/1211315 ?
14:08:32 <markwash> "optimize queries for image-list"
14:08:32 <iccha> markwash: its in review
14:08:39 <iccha> I had taken a look at it and looked good to me
14:08:58 <flwang> +1 from me for that
14:09:03 <iccha> wait one second
14:09:14 <zhiyan> markwash: yes, i kown that 1211315
14:09:25 <markwash> it appears to be abandoned?
14:09:31 <iccha> so the bug shows two reviews
14:09:36 <iccha> we should look at first one
14:09:42 <iccha> https://review.openstack.org/#/c/41404/
14:09:54 <zhiyan> actually i don't think it's a bug, i think it's a improvement...not sure should belong to rc1 as a bug...
14:10:16 <iccha> its more like a wishlist item. but does imporve speed of queries for glance
14:10:22 <venkatesh> "optimize queries for image-list": https://review.openstack.org/#/c/41404/
14:10:28 <iccha> will be a good performance improvement imo
14:10:43 <venkatesh> sorry will update the launchpad with the review link
14:11:38 <markwash> Okay. . as nice as it is, it looks like a ton of fresh code so I'm a little more worried about regressions and new bugs than otherwise
14:11:39 <zhiyan> iccha: yes, but it has regress risk
14:11:48 <markwash> so I think it makes a lot of sense to wait
14:12:36 <iccha> i would want all reviwers to play around with patch and feel comfortable before we approve it
14:13:27 <iccha> but its been around for a long time, even if we dont have it in rc1, it will be good for reviewers to check it out
14:13:37 <zhiyan> iccha: TBH, +1 to markwash, wait to I is more safe..
14:14:17 <iccha> yeah I am ok with it not being in rc1. But just want to make sure we dont push it to back burner :)
14:14:22 <markwash> anyone have issues with the rc1 schedule I proposed? if we don't fix something in our list by COB tomorrow eastern, it doesnt' make it and I cut the release anyway?
14:14:29 <zhiyan> iccha: actually i think the change can get my +1, but just a little worried on regress.. :)
14:14:59 <iccha> zhiyan: maybe after meeting u can express any concerns u have and we can talk to venkatesh ?
14:15:05 <zhiyan> markwash: i'm ok. but one question, do you think flwang's change #48401 should be included?
14:15:55 <zhiyan> iccha: sorry, i have a meeting later, but i like do it later? if you ok..
14:16:13 <iccha> yeah sure zhiyan
14:16:20 <zhiyan> iccha: cool.
14:17:09 <markwash> zhiyan: I'm interested in a fix but wonder how drastic is a workaround?
14:17:41 <markwash> yeah I wanna add that to rc1 I guess
14:17:59 <zhiyan> iccha: actually i'm ok on the code change, but the change have no any test case ...
14:18:08 <flwang> markwash: it's a fix not a workaround :)
14:18:10 <markwash> I've got a few questions from quickly scanning the review, so I'll note those down soon
14:18:22 <markwash> flwang: right, the code change is. . but I'm wondering about workarounds in case the fix is not in place
14:18:40 <zhiyan> markwash: yes, it's a fix. but flwang actually it also has some minor issue, i will put my comments asap later..
14:19:14 <zhiyan> markwash: end user can do necessary policy.json change to make workaround IMO.
14:19:34 <iccha> zhiyan: ah thats coz there is no change in functionaility. wish we could have performance tests. the current tests passing means it didnt break anything. but then again how much can we rely upon glance tests :)
14:20:11 <zhiyan> markwash: i mean to enable get_image_location always
14:20:43 <flwang> zhiyan: why?
14:21:02 <markwash> since I know its late in china I went ahead and added my notes on that review now
14:21:07 <markwash> but I am going to add that bug to rc1 also
14:21:16 <flwang> zhiyan: did you mean enable get_image_location for all user?
14:21:41 <zhiyan> flwang: for necessary user
14:22:01 <zhiyan> according to admin, maybe all user.
14:22:08 <zhiyan> markwash: cool. thanks.
14:22:22 <zhiyan> iccha: yes, i mean unit test :)
14:22:27 <iccha> i think it should be in rc1 too
14:22:36 <iccha> flwang: 's patch ^
14:22:56 <zhiyan> iccha: flwang +1 for rc1 bug
14:23:00 <markwash> I think that about covers the rc1 topic from my perspective. . anyone else have questions or items related to rc1?
14:23:14 <iccha> markwash: https://bugs.launchpad.net/glance/+bug/1229042
14:23:36 <rosmaita> markwash: in writing up the docs for image sharing, i found a few bugs
14:23:38 <zhiyan> iccha: aha, aready in
14:23:44 <flwang> markwash: how about this https://review.openstack.org/#/c/46986/ ? just a minor typo fix for glance api
14:23:46 <rosmaita> do you think we can get them in rc2?
14:23:51 <flwang> harmless
14:24:23 <markwash> rosmaita: if they're critical enough for rc2, we should delay releasing rc1 to address them
14:24:40 <markwash> rosmaita: otherwise, they can certainly be ported to stable/havana later
14:24:44 <zhiyan> markwash: humm TBH we'd better keep control on the list...
14:25:11 <markwash> rosmaita: at this point, critical is not the word I meant to use. . just trying to say if the importance of the bug would justify a "high" priority
14:25:15 <rosmaita> i think they can wait
14:25:20 <rosmaita> there are workarounds
14:25:49 <zhiyan> markwash: do not include more bug which not really need... to low the regress risk..
14:26:00 <rosmaita> although, i do have a patch up for one, it's an incorrect http response code, would be good to fix so people don't code against the response and then we change it
14:26:25 <rosmaita> https://bugs.launchpad.net/glance/+bug/1229366
14:26:27 <markwash> rosmaita: can you post links to the bugs.  it would only take a few minutes to double check the priority
14:26:30 <rosmaita> but i don't insist on it
14:26:50 <rosmaita> give me a sec to write up the other bugs, just confirmed them
14:27:51 <markwash> +2 on that one. . I don't see a big risk and its a wart that's nice to remove
14:28:37 <flwang> markwash: how about this? https://review.openstack.org/#/c/46986/
14:29:02 <flwang> but I don't insist on that :)
14:29:55 <markwash> flwang: seems low risk
14:30:08 <flwang> markwash: yep
14:30:10 <markwash> okay. . anyone okay with moving on to the next topic?
14:30:17 <iccha> https://review.openstack.org/#/c/46767/
14:30:18 <markwash> s/okay/not okay/ ?
14:30:47 <markwash> ah, that's a good point iccha
14:30:54 <markwash> there was some additional work with property protections
14:31:01 <markwash> that sort of popped up
14:31:17 <markwash> iccha: is there more than just that review you'd like to see come in under the FFE header?
14:31:19 <iccha> markwash: since the additional work is not gonna be in rc1. i think changes to documentation not needed for now
14:31:31 <markwash> ah okay
14:31:45 <iccha> markwash: i was thinking will think patch also go in rc1?
14:32:24 <iccha> in other words my question is there are some protected porperties patches like https://review.openstack.org/#/c/46283/5
14:32:28 <iccha> and the documentation one
14:32:37 <iccha> do they come under rc1?
14:32:43 <iccha> just prioritizing my reviews
14:32:46 <iccha> so the question
14:33:28 <iccha> i would like to see the documentation go in if not the other patches
14:33:45 <markwash> gotcha
14:33:51 <zhiyan> iccha: i think https://review.openstack.org/#/c/46767 doc change should wait, since the '@' role change is not landing yet.
14:34:32 <markwash> I think the docs could, but it looks like that dependency is real
14:34:36 <iccha> if we can get hold of thomas i feel that we can remove the dependency and see if we can get the documentation patch ?
14:34:52 <iccha> so it does not have the info related to the patch it is dependent on
14:34:57 <zhiyan> iccha: if so that's ok to me.
14:34:59 <markwash> if we can remove that dependency (removing @ ! from the docs themselves too) thenI'm cool
14:35:12 <iccha> i ll try and see if i can contact thomas in time
14:35:31 <iccha> becoz logically speaking it ll be good to have the documentation with the feature when packaage is cut
14:35:52 <zhiyan> iccha: make sense to me.
14:36:22 <iccha> anyone know his irc nick?
14:36:35 <markwash> not I
14:36:36 <zhiyan> :-[ seems we can landing a lot of things in rc1
14:36:46 <zhiyan> iccha: sorry
14:36:53 <iccha> ya I understand
14:37:01 <iccha> we re being too last minute about it too :(
14:37:07 <iccha> I should have brought it up earlier
14:38:02 <markwash> zhiyan: :-( indeed. . it feels to me like these are generally either really important or really low risk
14:38:13 <markwash> but please do make the case if you disagree about the risk or importance
14:38:30 <markwash> I don't want us to regress because groupthink miscalculated the risk
14:38:54 <markwash> indeed, everyone should be looking at these changes with a very skeptical eye
14:39:25 <zhiyan> :P my eyes is nice
14:39:43 <flwang> markwash: yep, we need heavily review for those in progressing rc1 patches
14:39:50 <iccha> also we need to think of it as a package . in terms of what can this release do with and without
14:39:51 <zhiyan> markwash: yes, i totally agree that
14:39:54 <iccha> flwang: +1
14:40:09 <iccha> all reviewers on board for next 2 days
14:40:27 * markwash is
14:41:00 <iccha> sorry mark took up too much time on rc1
14:41:13 <zhiyan> markwash: iccha next topic?
14:41:17 <markwash> okay. . bring up any other issues with rc-important bugs after meeting, and be very skeptical during reviews to play things safe
14:41:19 <zhiyan> ... if we have
14:41:21 <markwash> no worries!
14:41:28 <markwash> #topic summit topics
14:41:32 <zhiyan> cool
14:41:37 <markwash> just wanted to check in with those really fast
14:41:39 <markwash> we have two so far
14:41:50 <markwash> I think folks can start submitting them now
14:41:56 <markwash> (some time ago evidently)
14:41:58 <iccha> link http://summit.openstack.org/
14:42:00 <zhiyan> summit.openstack.org
14:42:01 <zhiyan> yes
14:42:22 <iccha> also markwash u said few topics in mind
14:42:42 <markwash> yes, I've just been too lazy to add them to the list
14:43:03 <iccha> does one of them have to do with testing :p
14:43:06 <markwash> #1 I wanna talk about the future of v1. . when do we deprecate? can we make it use the domain and delete all its custom stuff
14:43:19 <markwash> iccha: hmm I hadn't thought to, but we could
14:43:23 <flwang> markwash: iccha: do you think https://blueprints.launchpad.net/glance/+spec/image-error-state-management  can be re-discuss since it's a real problem for end user
14:44:21 <flwang> there are many bugs attached to it
14:44:42 <markwash> flwang: I'd really like for us to tackle that issue, but I want to rephrase the blueprint to (being frank) make the outcome I want to see more likely
14:45:26 <markwash> what I'd really like to see is nova tracking the progress of uploads
14:45:30 <flwang> markwash: yep, that's what I will do next
14:45:39 <markwash> we might even want to submit that as a nova talk honestly
14:46:01 <flwang> markwash: I think we may need a notification from Nova to tell Glance what happened
14:46:14 <markwash> I think having the conversation with both glance and nova folks will allow us to actually make progress
14:46:26 <markwash> instead of the neglect/stonewall stalemate we've had for many cycles now
14:46:30 <iccha> +1
14:46:32 <flwang> markwash: sure
14:46:43 <markwash> okay cool
14:47:14 <flwang> markwash: so do you think it a good topic for summit?
14:47:15 <markwash> so just a few items there, then. . I'm only looking to add the one I mentioned, anyone feel free to add what they want to that list and bug me to take a look at it
14:47:19 <markwash> flwang: yes
14:48:07 <flwang> markwash: cool
14:48:22 <markwash> all right. . feels like open discussion time to me :-)
14:48:31 <zhiyan> just thinking maybe bresnah probably need a proxy.. to talk about https://blueprints.launchpad.net/glance/+spec/framework-for-contributed-storage-systems
14:49:15 <zhiyan> and i have one  location-property. markwash ^
14:49:58 <markwash> yeah, go ahead and submit it or both
14:50:13 <markwash> its okay if our submissions overlap or have to be adjusted
14:50:40 <markwash> really its a good sign if we have a few of those problems
14:50:45 <markwash> #topic open discussion
14:50:48 <zhiyan> cool, roger
14:51:03 <iccha> i pinged thomas leaman for the documentation patch
14:51:36 <iccha> and he ll have another one today
14:51:57 <zhiyan> markwash: do we have a release plan on the glanceclient part?
14:52:06 <flwang> iccha: cool
14:52:14 * markwash frantically eats cereal before the bus arrives
14:52:32 <iccha> zhiyan: good question
14:52:51 * markwash wonders if he netsplit
14:53:00 <markwash> #endmeeting