20:03:59 <nikhil_k> #startmeeting Glance
20:03:59 <openstack> Meeting started Thu Nov 13 20:03:59 2014 UTC and is due to finish in 60 minutes.  The chair is nikhil_k. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:04:00 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:04:02 <openstack> The meeting name has been set to 'glance'
20:04:31 <nikhil_k> #link https://etherpad.openstack.org/p/glance-team-meeting-agenda
20:04:57 <nikhil_k> kragniz: abhishekk jokke_ : around ?
20:05:01 <jokke_> yup
20:05:10 <kragniz> nikhil_k: o/
20:05:10 <jokke_> are we the only ones? ;)
20:05:27 <nikhil_k> see there ack from hemanthm mfedosin sigmavirus24 ?
20:05:27 <sigmavirus24> I'm here
20:05:30 <kragniz> nope, there's a bunch more
20:05:34 <nikhil_k> okay
20:05:36 <mfedosin> +1
20:05:48 <curtis_p> +1
20:06:08 <nikhil_k> arnaud__: flaper87 ?
20:06:10 <jokke_> Mike & Curtis \o
20:06:29 <mfedosin> jokke_, o/
20:06:32 <nikhil_k> not sure if Travis and Lakshmi are here?
20:06:42 <nikhil_k> anyways, let's get started
20:06:50 <hemanthm> o/
20:06:58 <nikhil_k> #Topic Virtual mini summit survey results
20:07:12 <nikhil_k> so, haven't had anymore responses yet
20:07:17 <nikhil_k> just those 8
20:07:43 <nikhil_k> and it would be good to have input on the form to decide the location and time of the kilo mid-cycle meetup
20:08:08 <nikhil_k> so, strongly encourage for everyone to participate
20:08:23 <hemanthm> do we default to virtual mini-summit?
20:08:36 <nikhil_k> not sure yet
20:08:38 <jokke_> nikhil_k: so is it still open if we will have it together with Nova at VMWare or not?
20:08:47 <nikhil_k> current responses indicate that no
20:09:09 <abhishekk> nikhil, sorry I am here now
20:09:18 <nikhil_k> jokke_: current results seem to indicate that we need to have it in east coast area
20:09:24 <kragniz> nikhil_k: due to nova peeps, or scheduling glance folk?
20:09:47 <nikhil_k> am not sure if there is much to discuss sitting alongside Nova
20:10:06 <jokke_> nikhil_k: I think the info was also not there when that survey was posted
20:10:07 <nikhil_k> however, I can confirm that after syncing up with the Nova PTL
20:10:09 <kragniz> nikhil_k: those results were before we knew we could have the meetup alongside nova, though?
20:10:25 <jokke_> nikhil_k: I'm willing to join Palo Alto if we get together with Nova folks
20:10:50 <nikhil_k> if I start the survey again, people will not fill it out
20:11:02 <nikhil_k> if what I'm afraid about
20:11:04 <nikhil_k> is*
20:11:24 <nikhil_k> and we do not have participation at the meetings
20:11:26 <sigmavirus24> Can people update existing responses?
20:11:30 <kragniz> nikhil_k: why not just send out a very small new one?
20:11:30 <nikhil_k> which is kinda of annoying
20:11:39 <jokke_> so don't let the survey dominate ... make executive call based on new opportunities :P
20:12:05 <nikhil_k> jokke_: lol, however we want to respect the open-ness of the community
20:12:11 <sigmavirus24> ^^ PTL is kind of like BDFL, right?
20:12:22 * sigmavirus24 kids
20:12:25 <nikhil_k> what if people are not willing to travel that far, loses the purpose of mid-cycle meetup
20:12:33 <jokke_> nikhil_k: I doubt it will change a lot if it's at East Coast or West Coast ... it's Biatch for EMEA/APJ anyways
20:13:18 <nikhil_k> jokke_: ok, only if I'd some confirmation from the other people as well
20:13:33 <nikhil_k> let me try to figure it out offline
20:13:35 <jokke_> nikhil_k: that's easy, send e-mail to the mailing list "Is there anyone strongly opposing?" ... you probably won't get a single response and you have given opportunity :)
20:13:52 <nikhil_k> we'd decide on the location by next Thursday
20:13:58 <jokke_> +1
20:14:16 <kragniz> nikhil_k: sounds good
20:15:06 <nikhil_k> so, how do we decide on the Meeting time?
20:15:26 <nikhil_k> is everyone here okay with keeping all the meetings at 14UTC?
20:15:32 <nikhil_k> and stop the alternating time
20:15:37 <jokke_> I'm good with 1400
20:15:38 <hemanthm> works for me
20:15:46 <kragniz> works for me, too
20:15:54 <sigmavirus24> No objections here but i'm not a regular attendee
20:16:07 <hemanthm> guess we gotta check with zhiyan
20:16:10 <curtis_p> 1400 is good
20:16:19 <mfedosin> i like 2 pm every time
20:16:35 <nikhil_k> ok, so that might be another note of disappotment sent to ML and mention of the new time
20:16:45 <kragniz> at the very least, we can trial it for a few weeks
20:16:55 <hemanthm> I'll miss fewer meetings if we are not alternating :)
20:17:10 <kragniz> if a few people don't like it, we can change to something else
20:17:11 <nikhil_k> we want to avoid changes to reduce scheduling conflicts
20:17:19 <kragniz> nikhil_k: true
20:17:21 <abhishekk> 1400 sounds good
20:18:13 <nikhil_k> #topic Guidelines for: Core reviewers, review quality, membership etc
20:18:26 <nikhil_k> seems like a favority topic :)
20:18:50 <nikhil_k> we've had a lot of requests for core-reviewer membership lately
20:19:04 <jokke_> that's good
20:19:22 <nikhil_k> and in order to keep fairness for all those who are earning/earned it the hard way
20:19:25 <kragniz> I didn't know that was a thing
20:19:52 <sigmavirus24> ^^
20:19:53 <hemanthm> Isn't it only by nomination?
20:19:53 <nikhil_k> it seemed like a good idea to keep some guidelines
20:20:05 <nikhil_k> coming to that
20:20:27 <jokke_> That's what I thought as well until the summit :D
20:20:35 <jokke_> +1 for guidelines
20:20:40 <nikhil_k> all those who are already doing quality reviews, please keep up the good work
20:20:40 <hemanthm> +1
20:20:50 <abhishekk> +1
20:21:29 <nikhil_k> here I just wanted to say that we must understand that numbers help however, that'd not be the only deciding factor
20:22:08 <nikhil_k> please give your valuable feedback so that we will have good changes to the patch sets
20:22:23 <mfedosin> btw, what is a procedure of becoming a core?
20:22:25 * jokke_ stops writing that auto +1 script
20:22:34 <nikhil_k> and the changes go through smoothly with fewer patch sets and less of wait time for developers
20:22:38 <kragniz> jokke_: I knew it
20:22:49 <mfedosin> does another core needs to bite you? :)
20:22:55 <hemanthm> jokke_: mind sharing the script? :P
20:23:13 <hemanthm> mfedosin: lol
20:23:18 <nikhil_k> it helps developers working on the bugs/features, helps the community with less review queue and helps a ton on the infra gate
20:23:58 <nikhil_k> just wanted to mention that to whoever is proposing or is going to propose candidate names to me offline
20:24:43 <nikhil_k> best idea for all the developers joning the circle this cycle would be
20:25:28 <mfedosin> it's pretty clear
20:25:32 <nikhil_k> helping the core-reviewer assigned to a BP should help in getting that core-reviewer and me to nominate you
20:25:57 <nikhil_k> it will also help the BP and you will also have build a good rapport with community member
20:26:42 <nikhil_k> all in all, please keep focusing on the quality reviews
20:27:08 <hemanthm> I've just started putting together a brain dump for the review guidelines, https://etherpad.openstack.org/p/glance-review-guidelines
20:27:15 <nikhil_k> my feeling is that a good comments without a -1 or +1 is worth more than the rank in the stats
20:27:32 <nikhil_k> ok great
20:27:43 <hemanthm> It'd be helpful if you all can contribute
20:27:47 <nikhil_k> #topic Review Guidelines
20:27:50 <nikhil_k> #link  https://etherpad.openstack.org/p/glance-review-guidelines
20:28:24 <nikhil_k> there a bit more guidelines set out there
20:28:35 <jokke_> Could I ask, Please don't stop reviewing code at the point you see first -1 there
20:28:35 <hemanthm> feel free to drop in whatever comes to your head. I'd be happy to organize stuff into meaningful buckes
20:28:44 <nikhil_k> specifically, Nova team has a good page for their reviews
20:28:49 <nikhil_k> #link https://wiki.openstack.org/wiki/ReviewWorkflowTips
20:29:07 <jokke_> at least I tend to wait day or two to get all the wishes in the next PS not just the oneliner someone pointed out
20:29:29 <nikhil_k> jokke_: yes, unless it's something that indicates that the patch is not fully complete, not applicable to cycle , etc
20:30:27 <nikhil_k> does anyone have any comments, concerns?
20:30:42 <jokke_> nikhil_k: btw that wikipage is by no means Nova specific even the exampleas are having nova there
20:30:42 <kragniz> nikhil_k: not really, all sounds good
20:31:02 <nikhil_k> jokke_: yeah, I'd have rephrased that
20:31:08 <nikhil_k> we need something Glance specific
20:31:29 <nikhil_k> something that indicates - regular errors, gotchas, etc.
20:31:38 <jokke_> nikhil_k: what glance specific we need there?
20:31:49 <nikhil_k> so that the reviewers don't have to comment exact same thing on every single patch
20:32:11 <nikhil_k> this would be a good guideline for both reviewers and the developers
20:32:19 <kragniz> nikhil_k: +1
20:32:26 <jokke_> nikhil_k: good luck with that .... like hacking doc or logging guidelines had helped there ;)
20:32:59 <nikhil_k> Let's see where we land
20:33:05 <nikhil_k> We can try, for now
20:33:36 <nikhil_k> #topic glance_store in launchpad
20:33:47 <nikhil_k> #link https://launchpad.net/glance-store
20:33:52 <jokke_> nikhil_k: sounds good ... would be great get something more generic so that every project didn't have to have their own guidlines and if you're jumping between, trying to remember what was the one here
20:34:39 <nikhil_k> jokke_: sure, may be a sub page for glance specific things. Overall structure being generic
20:35:09 <nikhil_k> I believe hemanthm is leading that effort
20:35:23 <mfedosin> I can help him
20:35:25 <nikhil_k> so please co-ordinate with him
20:35:37 <mfedosin> got some experience in that
20:35:46 <hemanthm> mfedosin cool, thanks
20:35:51 <nikhil_k> moving to glance_store
20:35:52 <mfedosin> okay
20:36:02 <jokke_> nikhil_k: cool ... as this is going to be handholding cycle anyways, I'll reach Sean, Jay and Doug if we can get some generic best practises out there through openstack-specs
20:36:07 <jokke_> ok
20:36:17 <nikhil_k> jokke_: that sounds great. thanks!
20:36:25 <nikhil_k> glance_store: please do file bugs against that project
20:36:33 <nikhil_k> and not against glance
20:36:49 <jokke_> oh gr8
20:36:53 <kragniz> I moved a coule of existing ones against it this morning
20:37:11 <nikhil_k> release notes, dates, etc. info should be available on that page at some point
20:37:20 <nikhil_k> kragniz: excellent!
20:37:48 * kragniz thinks it needs a fancy openstack logo
20:37:50 <kragniz> :P
20:38:12 <nikhil_k> #topic Reviews, Specs, priorities, etc.
20:38:34 <nikhil_k> Am trying to come up with a good solution where we can collaborate on the most important things to do
20:38:51 <nikhil_k> etherpad, trello, etc. tools come to mind
20:39:05 <kragniz> nikhil_k: trello might be cool
20:39:10 <jokke_> etherpads are just pain
20:39:15 <kragniz> iirc another program uses it
20:39:17 <mfedosin> yep, trello is nice
20:39:23 <jokke_> never touched trello ... need to have a look
20:39:23 <hemanthm> +1
20:39:31 <nikhil_k> that would help keep the velocity going and keep people aware about the most important changes *now* in glance, glance_store and python-glanceclient
20:39:35 <kragniz> so using it in openstack isn't something new
20:39:44 <kragniz> nikhil_k: +1 to that
20:40:06 <nikhil_k> so, for tracking glance_store and client releases we just started using trello
20:40:07 <jokke_> nikhil_k: reg prioritization ... is it cores who are also bug drivers?
20:40:30 <nikhil_k> jokke_: well there is that bug team
20:40:39 <jokke_> nikhil_k: and if not, who is, because that is not working
20:40:45 <nikhil_k> however, glance used to be very small that there was literally no distinction
20:41:03 <jokke_> nikhil_k: yeah that was exactly my point ... who is on that bug team
20:41:24 <nikhil_k> ok, so may be we can keep that a part of core duty
20:41:44 <nikhil_k> jokke_: it's public, I think
20:41:47 <nikhil_k> need to double check
20:41:56 <kragniz> nikhil_k: pretty sure anyone can join
20:42:09 <nikhil_k> ok, so that will not work
20:42:12 <jokke_> it's great that we pay attention to BP's and specs and gate, but we have again 260 bugs logged only against glance and there is no noise about it outside the odd bug days
20:42:24 <nikhil_k> jokke_: agreed
20:42:44 <nikhil_k> let's make cores do it
20:43:04 <nikhil_k> that kinda makes sense as these people are aware about the project overall
20:43:05 <hemanthm> how many active cores do we have?
20:43:44 <nikhil_k> 4-5 maybe atm, not sure as the term "active" is rather hard to define ;)
20:43:50 <kragniz> hemanthm: well we have one nikhil_k at this meeting!
20:44:07 <nikhil_k> so, that's my disappointment
20:44:14 <jokke_> imho it's not acceptable that we have 111 new bugs and I think even more that has no priority set
20:44:17 <nikhil_k> no cores available to guide the meeting
20:44:26 <hemanthm> precisely my point. We are pushing more responsibility to cores but we don't have too many of them active
20:44:33 <hemanthm> no disrespect here
20:44:51 <nikhil_k> what do you propose hemanthm ?
20:45:01 <nikhil_k> I was a bit hesistent to create another team
20:45:23 <nikhil_k> as we already have, cores, bug team, glance drivers, glance_store cores!
20:45:41 <nikhil_k> just keeps getting complicated
20:45:53 <jokke_> nikhil_k: is this now for what?
20:46:18 <nikhil_k> jokke_: did not get you there
20:46:46 <nikhil_k> we may be running out of time here
20:46:54 <nikhil_k> so, I propose the following
20:46:58 <jokke_> nikhil_k: for what action you're referring with "I was a bit hesistent to create another team"
20:47:04 <sigmavirus24> I think we're stuck on the how, maybe take this to the ML?
20:47:12 <hemanthm> not sure I have a good suggestion. But, what privileges do we need for the people in bug team?
20:47:36 <abhishekk> in summit, we had a discussion in contributors meetup to check active cores
20:47:37 <nikhil_k> ML is getting a bit crowded these days ;)
20:47:58 <abhishekk> jokke_: remember?
20:48:07 <nikhil_k> #topic feedback
20:48:26 <nikhil_k> how about each one of you please send me your feedback on the following?
20:48:33 <jokke_> hemanthm: there is Glance Bug Team in the launchpad and those people are the only ones who can do stuff like set priority, close bug as "Won't fix" etc
20:48:41 <nikhil_k> 1. bug team ideology
20:48:50 <nikhil_k> 2. summit feedback
20:49:14 <nikhil_k> 3. review guidelines (if you did not get a chance today)
20:49:19 <nikhil_k> 4. anything else
20:49:39 <nikhil_k> and we can keep the rest of the 10 mins for the few items on the agenda list
20:49:46 <kragniz> nikhil_k: sounds good
20:49:52 <hemanthm> nikhil_k: sure
20:50:04 <nikhil_k> thanks!
20:50:15 <nikhil_k> #topic #openstack-glance IRC channel logging
20:50:22 <nikhil_k> kragniz: was that you?
20:50:37 <kragniz> does anyone have any objections to having #openstack-glance logged?
20:50:42 <jokke_> o/
20:50:54 <nikhil_k> we may want to discuss that with the infra team
20:51:01 <kragniz> there have been a couple of times when it would have been useful to point someone at the logs
20:51:12 <kragniz> nikhil_k: I already have, the patch is very easy
20:51:23 <jokke_> I do not want it to be publicly logged ... if you want to have logs out of it, have your client sitting there and logging it for you
20:51:37 <nikhil_k> okay cool
20:51:54 <nikhil_k> jokke_: yeah, good point.
20:51:56 <mfedosin> we have our own bouncer ant it's alright
20:52:01 <kragniz> jokke_: okay
20:52:15 <sigmavirus24> it might make sense to make an FAQ out of stuff we'd otherwise use logs for
20:52:15 <nikhil_k> However, freenode prolly means that you need to watch the language
20:52:26 <jokke_> I tend to do stuff like popping my e-mail address there and that will be flooded with spam if it gets indexed by some spambot
20:52:39 <nikhil_k> makes sense
20:52:41 <kragniz> jokke_: having those logs on http://eavesdrop.openstack.org/irclogs/ would have been nice a while back
20:52:48 <kragniz> jokke_: but it doesn't really matter that much
20:53:23 <kragniz> so let's not log, then
20:53:34 <kragniz> let's move on, since we're low on time
20:53:54 <nikhil_k> cool. yes, let's NOT log seems to be the direction (just in case)
20:54:04 <nikhil_k> #topic bug/1387973
20:54:13 <nikhil_k> abhishekk: please go ahead
20:54:34 <abhishekk> nikhil_k: I am ok with your approach
20:54:36 <nikhil_k> yt
20:54:50 <abhishekk> i have submitted patch against stable/juno
20:55:05 <nikhil_k> ok, cool. so, we can wait until next week and see what happens
20:55:14 <abhishekk> i hope will get positive response
20:55:14 <nikhil_k> should we move on?
20:55:20 <abhishekk> sure
20:55:28 <nikhil_k> #topic Multiple sort keys
20:55:41 <nikhil_k> who proposed this topic
20:55:42 <nikhil_k> ?
20:55:43 <mfedosin> okay, it's mine
20:55:55 <mfedosin> excuse me, I'm not a native speaker, so I can't type very fast, but I'll try my best
20:56:02 <mfedosin> There is an old bug
20:56:08 <mfedosin> https://bugs.launchpad.net/glance/+bug/1221274
20:56:11 <uvirtbot> Launchpad bug 1221274 in glance "allow sorting with more than one field" [Undecided,In progress]
20:56:24 <mfedosin> as for me it's a feature :)
20:56:31 <jokke_> +1
20:56:34 <hemanthm> +1
20:56:34 <mfedosin> some guy wants an ability to sort image list output with multiple keys
20:57:00 <mfedosin> I implemented these features both in v1 and v2, but then cores said me that v1 is will be soon deprecated and I removed this part of code
20:57:13 <nikhil_k> mfedosin: okay, then a spec ? we can discuss further there
20:57:17 <jokke_> sounds like that's something the bug team should have prioritized as "Wishlist"
20:57:20 <mfedosin> So, v2 is here and I want to ask you: do we need this feature in Glance? And if yes, then do we need it for task and metadefs (the last one is not my own private main, but I'm just interested)
20:57:58 <mfedosin> https://review.openstack.org/#/c/120727/ https://review.openstack.org/#/c/120777/
20:58:17 <mfedosin> The patch doesn't break backward compatibility, so, previous requests stay the same, but now it supports csv keys (owner, size for example) Glance client has support too and there is a normal test suit
20:58:41 <mfedosin> review it please :)
20:58:54 <mfedosin> I'm done here
20:59:05 <nikhil_k> sounds like we need some reviews on that soon
20:59:20 <nikhil_k> we'd talk a bit more about it later
20:59:27 <mfedosin> okay, thanks!
20:59:39 <nikhil_k> sigmavirus24: we will have to discuss your item on the regular channel
20:59:42 <nikhil_k> sorry about that
20:59:43 <kragniz> poor sigmavirus24 has had to sit though the whole meeting for this!
20:59:47 <sigmavirus24> lol
20:59:54 <sigmavirus24> it's okay. i joined the bug team
20:59:57 <sigmavirus24> so i have that going for me
21:00:02 <kragniz> sigmavirus24: heh
21:00:02 <jokke_> ok back to openstack-glance
21:00:05 <nikhil_k> #endmeeting