Monday, 2016-02-15

openstackgerritwangxiyuan proposed openstack/zaqar: Fix 500 error when create pools in wsgi v2.  https://review.openstack.org/27453701:09
*** stanchan has quit IRC01:22
openstackgerritwangxiyuan proposed openstack/zaqar: Don't return links if subscriptions are empty  https://review.openstack.org/28003501:35
openstackgerritwangxiyuan proposed openstack/zaqar: The subscriber should be unique in redis backend.  https://review.openstack.org/27994601:51
*** achanda has joined #openstack-zaqar01:58
*** flwang1 has quit IRC03:57
*** david-lyle has quit IRC05:21
*** achanda has quit IRC05:32
*** achanda has joined #openstack-zaqar05:41
*** mdnadeem has quit IRC06:10
*** mdnadeem has joined #openstack-zaqar06:11
*** achanda has quit IRC06:58
*** wanghao has joined #openstack-zaqar07:16
*** achanda has joined #openstack-zaqar07:34
*** mdnadeem_ has joined #openstack-zaqar07:51
*** mdnadeem has quit IRC07:55
*** mdnadeem has joined #openstack-zaqar07:55
*** mdnadeem_ has quit IRC07:57
*** jtomasek has joined #openstack-zaqar08:14
*** achanda has quit IRC08:26
*** achanda has joined #openstack-zaqar08:28
*** dynarro has joined #openstack-zaqar08:34
*** rcernin has joined #openstack-zaqar08:34
openstackgerritwangxiyuan proposed openstack/zaqar: Queue NotFound should return 404  https://review.openstack.org/28010608:41
*** achanda has quit IRC08:42
openstackgerritwangxiyuan proposed openstack/zaqar: Queue NotFound should return 404  https://review.openstack.org/28010608:52
*** rcernin has quit IRC09:17
*** rcernin has joined #openstack-zaqar09:17
openstackgerritwangxiyuan proposed openstack/zaqar: Queue NotFound should return 404  https://review.openstack.org/28010609:25
*** GB21 has joined #openstack-zaqar10:10
*** dynarro has quit IRC11:06
*** GB21 has quit IRC11:47
*** GB21 has joined #openstack-zaqar12:13
*** openstackgerrit_ has joined #openstack-zaqar12:27
*** GB21 has quit IRC12:36
*** shakamunyi has quit IRC13:20
*** dynarro has joined #openstack-zaqar13:30
*** GB21 has joined #openstack-zaqar14:08
*** malini has joined #openstack-zaqar14:12
*** sriram has joined #openstack-zaqar14:12
*** malini1 has joined #openstack-zaqar14:17
*** malini has quit IRC14:19
*** ametts has joined #openstack-zaqar14:52
*** GB21 has quit IRC15:10
*** GB21 has joined #openstack-zaqar15:15
*** mpanetta has joined #openstack-zaqar15:15
*** ametts has quit IRC15:20
*** ametts has joined #openstack-zaqar15:20
*** kgriffs|afk is now known as kgriffs15:22
ryansboh jeez, did I miss the meeting folks?15:24
Eva-iryansb: hello. Seems like meeting hasn't started in time.15:30
Eva-iflwang: it's meeting time now15:30
*** GB21 has quit IRC15:32
Eva-iryansb: seems like flwang isn't here. Would you like to have meeting later today?15:37
*** exploreshaifali has joined #openstack-zaqar15:37
Eva-ivkmc: hello, should we attend to meeting now or wait for Fei Long?15:38
*** itisha has joined #openstack-zaqar15:38
ryansbI think later sounds good15:39
*** pt_15_ has joined #openstack-zaqar15:40
Eva-iryansb: me too, let's do it at 21 UTC.15:42
ryansbkk15:44
ryansbEva-i: anything I can work on in the meantime (high-pri reviews etc)?15:44
Eva-iryansb: yes, there's a review I'd like you to do15:46
Eva-iryansb: here https://review.openstack.org/#/c/269398/15:48
ryansbcool, can do15:48
*** mpanetta has quit IRC15:50
*** mpanetta has joined #openstack-zaqar15:50
openstackgerritEva Balycheva proposed openstack/zaqar: Support binary messages over websocket  https://review.openstack.org/25697816:08
*** exploreshaifali has quit IRC16:23
Eva-iryansb: thank you. I'd like to discuss this during the meeting: http://paste.openstack.org/show/kYFFmShi2eiouoYnUz5B/16:43
Eva-iryansb: and this, maybe: http://paste.openstack.org/show/niT4ouOOEhedv2gCEWZO/16:45
*** rcernin has quit IRC16:48
*** csoukup has joined #openstack-zaqar17:19
*** rcernin has joined #openstack-zaqar17:24
*** openstackgerrit has quit IRC17:45
*** openstackgerrit has joined #openstack-zaqar17:46
*** achanda has joined #openstack-zaqar18:11
*** malini1 has quit IRC18:26
*** dynarro has quit IRC18:38
*** ametts has quit IRC18:53
*** achanda has quit IRC18:54
*** achanda has joined #openstack-zaqar19:35
*** achanda has quit IRC20:15
*** achanda has joined #openstack-zaqar20:16
*** openstackgerrit has quit IRC20:17
*** openstackgerrit_ is now known as openstackgerrit20:17
*** openstackgerrit has quit IRC20:17
*** openstackgerrit_ has joined #openstack-zaqar20:17
*** openstackgerrit_ is now known as openstackgerrit20:18
*** openstackgerrit_ has joined #openstack-zaqar20:18
*** ametts has joined #openstack-zaqar20:49
openstackgerritMerged openstack/zaqar: Add queue name to notifications  https://review.openstack.org/27503020:57
openstackgerritMerged openstack/zaqar: Updated from global requirements  https://review.openstack.org/27876020:57
ryansb\o/ review-a-thon20:58
flwangmeeting?20:59
flwangthank you guys for waiting me20:59
flwangand sorry i can't make the 4:00 am meeting20:59
vkmcflwang, o/21:02
flwanghalllllo, anybody around?21:02
flwanglet's go to #openstack-meeting-321:03
ryansbhey folks21:07
ryansbI think there's another mtg in openstack-mmeting-3 at this time21:07
ryansbwe should stay in here21:07
flwangryansb: really?21:09
flwangseems no body there except us21:09
ryansbflwang: yeah, there was last time we inverted timezones anyways21:09
flwangvkmc: would you mind coming back to our channel?21:10
vkmcnot a problem21:10
vkmclet's have the meeting here21:11
flwangok, cool21:11
vkmc:)21:11
flwangok, not sure if we can still use the command21:12
flwangfor meetings21:12
flwang#startmeeting zaqar21:12
openstackMeeting started Mon Feb 15 21:12:18 2016 UTC and is due to finish in 60 minutes.  The chair is flwang. Information about MeetBot at http://wiki.debian.org/MeetBot.21:12
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.21:12
openstackThe meeting name has been set to 'zaqar'21:12
flwangoh, it works21:12
flwang#topic code review21:12
flwangwe did a great a great job for zaqar client, both coding and reviewing21:13
flwanguntil now, we almost filled all the function gaps21:13
ryansbI guess it does, neat21:13
flwangryansb: :)21:13
Eva-ihello21:14
ryansb#link https://www.openstack.org/summit/austin-2016/vote-for-speakers/presentation/714521:14
ryansb#link https://www.openstack.org/summit/austin-2016/vote-for-speakers/presentation/709621:14
*** sriram has quit IRC21:14
ryansb#link https://www.openstack.org/summit/austin-2016/vote-for-speakers/presentation/713921:14
ryansb^ all talks on zaqar for summit, for everbody's convenience21:15
*** malini has joined #openstack-zaqar21:15
flwangawesome21:16
flwangthanks ryansb collecting those links21:16
flwangi even don't know 714521:16
flwanglet's discuss them in next topic21:16
ryansbyeah, the speaker tweeted at me the other day21:16
flwangfor zaqar server side, we do still have some patches need to be reviewed21:17
flwangTTL subscription issue for mongoDB https://review.openstack.org/27046421:17
Eva-iryansb: thank you21:17
flwangTTL subscription issue for redis https://review.openstack.org/27654821:17
flwangfor the redis, it's still WIP. i'm still trying to figure out a way to fix it21:18
flwangand this one https://review.openstack.org/265723   more attributes for queue21:18
flwangbinary support for websocket https://review.openstack.org/25697821:18
flwangBTW, you guys maybe noticed, we have 2 new contributors from Huawei21:19
flwangwanghao and wxy21:19
Eva-iflwang: redis one is hard one, I'll try to review it21:19
Eva-iflwang: yes21:19
*** malini has quit IRC21:19
flwangpls give them more care :)  thanks21:19
ryansbyup, love having new faces about21:20
Eva-iflwang: oki21:21
flwanganything else for code review topic?21:21
Eva-inot from me21:21
ryansbnope - I haven't been writing zaqar code. Bad me.21:22
flwang#topic zaqar UI21:23
flwangwith the big support from horizon team, we made a great progress on zaqar UI21:23
flwangnow with the queues panel, you can list all the queues and create queue21:23
Eva-iflwang: I saw it, good job21:24
flwangupdate, delete will come soon21:24
flwangpls try it and let me know your feedback21:24
flwangsubscriptions panels is coming as well21:24
Eva-iflwang: I can only try, I can't understand the code written for Zaqar UI21:25
flwang#link https://review.openstack.org/#/q/project:openstack/zaqar-ui21:25
ryansb\o/21:25
flwang#link https://review.openstack.org/#/c/277257/ the subscriptions panel21:25
flwangsubscriptions panel is little bit tricky since zaqar needs the queue name to list subscriptions21:26
flwangso we need some ideas on this one21:26
flwanghow to show them21:26
flwangagain, pool and flavor won't come until Newton21:26
flwangin Mitaka, we will focus on the user part.   Sorry, operators21:27
openstackgerritMerged openstack/python-zaqarclient: Improve subscription listing  https://review.openstack.org/27290921:27
flwang#topic puppet and ansible21:28
flwangjasondotstar said he don't have much bandwidth on the puppet zaqar work21:28
flwangdprice will take over21:29
*** achanda has quit IRC21:29
flwangi haven't sync with dprice, but will do21:29
ryansb*dprince21:29
flwangryansb: sorry :(21:29
flwangdprince21:29
ryansbno worries21:29
flwangryansb: another redhater  IIRC21:30
ryansbhe is21:30
flwangbtw,  the ansible zaqar spec has been approved   https://review.openstack.org/#/c/269889/21:30
vkmcsweeeet21:30
flwangbut i haven't got time to work on that, since i don't know much about ansible :(21:30
ryansboh, neato. I happen to write lots of ansible :)21:31
flwangryansb: kidding me?21:31
flwangryansb: would you like to take it?21:31
ryansbI would :) I can't guarantee I'll get to it fast, but I'll try21:31
flwangagain, we(Catalyst IT) is very keen to deploy zaqar21:31
ryansbheh, so ansible would be handy21:31
flwangbut unfortunately, now there is no way21:32
flwangryansb: i will give you some example project21:32
flwangironic and designate ansible21:32
flwangryansb: awesome awesome awesome21:32
ryansbK, sounds good. Next topic then?21:34
Eva-iI have an idea about subscription listing in Zaqar-UI, maybe we can discuss it after the meeting21:35
flwangEva-i: cool, sure21:35
flwangi don't have much topics, TBH, unless you guys want to discuss the summit sessions and the design summit topics21:36
flwangnot sure if it's too early21:36
ryansbEva pasted a couple discussion items21:37
ryansbhttp://paste.openstack.org/show/niT4ouOOEhedv2gCEWZO/21:37
ryansband http://paste.openstack.org/show/kYFFmShi2eiouoYnUz5B/21:37
Eva-iryansb: yes, thank you21:37
Eva-ifirst paste about possible design summit topic/session21:38
flwangreading...21:40
vkmcregarding that one21:41
vkmcwe need to decouple wsgi transport from the API21:41
Eva-iyes21:41
vkmcright now websocket transport is independent from the API, but the wsgi transport is not21:41
ryansbyeah, that's not great. I think the one parameter I'd want on any solution is it can't change any of the existing stuff we expose21:42
flwangryansb: +121:43
vkmcI see that coupling as a bug, and it's something I'd really like to change from our current code21:43
Eva-iryansb: what do you mean?21:43
ryansbso the problem stated was that there's coupling between the WSGI & backends21:44
flwangvkmc: Eva-i: so both of you're talking about a refactor, right?21:45
vkmcflwang, yes21:45
Eva-iflwang: right21:45
ryansband I was just saying that I don't have an incredibly strong opinion on how to break it apart, just that I don't want to change anything we expose as-is over the API (which is part of the definition of refactor)21:45
flwangi see. then i would suggest register a blueprint to track it, personally, i would like to see a detailed plan21:46
Eva-iryansb: if we refactor things properly, the API change will be unnoticed by the user21:46
flwangand what's the part we should improve21:46
ryansbyup :) that's the idea. Anyways, I think a blueprint would be a good start, and we shouldn't dealy until summit21:46
ryansb*delay21:46
Eva-iI can write the blueprint21:46
flwangEva-i: cool. thanks21:47
ryansbEva-i++21:47
flwangi have another interesting topics if you guys still have time21:47
Eva-ithank you guys =)21:47
flwangqueue's metadata21:47
flwangas you know, we were trying to remove queue's metadata in v1.1 since we would like to make queue as lazy21:47
flwanghowever, finally, we decide to keep it21:48
vkmcEva-i++21:48
flwangnow, the problems are coming21:48
flwang1. there is no way to update queue's metadata since v1.121:48
Eva-ihm, I don't remember deciding to keep it21:49
Eva-ibut I'm a newbie21:49
flwangEva-i: i can't remember the details21:49
flwangbut that's the decision21:49
vkmcflaper87, ^ can you relate to this?21:49
flwangand as the code saying, we're keeping it21:49
flwangwe run into this issue when working on the zaqar UI21:50
*** mdnadeem has quit IRC21:50
flwangnow with the zaqar client v2, there is no way to create a queue with metadata21:50
*** mdnadeem has joined #openstack-zaqar21:50
flwangand worse, no way to update metadata after the queue created21:50
Eva-iflwang: even on API v1?21:52
flwangEva-i: no21:52
flwangfor v1, user can set the metadata21:52
vkmcv1.1 and v2 are the problematics21:53
flwang#link https://bugs.launchpad.net/python-zaqarclient/+bug/154390021:53
openstackLaunchpad bug 1543900 in zaqar "Can't update metadata after create queue for v1.1 and v2" [High,New] - Assigned to Fei Long Wang (flwang)21:53
flwangvkmc: ryansb: flaper87: i would like to know your comments if we should support metadata for v1.1 and v2 like v121:53
flwanggiven we're keeping it21:54
ryansbyeah, I think we should keep update support21:54
flwangand queue is very important for subscriptions21:54
ryansberr, support for updating queue metadata21:54
ryansbindeed it is21:54
vkmcI think it's useful and we should have a way to change it21:55
vkmcmetadata I mean21:55
vkmcI don't remember why we removed that21:55
flwangin the future, we may split notification from the messaging/queuing, but for now, we still need queue21:55
flwangvkmc: since the lazy queue21:55
flwangand we did some research from rackspace21:55
vkmcflwang, but why we didn't create queues with empty metadata and allowed update?21:55
vkmchmm21:55
flwangseems there are not too much user using metadata21:56
vkmcI see21:56
flwangbut finally, we got some feedback from operators (not sure), then we decide to keep metadata21:56
vkmcI see21:56
ryansbThis may not be public info, but how many is "not many"21:56
Eva-ihttps://blueprints.launchpad.net/zaqar/+spec/api-v1.1-remove-queue-metadata21:57
flwangryansb: i can't remember :(21:57
ryansbbecause "not many" of all the rackspace users may still be lots21:57
flwangBased on feedback at the summit, we decided as a team to leave metadata as-is. --kgriffs21:57
flwangsee the last line of the blueprint21:57
flwangso, I would say let's fill the gaps21:58
ryansb+121:58
vkmchey guys, gotta rush21:58
vkmcwill read the backlog later21:58
vkmco/21:58
ryansbkk, catch you later21:58
Eva-ioki21:58
flwangvkmc: ok, ttyl21:58
flwangso personally, i would like to support the metadata actions for v1.1 and v221:59
Eva-iokay, we may support queue metadata in API v1.1 and API v221:59
flwangthat means, we need to fix the metadata update bug on server side21:59
flwangand provide the metadata support for client side21:59
flwangryansb: vkmc: flaper87: how do you think?22:00
ryansbI think that's the way to go22:00
ryansbIMO metadata is a useful thing to have, and doesn't mean queues can't be lazy22:01
Eva-iryansb: right22:01
Eva-iryansb: I also wanted to ask your opinion about queue lazyness in subscriptions22:02
ryansbk, shoot22:02
flwangryansb: cool, thanks for the feedback22:02
flwangEva-i: what's the design? re queue lazyness in subscriptions22:03
Eva-iryansb: the problem is queues stop being lazy on operations with subscriptions22:03
flwangIMHO, subscribe a non-existing object is weird22:03
Eva-iryansb: though in API v2 queues are considered to be lazy22:03
Eva-iryansb: maybe we should make it possible to subscribe to unexisting queue22:04
ryansbhm. Good thought. Let's get into a user's mind first22:05
ryansbso if you're a user, and you can send messages to queues that don't exist22:05
ryansbwithout creating them first22:05
ryansbthen you, the user, would probably think you were also able to subscribe to a queue that doesn't yet exist22:06
ryansbdoes that make sense to you?22:06
*** flwang1 has joined #openstack-zaqar22:06
Eva-ifor me it makes sense22:07
flwangryansb: for that case, i would agree only if the queue is crated in background just like posting messages22:07
flwanginstead of created a subscription on a non-existing queue22:07
ryansbyeah, for messages we're actually creating queues in the background22:08
Eva-iqueues are topics22:08
ryansbso subs could be the same22:08
ryansbflwang: That's what you mean, right?22:08
flwangqueue is not existing before creating the subscription, but after created the subscriptions, the queue should be created already22:08
flwangryansb: yes22:08
Eva-iyou can say to Zaqar: I would to listen about this topic. If there will be someting, please notify me.22:08
Eva-i*would like to22:08
flwangryansb: but22:09
flwangit's still a little bit weird for me, TBH22:09
flwangsome users may like it22:09
ryansbhm, which users wouldn'22:09
ryansb*wouldn't like it22:09
Eva-ifor me it's not weird22:09
flwangryansb: haha, me :)22:09
flwangryansb: because22:10
ryansbwell what *type* of user? flwang isn't a type ;)22:10
flwangi maybe wrong22:10
flwangas a user, i create a subscription with a non-existing queue name 'horizon'22:11
flwangthen there is another user in the same tenant22:11
flwanghe may post messages on 'horizon' before creating it22:11
flwangi haven't think it very clearly22:12
flwangbut i'm just wondering if there is a case like above, user A may get messages from user B who don't want his message spread unintentionally22:13
flwangeven though they're in same tenant22:13
ryansbif they want isolation, shouldn't they be in different tenants?22:13
ryansbI mean, I feel like that's what tenants/projects are for22:14
*** AAzza_ has joined #openstack-zaqar22:14
flwangryansb: i see22:14
Eva-ihm, I'll think about it.22:15
flwangEva-i: could you please register a bp and propose a spec?22:15
flwangEva-i: generally, i think it's a good idea22:15
flwangbut i just need some more thoughts22:15
Eva-iflwang: okay, I can make it. I'll invite flaper87 to spec review =)22:15
flwangmaybe i'm too worry22:15
flwangEva-i: cool22:16
flwangEva-i: thanks22:16
Eva-iflwang: thank you too =)22:16
*** AAzza has quit IRC22:17
*** AAzza_ is now known as AAzza22:17
flwangryansb: as for the summit sessions, how can i see the vote status?22:17
ryansbI don't think we get to know until the end22:17
flwangryansb: ok, i see22:18
flwangryansb: i will spread them into China openstack chat groups22:18
ryansb:) Thank you22:18
flwangthank you!22:19
ryansbI think that covers all our topics, yes?22:20
Eva-iI don't have anything else to discuss now22:20
*** AAzza has quit IRC22:21
Eva-imaybe just an idea for Zaqar-UI subscription listing22:21
flwangryansb: yep, thank you guys22:22
flwangEva-i: what's the idea?22:22
ryansbflwang: remember to #endmeeting22:22
flwang#endmeeting22:22
openstackMeeting ended Mon Feb 15 22:22:40 2016 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)22:22
openstackMinutes:        http://eavesdrop.openstack.org/meetings/zaqar/2016/zaqar.2016-02-15-21.12.html22:22
Eva-iI think subscriptions panel must have an interactive text field for subscription listing. After user stops typing into this text field, after 1 second a request to Zaqar will be made to get list of subscriptions.22:22
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/zaqar/2016/zaqar.2016-02-15-21.12.txt22:22
openstackLog:            http://eavesdrop.openstack.org/meetings/zaqar/2016/zaqar.2016-02-15-21.12.log.html22:22
flwangryansb: hah22:22
Eva-iflwang: ^^^22:23
Eva-iflwang: a text field with button will be fine too, but not as convenient22:23
Eva-iby text field I meant the place where the user inputs queue name22:24
ryansbso you're thinking like a progressive search?22:25
Eva-iryansb: progressive search is the one that gives suggestions? No, it's much harder to make.22:26
ryansbI mean the kind that you type a bit, then it checks and shows options that match, then you type more and it shows more options22:27
*** AAzza has joined #openstack-zaqar22:27
ryansbso yeah, that would be harder. I see you're just thinking of auto-running the search after the user finishes typing.22:27
Eva-iryansb: yes, the last one22:27
flwangEva-i: it's fancy, but maybe not the best way for now22:28
flwangEva-i: for now, we need a simple, clean way to show the data22:28
flwangsince it's the first version22:28
Eva-iflwang: okay. So you will probably do a queue name field and button "List"?22:28
flwangso the simpler the better and we can enhance it later22:28
flwangEva-i: i don't know :D22:28
flwangmaybe just a dropdown list22:29
Eva-iflwang: that's how I see the most simple way22:29
*** achanda has joined #openstack-zaqar22:30
Eva-iOh, I haven't thinked about dropdown list. Maybe it's a good idea. But it might be not good if there are too many queues.22:30
ryansbyeah, dropdown only works to a certain size22:30
ryansbok, I've gotta head off to make dinner - see you folks tomorrow. flwang I'll take a look at the openstack-ansible spec too :)22:34
Eva-iI'm gonna review one patch and then sleep22:34
Eva-iryansb: see you22:34
*** mpanetta has quit IRC22:35
flwang1ryansb: ttyl22:42
flwang1ryansb: i will send you links about ansible22:42
*** rcernin has quit IRC23:05
*** achanda has quit IRC23:25
*** achanda has joined #openstack-zaqar23:33
*** csoukup has quit IRC23:35
*** ametts has quit IRC23:43

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!