Friday, 2016-03-18

« Thursday, 2016-03-17 Index Saturday, 2016-03-19 »
*** wanghao has quit IRC00:04
*** wanghao has joined #openstack-zaqar00:05
*** kgriffs|afk is now known as kgriffs00:08
*** JAHoagie has quit IRC00:08
*** eliqiao_ has joined #openstack-zaqar00:08
*** eliqiao has quit IRC00:11
openstackgerritFei Long Wang proposed openstack/zaqar: Fix PATCH queue's metadata  https://review.openstack.org/29284300:12
flwang1Eva-i: pls revisit it00:13
Eva-iflwang1: yes00:13
*** davideagnello has joined #openstack-zaqar00:13
flwang1ryansb: still around?00:15
*** tqtran-afk has joined #openstack-zaqar00:17
flwang1currently, my feeling is helpless :(00:19
Eva-iflwang1: why?00:20
Eva-iflwang1: are you alright?00:20
flwang1bad, we need core reviewer for those patches00:21
flwang1so that we won't miss RC100:21
flwang1today is the last day00:21
*** tqtran-afk has quit IRC00:21
Eva-iflwang1: we can try our irregular core reviewers00:22
Eva-i=)00:22
flwang1what's the irregular core?00:22
Eva-ilike vkmc, flaper8700:23
flwang1vkmc: flaper87: if you can see this, pls pls pls review our patches00:23
flwang1especially this one https://review.openstack.org/29284300:23
flwang1ryansb: ^00:24
flwang1i'm going to nijia approve this one https://review.openstack.org/#/c/293810/00:24
Eva-iflwang1: sure00:24
flwang1Eva-i: https://review.openstack.org/#/c/290923/00:26
flwang1can you give it a quick review?00:26
*** Eva-i has quit IRC00:32
openstackgerritMerged openstack/zaqar: Remove unnecessary assert  https://review.openstack.org/29381000:40
*** wxy has joined #openstack-zaqar00:55
openstackgerritFei Long Wang proposed openstack/zaqar: Fix PATCH queue's metadata  https://review.openstack.org/29284301:02
*** boris-42 has quit IRC01:04
*** Eva-i has joined #openstack-zaqar01:06
Eva-iflwang1: sorry for delay, crap is happening with my laptop01:08
flwang1Eva-i: no worries01:09
flwang1thank you for staying with me at this moment :)01:09
openstackgerritMerged openstack/zaqar: Add missing /ping for v1.1 homedoc  https://review.openstack.org/29092301:28
wxyI'm writing the PATCH code for client side now. I hope the server side's patch could be merged today.01:38
*** achanda has joined #openstack-zaqar01:40
Eva-iflwang1: wxy: found some problems in PATCH patch, trying to triage one of them01:45
Eva-iwxy: that's very nice01:45
flwang1Eva-i: what's the problem?01:46
Eva-iflwang1: I'll describe it in review now01:48
flwang1major issue?01:48
flwang1or minor?01:48
*** csoukup has joined #openstack-zaqar01:53
Eva-iflwang1: one major - 500 error01:57
*** csoukup has quit IRC01:57
Eva-iflwang1: and we have to decide if queues should be lazy or not in PATCH. This is a major too, I guess.01:57
Eva-iflwang1: I commented your patch.01:57
Eva-iflwang1: the patch is hard... I'm trying as I can.01:59
wxyInstead of rewrite "set metadata" in v2, I think it's better to add three new commands : "add/replace/delete metadata". What do you think, guys?02:06
*** tqtran-afk has joined #openstack-zaqar02:18
*** tqtran-afk has quit IRC02:22
*** Eva-i has left #openstack-zaqar02:23
*** Eva-i has joined #openstack-zaqar02:26
*** tqtran-afk has joined #openstack-zaqar02:29
*** tqtran-afk has quit IRC02:29
openstackgerritwangxiyuan proposed openstack/python-zaqarclient: Add update queue function in v2  https://review.openstack.org/29436802:40
*** eliqiao has joined #openstack-zaqar02:45
*** eliqiao_ has quit IRC02:47
flwang1wxy: if we can  i would recommend avoid adding those new three commands02:59
flwang1since the name will be really confused02:59
flwang1Eva-i: still around?02:59
Eva-iwxy: I'm not sure about forcing the user to use update commands, it will be a breaking change. But I know that our metadata code is complicated now (too many special conditions) and adding separate metadata method to v2 seems nice, just without breaking compatibility.03:03
Eva-iflwang1: yes, I'm here03:03
flwang1Eva-i: as for your question about lazy queue03:03
flwang1for patch, i will say no, let's throw 404 if the queue is not existing03:03
flwang1since it's not the case lazy queue should cover03:04
Eva-iflwang1: oki, I don't know the right answer, so maybe I should rely on your wisdom. So let's make it throw 404.03:05
flwang1Eva-i: cool, thanks,03:06
flwang1Eva-i: and i'm confused for your comments on https://review.openstack.org/#/c/292843/11/zaqar/transport/wsgi/v2_0/queues.py03:09
flwang1you said "This method throws DoesNotExist exception, but our queues are lazy. Is  it okay for PATCH method? Maybe it's better to pre-create queue? Please  decide."03:09
flwang1and then you said "When queue does not exist, this message is shown. Maybe it's okay,  because our queues are lazy. But this is PATCH, maybe they must not  behave lazy in PATCH. I don't know. Please decide."03:09
Eva-iflwang1: yes, here in chat we decided that queues should not act as lazy on PATCH.03:10
flwang1i don't think the 2nd comment can happen, does it?03:10
flwang1at line 17203:10
flwang1zaqar will try to get the metadata of queue, that means if the queue is not eixsting03:11
flwang1the following code shouldn't be called03:11
flwang1no?03:11
Eva-iflwang1: for some reason, when I was testing replace and delete PATCH on unexisting queue, L172 didn't throw 404 DoesNotExist error...03:12
Eva-iflwang1: you can try it yourself03:12
flwang1ok, let me try03:13
flwang1i can're reproduce it03:14
Eva-iflwang1: so L172 throws DoesNotExist error for you?03:15
flwang1yes03:16
flwang1as it should03:16
Eva-iI'll try to paste reproduce manual. What I do.03:16
flwang1i'm going to submit another patchset03:17
flwang1so let me know if you find something03:17
Eva-iflwang1: okay, I'll try to reproduce with your fresh patch set03:18
openstackgerritFei Long Wang proposed openstack/zaqar: Fix PATCH queue's metadata  https://review.openstack.org/29284303:21
flwang1i will be off after 10-15 mins and be back after 4-5 hours03:21
flwang1we need to get this one in today03:21
Eva-iflwang1: you're right... I can't reproduce now :o03:21
Eva-iflwang1: oki03:21
flwang1so feel free post a new patch set if you think it's right03:21
*** eliqiao_ has joined #openstack-zaqar03:21
Eva-iflwang1: yes03:22
flwang1Eva-i: thanks03:22
flwang1wxy: wanghao: pls leave your comments on this one https://review.openstack.org/29270203:22
*** eliqiao has quit IRC03:24
openstackgerritwangxiyuan proposed openstack/python-zaqarclient: Add update queue function in v2  https://review.openstack.org/29436803:30
*** flwang has quit IRC03:31
Eva-iflwang1: strange, sometimes metadata = self._queue_controller.get(queue_name, project=project_id) throws an exception. Now I can reproduce the problem again.03:36
*** eliqiao_ is now known as eliqiao03:41
Eva-isometimes I can reproduce it, sometimes not...03:42
Eva-i=(03:43
Eva-ioh, figured it out03:45
Eva-iflwang1: good night, I'll fix everything03:48
openstackgerritwangxiyuan proposed openstack/python-zaqarclient: Add update queue function in v2  https://review.openstack.org/29436803:54
openstackgerritwangxiyuan proposed openstack/python-zaqarclient: Add update queue function in v2  https://review.openstack.org/29436803:59
*** shu-mutou has joined #openstack-zaqar04:07
openstackgerritEva Balycheva proposed openstack/zaqar: Fix PATCH queue's metadata  https://review.openstack.org/29284304:11
Eva-iwxy: thank you for today. I'm going to sleep soon. I'll check your patches later. See you!04:11
wxyEva-i: Ok, Good night. :)04:12
wxyEva-i: flwang1: I'll update the test code after the server side's patch is merged. Please review the functional code first.:)04:15
openstackgerritwangxiyuan proposed openstack/python-zaqarclient: Add update queue function in v2  https://review.openstack.org/29436804:16
*** shu-mutou has quit IRC04:44
*** achanda has quit IRC05:01
*** GB21 has joined #openstack-zaqar05:16
*** itisha has quit IRC05:19
*** shu-mutou-AFK has joined #openstack-zaqar05:26
*** shu-mutou-AFK has quit IRC05:32
*** pcaruana has quit IRC06:18
*** rcernin has joined #openstack-zaqar06:37
*** GB21 has quit IRC07:25
*** achanda has joined #openstack-zaqar07:44
*** kgriffs is now known as kgriffs|afk07:45
*** kgriffs|afk is now known as kgriffs07:45
*** achanda has quit IRC07:48
*** pcaruana has joined #openstack-zaqar07:57
*** shu-mutou has joined #openstack-zaqar07:58
*** flwang has joined #openstack-zaqar08:02
flwangEva-i: thanks a lot for the patchset08:03
flwangit's a good improvement08:03
flwangvkmc: flaper87: ryansb: pls review https://review.openstack.org/#/c/292843 we need to get it in today08:04
*** GB21 has joined #openstack-zaqar08:07
*** flwang has quit IRC08:18
*** tesseract has joined #openstack-zaqar08:20
*** tesseract is now known as Guest5995408:21
*** Guest59954 is now known as new08:25
*** new has quit IRC08:26
*** tesseract- has joined #openstack-zaqar08:26
*** GB21 has quit IRC08:27
*** GB21 has joined #openstack-zaqar08:27
*** flaper87 has quit IRC08:33
*** flaper87 has joined #openstack-zaqar08:34
*** flwang has joined #openstack-zaqar08:41
*** GB21 has quit IRC08:45
*** GB21 has joined #openstack-zaqar08:45
*** GB21 has quit IRC08:55
*** flwang has quit IRC09:00
*** JAHoagie has joined #openstack-zaqar09:02
*** AAzza_ has joined #openstack-zaqar09:06
*** JAHoagie has quit IRC09:06
*** AAzza has quit IRC09:08
*** AAzza has joined #openstack-zaqar09:10
*** AAzza_ has quit IRC09:12
openstackgerritShu Muto proposed openstack/zaqar-ui: Add DevStack plugin settings  https://review.openstack.org/29447109:20
*** GB21 has joined #openstack-zaqar09:23
*** flwang has joined #openstack-zaqar09:48
flwanganybody around?09:49
openstackgerritShu Muto proposed openstack/zaqar-ui: Add DevStack plugin settings  https://review.openstack.org/29447109:54
*** JAHoagie has joined #openstack-zaqar10:02
*** AAzza has quit IRC10:02
*** JAHoagie has quit IRC10:06
*** GB21 has quit IRC10:17
*** AAzza has joined #openstack-zaqar10:19
*** shu-mutou is now known as shu-mutou-AFK10:34
*** flaper87 has quit IRC10:49
*** flaper87 has joined #openstack-zaqar10:49
openstackgerritMerged openstack/zaqar: Fix PATCH queue's metadata  https://review.openstack.org/29284310:51
*** JAHoagie has joined #openstack-zaqar11:03
flwangthanks flaper8711:06
flwangflaper87: would you mind leaving your comments on this https://review.openstack.org/292702 ?11:06
*** JAHoagie has quit IRC11:07
flwangflaper87: vkmc: ryansb: let's get this tiny one merged https://review.openstack.org/#/c/292699/11:08
flaper87flwang: looking11:11
flwangflaper87: i love you, Fla11:12
flaper87flwang: <3 u too11:13
flaper87flwang: https://review.openstack.org/#/c/292702/ <- this does not prevent users from setting private keys11:15
flaper87It just verifies the keys values11:16
flaper87Should we prevent v1 from setting private keys altogether?11:16
flaper87I think it'd be better if we do, tbh11:17
flaper87but that could be seen as a breaking change11:17
flaper87:/11:17
flaper87mmhh11:17
flaper87but I guess it's still fine to do so11:18
flaper87flwang: commented11:20
flwangflaper87: cool, do you think it's really necessary for Mitaka RC1?11:28
flwangflaper87: i'm going to release RC111:28
flaper87flwang: we can fix it in RC2, I guess. Thing is, if we let this go as is, we can't just go and disable all `_` later11:29
Eva-iflaper87: flwang: hello! I'm going to read your comments11:30
flwangflaper87: maybe i'm wrong, but i'm hesitating to add change for old version because a new feature in current version11:31
flwangflaper87: thoughts?11:31
flaper87flwang: well, we're basically doing that now, aren't we?11:32
flaper87I mean, you're preventing v1 to use `_max_.._..`11:32
flaper87s/you're/we're/11:32
flaper87so, instead of forbidding v1 to set just 1 key, I think we should be consistent and reserve a set of keys11:32
flwangflaper87: we're not, i think. v1 still can set it. but it means nothing for v111:32
flaper87mmmh, fair point11:33
flaper87but we're validating it11:33
flaper87it just feels weird that we're validating just one key11:33
flwangflaper87: the only problem is11:33
flaper87We should probably reserve all keys that start with "$."11:33
flwangif user set a wrong value with v1, like give '_max_xxx' a string11:34
flaper87right, and that's not really nice11:34
flwangand using v2 to posting messages, then they will see something wrong11:34
flaper87So, what do you think about my proposal?11:34
flaper87We change the prefix for private keys to `$.`11:34
Eva-ibecause our APIs share the same database and both have instruments to edit metadata, and metadata in v2 is used in logic, we can't just validate metadata only in API v2.11:34
flaper87and we forbid them altogether11:34
flaper87Eva-i: the problem is clear, the solution is the one I think is not the most ideal one from an user perspective11:35
flwangflaper87: personally, i don't think we should forbid all private keys for v1, given v1 is deprecated and v1.1 is going to be deprecated11:36
openstackgerritMerged openstack/zaqar: Fix cleanup in metadata putting test  https://review.openstack.org/29269911:36
flwangi can't see any reason why we would like to change a deprecated api version11:36
Eva-iflaper87: it's more breaking to forbid all private keys with prefix $. I see nothing wrong in checking each one reserved keys. They can be different and require special validation. They're not too many.11:36
Eva-i*each one of reserved keys11:37
flaper87My only issue with this change is that it does not provide a good experience for the user. We're preventing the user from doing whatever with the `_max_..._..` key11:37
flaper87That said, I think I don't care much as both the API's should be turned off in Newton11:37
* flaper87 is working on a patch for that already11:37
flwangflaper87: oh, you're working on a patch to deprecate v1.1?11:38
flwangflaper87: if so, i prefer to hold this patch, thoughts?11:39
flaper87well, to officially deprecate v1 and put both v1 and v1.1 behind a config option that is off by default11:39
flwangflaper87: that's what i want to see11:39
flaper87flwang: what do you think about we introduce the warning for deprecation in RC2 ?11:39
flwangsince for now, operator can't disable v1 and v1.111:39
flaper87and we turn it off by default in newton ?11:39
flwangflaper87: i like the idea11:40
* flaper87 gets that done11:40
flwangso let's hold this one?11:40
flaper87yup11:41
flaper87let's wait until RC211:41
flwangcool, thanks for the feedback11:42
flaper87Anytime, Boss!11:43
*** JAHoagie has joined #openstack-zaqar12:02
openstackgerritFlavio Percoco proposed openstack/zaqar: Warn on upcoming deprecations for v1.0 and v1.1  https://review.openstack.org/29454512:04
flaper87flwang: ^12:05
flwangflaper87: lovely12:05
flaper87Now I'll add the config option12:05
flaper87but that one is for Newton12:05
flwangflaper87: cool, thanks12:07
*** JAHoagie has quit IRC12:07
flwangflaper87: do you think we need some work like this https://review.openstack.org/#/c/50491/ ?12:07
flaper87flwang: what do you mean? In the commit message?12:08
flaper87oh crap, I should add release notes for that12:08
flaper87or not12:08
flaper87it's just a warning12:08
flaper87not a deprecation12:09
flwangflaper87: i mean something like this https://review.openstack.org/#/c/50486/14/keystone/common/utils.py12:09
flwanga more formal deprecation12:09
flaper87flwang: I'm planning to do that in the next patch for Newton. I wanted to use debtcollator but it's too late to add a new depdendency12:11
flaper87It's fine for now, we can add that later and re-use the info we have in `VERSION`12:11
flaper87Planning to work on a good decorator for that12:11
flwangflaper87: oh, yep, the work definitely for newton, not rc212:11
flwangflaper87: cool, glad to see we're on the same page. i think we need the same thing for glance to deprecate v112:12
flaper87flwang: ah ok, then yes! I'm actually already working on that12:12
flaper87:D12:12
flwangif we can get nova use v212:12
flaper87w000h000012:17
flaper87RC1 is out12:17
flwangyeah!12:17
flwangi can go to sleep now12:17
flaper87flwang: NO!12:18
flaper87you stay there12:18
flwang:-(12:18
* flaper87 ties flwang to his desk12:18
flaper87ok, ok ok!12:18
flaper87go to sleep12:18
flwangflaper87: you can't know how i'm really happy to see you're around again12:19
flaper87I'm happy to be back :)12:19
flaper87and to have you as PTL12:19
flaper87so I can bother you12:19
flaper87annoy you12:19
flaper87and make your life impossible12:19
flaper87:D12:19
flaper87ok ok12:19
flaper87I won't do that12:19
flaper87just keep the gummybears coming12:20
flwangflaper87: i will return the favor when you're the PTL next time12:20
flaper87:P12:20
flwangok, i have to offline now12:20
flwangsee u later, man12:21
Eva-iflwang: good night12:21
Eva-iflaper87: thank you for your help12:21
flwangEva-i: thanks for all your hard work12:21
openstackgerritDoug Hellmann proposed openstack/zaqar: Update reno for stable/mitaka  https://review.openstack.org/29455112:21
flaper87Eva-i: np :)12:21
openstackgerritDoug Hellmann proposed openstack/zaqar: Update reno for stable/mitaka  https://review.openstack.org/29455112:22
*** openstackgerrit has quit IRC12:48
*** openstackgerrit has joined #openstack-zaqar12:48
*** csoukup has joined #openstack-zaqar12:56
*** tesseract- has quit IRC12:59
*** JAHoagie has joined #openstack-zaqar13:02
*** JAHoagie has quit IRC13:06
*** tesseract has joined #openstack-zaqar13:08
*** tesseract is now known as Guest8468813:08
*** itisha has joined #openstack-zaqar13:13
*** JAHoagie has joined #openstack-zaqar13:20
*** JAHoagie has quit IRC13:25
*** ametts has joined #openstack-zaqar13:37
*** JAHoagie has joined #openstack-zaqar13:38
*** amitgandhinz has joined #openstack-zaqar13:50
*** rcernin has quit IRC14:01
openstackgerritFlavio Percoco proposed openstack/zaqar: Disable DEPRECATED API versions by default  https://review.openstack.org/29461114:16
*** JAHoagie has quit IRC14:19
*** achanda has joined #openstack-zaqar14:34
*** pcaruana has quit IRC14:38
*** achanda has quit IRC14:47
*** Guest84688 has quit IRC14:48
*** ametts has quit IRC14:57
*** tesseract has joined #openstack-zaqar15:00
*** tesseract is now known as Guest7688315:00
*** david-lyle has quit IRC15:32
*** david-lyle has joined #openstack-zaqar15:33
*** amitgandhinz has quit IRC15:49
*** david-lyle has quit IRC16:00
*** david-lyle has joined #openstack-zaqar16:01
*** amitgandhinz has joined #openstack-zaqar16:21
*** amitgandhinz has quit IRC16:26
*** kgriffs is now known as kgriffs|afk16:44
*** pcaruana has joined #openstack-zaqar16:47
*** achanda has joined #openstack-zaqar17:03
*** amitgandhinz has joined #openstack-zaqar17:15
*** amitgandhinz has quit IRC17:21
*** tqtran-afk has joined #openstack-zaqar17:21
*** achanda has quit IRC17:25
*** amitgandhinz has joined #openstack-zaqar17:29
*** david-lyle_ has joined #openstack-zaqar17:44
*** david-lyle has quit IRC17:45
*** david-lyle_ is now known as david-lyle17:46
*** itisha has quit IRC17:59
*** Guest76883 has quit IRC18:15
*** itisha has joined #openstack-zaqar18:29
*** amitgandhinz has quit IRC19:02
*** david-lyle has quit IRC21:06
*** david-lyle has joined #openstack-zaqar21:07
*** ryansb_ has joined #openstack-zaqar21:24
*** ryansb_ has quit IRC21:24
*** ryansb_ has joined #openstack-zaqar21:24
*** ryansb has quit IRC21:25
*** ryansb_ is now known as ryansb21:25
*** dhellmann has quit IRC21:26
*** dhellmann has joined #openstack-zaqar21:26
openstackgerritMerged openstack/zaqar: Update reno for stable/mitaka  https://review.openstack.org/29455121:31
*** csoukup has quit IRC22:28
*** jtomasek has quit IRC23:09
*** tqtran-afk has quit IRC23:35
*** akanksha_ has joined #openstack-zaqar23:54
« Thursday, 2016-03-17 Index Saturday, 2016-03-19 »

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