Tuesday, 2017-04-11

*** gcb has quit IRC00:03
*** Administrator_ has quit IRC00:04
*** Administrator_ has joined #openstack-manila00:05
*** markstur has joined #openstack-manila00:05
*** markstur has quit IRC00:10
*** catintheroof has quit IRC00:19
*** mtanino_ has joined #openstack-manila00:23
*** mtanino has quit IRC00:25
*** cdelatte has joined #openstack-manila00:30
*** cdelatte has quit IRC00:30
*** markstur has joined #openstack-manila00:30
*** hongbin has quit IRC00:36
*** kaisers has joined #openstack-manila00:52
*** kaisers has quit IRC00:57
*** gcb has joined #openstack-manila01:16
*** mtanino_ has quit IRC02:11
*** tuanluong has joined #openstack-manila02:15
*** catintheroof has joined #openstack-manila02:16
*** catintheroof has quit IRC02:34
*** catintheroof has joined #openstack-manila02:35
*** catintheroof has quit IRC02:40
*** markstur has quit IRC02:46
*** chenying_ has quit IRC03:36
*** chlong has quit IRC03:47
*** gouthamr has quit IRC03:51
*** kaisers has joined #openstack-manila03:54
*** kaisers has quit IRC03:58
*** kaisers has joined #openstack-manila04:00
*** kaisers has quit IRC04:03
*** kaisers has joined #openstack-manila04:04
*** arnewiebalck_ has joined #openstack-manila04:08
*** kaisers has quit IRC04:09
*** arnewiebalck_ has quit IRC05:00
*** arnewiebalck_ has joined #openstack-manila05:01
*** kaisers has joined #openstack-manila05:05
*** kaisers has quit IRC05:09
*** rraja has joined #openstack-manila05:14
*** jprovazn has joined #openstack-manila05:22
*** arnewiebalck_ has quit IRC05:27
*** tuanluong has quit IRC05:33
*** tuanluong has joined #openstack-manila05:58
*** arnewiebalck_ has joined #openstack-manila06:02
*** markstur has joined #openstack-manila06:05
*** kaisers has joined #openstack-manila06:06
*** markstur has quit IRC06:10
*** mkoderer_ has joined #openstack-manila06:13
openstackgerritM V P Nitesh proposed openstack/manila master: Optimize the link address  https://review.openstack.org/45552306:15
*** mkoderer_ has quit IRC06:15
openstackgerritM V P Nitesh proposed openstack/manila-image-elements master: Optimize the link address  https://review.openstack.org/45552506:21
openstackgerritM V P Nitesh proposed openstack/manila-specs master: Optimize the link address  https://review.openstack.org/45552906:27
jprovazntommylikehu, good afternoon06:44
openstackgerritM V P Nitesh proposed openstack/manila-ui master: Optimize the link address  https://review.openstack.org/45554506:44
jprovazntommylikehu, I wonder if you be around later today when goutham is online - we could discuss event_id column to speed up things06:45
jprovazns/you be/you will be/06:45
openstackgerritM V P Nitesh proposed openstack/python-manilaclient master: Optimize the link address  https://review.openstack.org/45554706:48
tommylikehuhey jprovazn  I am here07:01
tommylikehuwould be online before 10pm utc+8, I think goutham is more like the first version in cinder.07:03
*** pcaruana has joined #openstack-manila07:06
openstackgerritzhongjun proposed openstack/manila-specs master: Add spec for infinite share  https://review.openstack.org/45209707:16
*** chenying has joined #openstack-manila07:20
jprovazntommylikehu, hi, if you will be online before 10pm utc+8, goutham should be around at that time07:21
jprovazntommylikehu, not sure what you mean by "goutham is more like the first version in cinder"?07:21
tommylikehugoutham is more like the implementation we already have in cinder :)07:24
tommylikehujprovazn:  he doesn't think we would have that much defined event_ids in manila, so we don't need to add the autpgenerated event_ids07:25
jprovazntommylikehu, my opinion regarding event_id is quite neutral - I prefer to align implementation in cinder and manila. I can see it beneficial as it provides the constant of detail field, on the other side I agree that if this detail constant would be returned thorugh api (so we return all parts from which event_id is generated), then there is no reason to have this event_id present and client can easily categorize messages hims07:29
jprovaznelf07:29
tommylikehujprovazn: yes, I think the true issue behind the auto-generated decision  is whether we should use event_id or action+detail to describe what happened inside07:33
tommylikehuIf gouthamr has a strong opinion on that, I think maybe we could take the same step as cinder did, because it's easy to change or optimise in the future,07:35
jprovazntommylikehu, well, for manila spec gouthamr argued against presence of event_id, use of action+detail is probably fine (at least based on spec comments)07:37
jprovazntommylikehu, yes, if it turns out event_id is needed/wanted, it can be added in a next step. So if you are fine with not having it in cinder spec, I'll remove it from manila spec :)07:38
tommylikehuI am surely ok with that, thanks07:39
jprovazntommylikehu, cool07:39
openstackgerritJan Provaznik proposed openstack/manila-specs master: Add spec for user messages  https://review.openstack.org/43427707:47
openstackgerritOpenStack Proposal Bot proposed openstack/manila-ui master: Imported Translations from Zanata  https://review.openstack.org/45556707:54
*** ianychoi has quit IRC07:54
*** ianychoi has joined #openstack-manila07:54
*** a-pugachev has joined #openstack-manila07:55
*** ianychoi has quit IRC08:20
*** ianychoi has joined #openstack-manila08:21
*** ianychoi has quit IRC08:35
*** swapnil-nilangek has joined #openstack-manila08:37
*** ianychoi has joined #openstack-manila08:38
*** arnewiebalck_ has quit IRC08:45
*** kaisers has quit IRC08:50
zhongjun_bswartz: ping08:51
*** openstackgerrit has quit IRC09:03
*** lpetrut has joined #openstack-manila09:04
*** ociuhandu has quit IRC09:05
*** tinwood has quit IRC09:18
*** openstackgerrit has joined #openstack-manila09:27
openstackgerritzhongjun proposed openstack/manila-specs master: Enhance update shares  https://review.openstack.org/45355309:27
openstackgerritzhongjun proposed openstack/manila-specs master: Add spec for infinite share  https://review.openstack.org/45209709:28
openstackgerritzhongjun proposed openstack/manila-specs master: Enhance update shares  https://review.openstack.org/45355309:32
*** tinwood has joined #openstack-manila09:35
*** markstur has joined #openstack-manila09:42
*** arnewiebalck_ has joined #openstack-manila09:44
*** kaisers has joined #openstack-manila09:46
*** markstur has quit IRC09:46
*** belmoreira has joined #openstack-manila09:47
*** ociuhandu has joined #openstack-manila09:57
*** arnewiebalck_ has quit IRC10:00
*** tuanluong_ has joined #openstack-manila10:02
*** rraja has quit IRC10:03
*** tuanluong has quit IRC10:05
*** swapnil-nilangek has quit IRC10:09
*** swapnil-nilangek has joined #openstack-manila10:13
*** makowals_ has quit IRC10:32
*** ianychoi has quit IRC10:36
*** ianychoi has joined #openstack-manila10:38
*** kaisers has quit IRC10:48
*** dsariel has joined #openstack-manila10:48
*** swapnil-nilangek has quit IRC10:56
*** ganso has joined #openstack-manila11:14
*** rraja has joined #openstack-manila11:26
*** kaisers has joined #openstack-manila11:43
*** tuanluong_ has quit IRC12:08
*** dsariel has quit IRC12:08
*** dsariel has joined #openstack-manila12:11
*** gouthamr has joined #openstack-manila12:14
*** dustins has joined #openstack-manila12:17
openstackgerritValeriy Ponomaryov proposed openstack/manila master: [WIP] Add quotas per share type support  https://review.openstack.org/45215812:18
*** ianychoi has quit IRC12:21
zhongjun_vponomaryov: ping12:24
*** ianychoi has joined #openstack-manila12:26
zhongjun_vponomaryov: I only have two goals in two specs. 1. change logic of "ensure share" in more efficient way compared to current one.(need restart) 2.  update share info without restart in enhance-ensure-share. I split of one spec to two, becuase of we wants spec goal more clear in one spec:  please see this link. https://review.openstack.org/#/c/446494/13/specs/pike/ensure-share.rst@14   We could discuss with bswartz.12:35
*** scottda has joined #openstack-manila12:45
*** rraja has quit IRC12:59
vponomaryovzhongjun_: pong13:10
*** rraja has joined #openstack-manila13:16
vponomaryovzhongjun_: second statement says "This spec should not refer to ensure_share", and you refer to it in both specs13:17
vponomaryovzhongjun_: so, split is not correct now considering this comment13:18
*** erlon has joined #openstack-manila13:24
*** timcl1 has joined #openstack-manila13:31
*** porrua has joined #openstack-manila13:31
*** timcl has quit IRC13:34
*** eharney has joined #openstack-manila13:34
*** lpetrut has quit IRC13:53
*** kaisers has quit IRC14:00
*** xyang_ has joined #openstack-manila14:00
*** ianychoi has quit IRC14:05
*** ianychoi has joined #openstack-manila14:06
*** timcl1 has quit IRC14:11
*** lpetrut has joined #openstack-manila14:11
*** Zhongjun__ has joined #openstack-manila14:15
openstackgerritValeriy Ponomaryov proposed openstack/manila-specs master: Add spec for share_groups quota resource  https://review.openstack.org/45227014:16
openstackgerritValeriy Ponomaryov proposed openstack/manila-specs master: Add spec for share_groups quota resource  https://review.openstack.org/45227014:18
*** chlong has joined #openstack-manila14:21
*** rraja has quit IRC14:24
*** porrua_ has joined #openstack-manila14:24
-openstackstatus- NOTICE: latest base images have mistakenly put python3 in some places expecting python2 causing widespread failure of docs patches - fixes are underway14:26
*** porrua has quit IRC14:27
gouthamrvponomaryov: ^ :P14:33
vponomaryovgouthamr: i knew it is Ci problem, because it works locally14:34
gouthamrvponomaryov: i was going to give you credit for breaking it14:34
vponomaryovgouthamr: just it is unclear when starts workign again14:34
gouthamrvponomaryov: :) yes... 'spose we'll get another status update.14:35
vponomaryovgouthamr: sure, in any unknown situation you can blame me ^_^14:35
gouthamr:D #thanksvponomaryov14:35
*** eharney has quit IRC14:36
*** eharney has joined #openstack-manila14:36
vponomaryovgouthamr: is Easter day is working day for you?14:36
gouthamrvponomaryov: it's sunday, isn't it..14:37
vponomaryovgouthamr: monday, after it14:37
vponomaryovgouthamr: you know, each holiday that appears to happen on Sunday or Sat should be compensated with some working day14:38
tommylikehuping zhongjun_ , gouthamr14:38
gouthamrvponomaryov: ah... no monday is a working day at NetApp.14:38
vponomaryovgouthamr: ok, ty for info14:38
tommylikehugouthamr: thanks you are here14:39
gouthamrtommylikehu: hey!14:39
Zhongjun__goutharmr: https://review.openstack.org/#/c/266688/ hi14:39
tommylikehuI think we are proposing the same thing in manila and cinder14:40
Zhongjun__gouthamr: Did you want to add regex filter in this spec?14:40
tommylikehugouthamr:  so you misunderstand the propose in manila that we are going to support regex?14:40
gouthamrtommylikehu: currently we support 'exact' filters... imo we shouldn't overload them14:41
gouthamrtommylikehu: i think we can accept regex as part of the same filters if wildcards are present14:42
gouthamralso, the spec doesn't propose limiting these filters to certain fields...14:43
gouthamrs/filters/'LIKE' searches14:43
tommylikehuwhy we should also introduce regex as well as like operator14:43
tommylikehuthe last point make sense14:43
tommylikehuZhongjun__:14:44
tommylikehugouthamr:  I am a little afraid whether I get your point on overload the exact filters14:46
-openstackstatus- NOTICE: we have rolled back centos-7, fedora-25 and ubuntu-xenial images to the previous days release. Feel free to recheck your jobs now.14:47
*** markstur has joined #openstack-manila14:48
*** ianychoi has quit IRC14:48
Zhongjun__gouthamr:  As the spec said: "Although we can provide filtering resource based on regex filter, but there is a possibility that we could have ReDos `[3]`_ attack."  We don't support regex in this filter14:48
*** mtanino has joined #openstack-manila14:49
tommylikehuZhongjun__: based on my understanding of gouthamr 's comments14:50
tommylikehuI think his concerns are about these two:14:51
tommylikehu1. we should announce which column we would apply 'like' filter14:51
*** ianychoi has joined #openstack-manila14:52
*** markstur has quit IRC14:52
tommylikehu2. this change should not affect the original usages...14:53
gouthamrtommylikehu: +114:53
gouthamrtommylikehu: +114:53
gouthamrfor both points...14:53
gouthamrZhongjun__: sorry, was in a short meeting and will be leaving soon for another one...14:53
gouthamrZhongjun__: i may not be understanding this entirely... correct me if i'm wrong.. we now have name=xyz returning all shares that are named xyz14:55
gouthamrZhongjun__: after your change, we'll have name=xyz returning shares that begin with xyz14:55
tommylikehuinclude xyz14:55
tommylikehuor contains14:55
Zhongjun__gouthamr:  yes include xyz( 1_xyz_1  4_xyz)14:56
*** kaisers has joined #openstack-manila14:56
Zhongjun__gouthamr:.op('LIKE')(u'%' + name + u'%'))14:56
gouthamrtommylikehu: okay.. i get your point about regex filtering, thanks. However, i still think we shouldn't overwrite the semantics of name=xyz for that..14:57
gouthamrtommylikehu, Zhongjun__: sorry got to run to this meeting. will comment on the patch if you respond to those comments14:57
tommylikehugouthamr: what's your main concerns on that ?14:58
*** hongbin has joined #openstack-manila14:58
tommylikehuoverwrite the semantics of name=xyz..14:58
*** belmoreira has quit IRC14:59
tommylikehugouthamr:  you should be in meeting 2, so maybe please leave your comments there, thanks14:59
tbarrongouthamr: tommylikehu Zhongjun__ I think gouthamr should make his comment about overloading on the cinder spec too.  We should drive that conversation to a consistent conclusion across the two projects since15:02
tbarronthere is really nothing manila-specific or cinder-specific about what the api should look like for this15:03
tommylikehutbarron: in cinder this change will only affect the filter and list APIs15:03
tbarrontommylikehu: I thought in manila it was only going to affect list apis?15:04
* tbarron may not be up-to-date15:04
tommylikehuhaha, my point is based on the implementation15:04
tbarronI know the sqlalchemy details are different in the two projects, but why wouldn't we have a consistent REST api?15:05
tommylikehutbarron: if we don't affect the orignal list APIs, the new one would be a little strange for end user15:06
tommylikehusomething like manila list name like 123?15:06
tommylikehumaybe I am a little misunderstanding your point, my point is we should change the list API behaviours15:07
tbarronmy only point is that we should do it the same way in both projects at the REST api level; I'd hate to see manila insist on no overload and cinder go on and overload, or whatever.15:08
tbarronit's an inconsistent experience for actual cloud users for no good reason.15:09
Zhongjun__tbarron: Do you want to add a flag in list API, like this "manila list --name=text --exactly=False/True"15:09
tommylikehutbarron: ok, that point you always have at the begining15:09
tbarronZhongjun__: I just want gouthamr to make exactly the same argument he's making in manila on the cinder spec15:10
tbarronZhongjun__: and then for you and tommylikehu to work through that issue in both projects to a consistent resolution.15:10
tbarronZhongjun__: I'll watch both specs and offer my opinions there, but I think those opinions aren't as important as the one I'm expressing right now: let's do it in a consistent way across the two projects unless there's a really good reason not to do so.15:11
* tbarron can't imagine what that would be at the moment15:11
tommylikehuZhongjun__:  +115:12
Zhongjun__tbarron: We always have a consistent way in the end.15:12
tbarronZhongjun__: well, maybe when you and tommylikehu work on both projects, we do. But historically15:13
tbarronsometimes cinder and manila have done APIs differently for no good reason.15:13
tommylikehuping jprovazn15:17
Zhongjun__tommylikehu, tbarron: We will do it ASAP, but two projects has many different things, such as: spec deadline15:17
tommylikehuone question about your user messages spec, why are you going to clear out the expired messages by administrator's command instead of clean them automatically15:19
tommylikehujprovazn: :15:19
*** gcb has quit IRC15:24
Zhongjun__vponomaryov: If we don't care about another one(enhance update shares), is it ok for this spec(ensure share) https://review.openstack.org/#/c/446494/.15:27
*** chlong_ has joined #openstack-manila15:29
*** chlong has quit IRC15:32
*** eharney_ has joined #openstack-manila15:35
*** eharney has quit IRC15:39
*** eharney_ is now known as eharney15:39
*** chlong has joined #openstack-manila15:44
jprovazntommylikehu, either would work, I consider manage command as a more flexible way where admin can better manage this15:46
*** cknight has joined #openstack-manila15:46
*** Administrator_ has quit IRC15:51
*** Administrator_ has joined #openstack-manila15:52
*** kaisers has quit IRC15:55
*** xyang_ has quit IRC16:05
*** ianychoi has quit IRC16:07
Zhongjun__gouthamr: hi, I respond to your inline comments, please take a look when you out of meeting. thanks in advance   https://review.openstack.org/#/c/447775/9/specs/pike/add-like-filter.rst16:09
*** ianychoi has joined #openstack-manila16:12
*** xyang_ has joined #openstack-manila16:12
openstackgerritzhongjun proposed openstack/manila-specs master: Add spec for infinite share  https://review.openstack.org/45209716:27
*** pcaruana has quit IRC16:29
*** a-pugachev has quit IRC16:35
*** a-pugachev_ has joined #openstack-manila16:35
*** a-pugachev_ has quit IRC16:36
openstackgerritzhongjun proposed openstack/manila-specs master: Add spec for infinite share  https://review.openstack.org/45209716:36
openstackgerritValeriy Ponomaryov proposed openstack/manila master: Add quotas per share type  https://review.openstack.org/45215816:38
*** lpetrut has quit IRC16:39
*** gouthamr has quit IRC16:47
*** kaisers has joined #openstack-manila16:52
*** gouthamr has joined #openstack-manila17:10
*** mtanino has quit IRC17:21
*** xyang_ has quit IRC17:24
*** xyang_ has joined #openstack-manila17:25
*** xyang_ has quit IRC17:29
*** kaisers has quit IRC17:32
*** kaisers has joined #openstack-manila17:32
gouthamrzhongjun_ tommylikehu tbarron: commented on both patches...17:36
*** ociuhandu has quit IRC17:45
*** xyang_ has joined #openstack-manila17:53
*** cknight has quit IRC17:54
*** mtanino has joined #openstack-manila18:06
*** jprovazn has quit IRC18:08
*** ociuhandu has joined #openstack-manila18:25
*** mtanino has left #openstack-manila18:35
*** ociuhandu has quit IRC18:40
*** ociuhandu has joined #openstack-manila18:41
*** kaisers has quit IRC18:47
*** Zhongjun__ has quit IRC18:51
*** kaisers has joined #openstack-manila18:53
*** kaisers has quit IRC18:57
*** xyang_ has quit IRC18:58
*** kaisers has joined #openstack-manila18:59
*** xyang_ has joined #openstack-manila19:02
*** ociuhandu has quit IRC19:03
*** ociuhandu has joined #openstack-manila19:03
*** xyang_ has quit IRC19:04
*** ianychoi has quit IRC19:34
*** ianychoi has joined #openstack-manila19:35
*** xyang_ has joined #openstack-manila19:55
*** xyang_ has quit IRC19:56
*** kaisers has quit IRC19:57
*** porrua_ has quit IRC20:00
*** eharney has quit IRC20:04
*** xyang_ has joined #openstack-manila20:13
*** xyang1 has joined #openstack-manila20:13
*** eharney has joined #openstack-manila20:18
*** xyang_ has quit IRC20:19
*** kaisers has joined #openstack-manila20:22
*** kaisers has quit IRC20:24
*** kaisers has joined #openstack-manila20:24
openstackgerritMerged openstack/manila-specs master: Add spec for user messages  https://review.openstack.org/43427720:27
*** gouthamr has quit IRC20:28
*** xyang_ has joined #openstack-manila20:28
*** xyang_ has quit IRC20:38
*** dmellado has quit IRC20:48
*** xyang_ has joined #openstack-manila20:51
*** xyang1 has quit IRC21:00
*** xyang1 has joined #openstack-manila21:00
*** chlong has quit IRC21:09
*** chlong_ has quit IRC21:09
*** gouthamr has joined #openstack-manila21:13
*** eharney has quit IRC21:19
*** catintheroof has joined #openstack-manila21:19
*** catintheroof has quit IRC21:21
*** catintheroof has joined #openstack-manila21:21
*** kaisers has quit IRC21:37
*** ociuhandu has quit IRC21:43
*** ociuhandu has joined #openstack-manila21:53
*** xyang_ has quit IRC22:23
*** dgonzalez has quit IRC22:24
*** ganso has quit IRC22:26
*** xyang1 has quit IRC22:26
*** dgonzalez has joined #openstack-manila22:28
*** ociuhandu has quit IRC22:29
*** ociuhandu has joined #openstack-manila22:31
*** kaisers has joined #openstack-manila22:37
*** cknight has joined #openstack-manila23:14
*** erlon has quit IRC23:15
*** cknight has quit IRC23:21
*** catintheroof has quit IRC23:29
*** catintheroof has joined #openstack-manila23:30
*** catintheroof has quit IRC23:30
tommylikehuPing gouthamr from my cellphone23:44
gouthamrtommylikehu: pong23:44
tommylikehuI think there is a little difference between cinder and manila on like filter23:45
tommylikehuIf cinder is finally combined the like filter spec and generilaze filter spec23:46
gouthamrtommylikehu: that would be good, however, things that can be searched with field like will still be limited.23:46
gouthamr"field like xyz"23:47
tommylikehuThe possible result of that is the administrator will have the ability to decide which one is a filter and which one is a like filter23:47
gouthamrwhy not allow both variants to be present?23:47
gouthamrname=xyz, name_like=xyz23:48
tommylikehuI reckon it's a little strange23:48
tommylikehuDo you know any other projects have this behaviour as yoi said23:48
*** cknight has joined #openstack-manila23:48
gouthamrtommylikehu: related q: have you seen searchlight's API (in turn, elastic search API)?23:49
tommylikehuI mean the peoject which using this searchlight23:49
tommylikehuI havn't used searchlight23:50
gouthamrtommylikehu: name_regex was proposed to ironic as zhongjun noted, i tried looking for it in implementation: https://review.openstack.org/#/c/266688/23:50
tommylikehuyes?23:51
tommylikehuOk23:51
gouthamrtommylikehu: me neither, but i've used elastic search REST API, and this here: https://developer.openstack.org/api-ref/search/ it seems to be built on top of it23:52
tommylikehuAnd we have nova and karbor which doesn't act in this way23:52
tommylikehuSo which one should we choose23:52
*** hongbin has quit IRC23:54
gouthamrtommylikehu: i'm for clear separation between exact and non-exact (contains, does not contain etc.) querying.... it seems semantically easier to consume without confusion23:58
tommylikehuquerying from inside the projects or outside?23:59

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