Thursday, 2018-02-22

*** ywat has joined #openstack-meeting-300:04
*** yamamoto has joined #openstack-meeting-300:08
*** yamamoto has quit IRC00:14
*** chyka has quit IRC00:37
*** d0ugal has quit IRC00:47
*** wanghao has joined #openstack-meeting-300:52
*** wanghao has quit IRC00:53
*** wanghao has joined #openstack-meeting-300:53
*** d0ugal has joined #openstack-meeting-300:57
*** mjturek has quit IRC00:58
*** yamamoto has joined #openstack-meeting-301:10
*** yamamoto has quit IRC01:16
*** slaweq has quit IRC01:34
*** slaweq has joined #openstack-meeting-301:35
*** wanghao_ has joined #openstack-meeting-301:39
*** wanghao has quit IRC01:39
*** slaweq has quit IRC01:39
*** salv-orlando has joined #openstack-meeting-301:41
*** salv-orlando has quit IRC01:46
*** wanghao has joined #openstack-meeting-301:53
*** wanghao_ has quit IRC01:56
*** chyka has joined #openstack-meeting-302:10
*** yamamoto has joined #openstack-meeting-302:12
*** hongbin has joined #openstack-meeting-302:13
*** chyka has quit IRC02:14
*** dklyle has joined #openstack-meeting-302:17
*** yamamoto has quit IRC02:18
*** david-lyle has quit IRC02:21
*** devananda has quit IRC02:28
*** zhenguo has joined #openstack-meeting-302:32
*** yamahata has quit IRC02:34
*** yamamoto has joined #openstack-meeting-303:14
*** yamamoto has quit IRC03:20
*** yamamoto has joined #openstack-meeting-303:37
*** salv-orlando has joined #openstack-meeting-303:43
*** salv-orlando has quit IRC03:49
*** slaweq has joined #openstack-meeting-303:59
*** d0ugal has quit IRC04:00
*** slaweq has quit IRC04:04
*** anilvenkata has quit IRC04:14
*** VW has joined #openstack-meeting-304:26
*** d0ugal has joined #openstack-meeting-304:27
*** psachin has joined #openstack-meeting-304:34
*** zhenguo has quit IRC04:42
*** salv-orlando has joined #openstack-meeting-304:45
*** salv-orlando has quit IRC04:49
*** pgadiya has joined #openstack-meeting-304:53
*** VW has quit IRC04:55
*** VW has joined #openstack-meeting-304:55
*** VW has quit IRC05:00
*** hongbin has quit IRC05:02
*** slaweq has joined #openstack-meeting-305:25
*** slaweq has quit IRC05:30
*** salv-orlando has joined #openstack-meeting-305:45
*** salv-orlando has quit IRC05:50
*** tdasilva has quit IRC06:10
*** lpetrut has joined #openstack-meeting-306:18
*** salv-orlando has joined #openstack-meeting-306:19
*** tdasilva has joined #openstack-meeting-306:25
*** slaweq has joined #openstack-meeting-306:27
*** slaweq has quit IRC06:31
*** alexchadin has joined #openstack-meeting-306:41
*** marios has joined #openstack-meeting-306:48
*** salv-orlando has quit IRC06:51
*** salv-orlando has joined #openstack-meeting-306:51
*** salv-orlando has quit IRC06:56
*** ianychoi has quit IRC07:01
*** ianychoi has joined #openstack-meeting-307:01
*** lpetrut has quit IRC07:11
*** Swami has quit IRC07:13
*** pcaruana has joined #openstack-meeting-307:37
*** lpetrut has joined #openstack-meeting-307:37
*** slaweq_ has joined #openstack-meeting-307:47
*** salv-orlando has joined #openstack-meeting-307:52
*** slaweq_ has quit IRC07:52
*** salv-orlando has quit IRC07:56
*** chyka has joined #openstack-meeting-307:56
*** slaweq_ has joined #openstack-meeting-308:00
*** chyka has quit IRC08:01
*** slaweq_ has quit IRC08:04
*** lpetrut has quit IRC08:09
*** MarkBaker_ has quit IRC08:11
*** shu-mutou-AWAY is now known as shu-mutou08:12
*** slaweq_ has joined #openstack-meeting-308:28
*** rossella_s has joined #openstack-meeting-308:31
*** slaweq_ has quit IRC08:45
*** salv-orlando has joined #openstack-meeting-308:52
*** tssurya has joined #openstack-meeting-308:54
*** salv-orlando has quit IRC08:57
*** salv-orlando has joined #openstack-meeting-308:57
*** alexchadin has quit IRC09:03
*** dougsz has joined #openstack-meeting-309:13
*** dougsz has quit IRC09:16
*** MarkBaker_ has joined #openstack-meeting-309:31
*** wanghao has quit IRC09:32
*** mab__ has joined #openstack-meeting-309:36
*** MarkBaker_ has quit IRC09:38
*** e0ne has joined #openstack-meeting-309:38
*** stendulker has joined #openstack-meeting-309:42
*** stendulker has quit IRC09:46
*** stendulker has joined #openstack-meeting-309:46
*** egallen has joined #openstack-meeting-309:50
*** anilvenkata has joined #openstack-meeting-309:53
*** shu-mutou is now known as shu-mutou-AWAY09:54
*** yamamoto has quit IRC10:06
*** ywat has quit IRC10:09
*** mab__ has quit IRC10:12
*** lpetrut has joined #openstack-meeting-310:17
*** rossella_s has quit IRC10:20
*** rossella_s has joined #openstack-meeting-310:23
*** egallen has quit IRC10:26
*** egallen has joined #openstack-meeting-310:28
*** yamahata has joined #openstack-meeting-310:32
*** alexchadin has joined #openstack-meeting-310:34
*** kei-ichi_ has joined #openstack-meeting-310:57
*** purp_too has joined #openstack-meeting-310:57
*** kei-ichi has quit IRC10:59
*** witek has quit IRC10:59
*** purp has quit IRC10:59
*** witek has joined #openstack-meeting-310:59
*** salv-orl_ has joined #openstack-meeting-311:00
*** yamahata has quit IRC11:00
*** kei-ichi_ has quit IRC11:01
*** purp_too has quit IRC11:01
*** PagliaccisCloud has quit IRC11:02
*** salv-orlando has quit IRC11:03
*** kei-ichi has joined #openstack-meeting-311:04
*** PagliaccisCloud has joined #openstack-meeting-311:04
*** purp has joined #openstack-meeting-311:04
*** yamamoto has joined #openstack-meeting-311:06
*** pcaruana has quit IRC11:07
*** egallen_ has joined #openstack-meeting-311:09
*** egallen has quit IRC11:10
*** egallen_ is now known as egallen11:10
*** yamamoto has quit IRC11:13
*** egallen has quit IRC11:16
*** stendulker has quit IRC11:17
*** mab__ has joined #openstack-meeting-311:19
*** Faster-Fanboi_ has joined #openstack-meeting-311:32
*** Faster-Fanboi has quit IRC11:34
*** luzC has quit IRC11:36
*** luzC has joined #openstack-meeting-311:39
*** pbourke has quit IRC11:42
*** pbourke has joined #openstack-meeting-311:47
*** pbourke_ has joined #openstack-meeting-311:52
*** anilvenkata has quit IRC11:57
*** pbourke has quit IRC11:58
*** diman has joined #openstack-meeting-312:06
*** yamamoto has joined #openstack-meeting-312:09
*** pcaruana has joined #openstack-meeting-312:09
*** raildo has joined #openstack-meeting-312:10
*** yamamoto has quit IRC12:14
*** yamamoto has joined #openstack-meeting-312:27
*** chyka has joined #openstack-meeting-312:39
*** chyka has quit IRC12:43
*** mattw4 has quit IRC12:51
*** mattw4 has joined #openstack-meeting-312:51
*** wxy has quit IRC12:52
*** wxy has joined #openstack-meeting-312:52
*** salv-orl_ has quit IRC13:10
*** salv-orlando has joined #openstack-meeting-313:11
*** betherly has quit IRC13:15
*** salv-orlando has quit IRC13:15
*** salv-orlando has joined #openstack-meeting-313:15
*** betherly has joined #openstack-meeting-313:15
*** ying_zuo has quit IRC13:16
*** zhuli has quit IRC13:16
*** zhuli has joined #openstack-meeting-313:16
*** ying_zuo has joined #openstack-meeting-313:17
*** zhipeng has joined #openstack-meeting-313:28
*** yamamoto has quit IRC13:34
*** yamamoto has joined #openstack-meeting-313:44
*** yamamoto has quit IRC13:44
*** lyan has joined #openstack-meeting-313:46
*** pgadiya has quit IRC13:49
*** VW has joined #openstack-meeting-313:51
*** jeremyfreudberg has joined #openstack-meeting-313:53
tellesnobrega#startmeeting sahara14:00
openstackMeeting started Thu Feb 22 14:00:11 2018 UTC and is due to finish in 60 minutes.  The chair is tellesnobrega. Information about MeetBot at http://wiki.debian.org/MeetBot.14:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.14:00
*** openstack changes topic to " (Meeting topic: sahara)"14:00
openstackThe meeting name has been set to 'sahara'14:00
jeremyfreudbergo/14:00
*** tosky has joined #openstack-meeting-314:01
toskyo/14:02
tellesnobregaseems like it is only us 3 today14:02
tellesnobrega#topic News/Updates14:02
*** openstack changes topic to "News/Updates (Meeting topic: sahara)"14:02
tellesnobregaI finally worked the schedule for PTG, please take a look14:02
tellesnobrega#link https://etherpad.openstack.org/p/sahara-rocky-ptg14:03
tellesnobregaalso I started working on plugins outside sahara core14:03
jeremyfreudbergi'm working on things related to vanilla 2.8.2, as you saw14:04
* jeremyfreudberg looks at etherpad14:04
* tosky looking too14:05
toskysome space for the documentation discussion, reviewing the notes collected two weeks ago?14:06
tellesnobregaI thought about it, don't know where we can put it, maybe we can get together wednesday to do it14:06
toskymaybe also a "go through the list of bugs" and see if there is something relevant14:07
tellesnobregatosky, makes sense, I will put it together with something that appears not to take the whole slot14:07
toskyalso, the storyboard team is meeting on Wednesday too, we could crash there and ask some questions: http://lists.openstack.org/pipermail/openstack-dev/2018-February/127620.html14:08
tellesnobregayes, that sounds great14:08
*** alexchadin has quit IRC14:08
jeremyfreudbergon that thursday schedule (doesn't affect me but thought i would say), i think they said lunch is actually 12:30-13:30, so you can find a few minutes there to inject the missing topics14:09
tellesnobregajeremyfreudberg, thanks, I will update it14:09
jeremyfreudbergschedule looks fine otherwise,  looks like enough padding on friday that you guys will have time to give me thursday's summary14:10
tellesnobregaI didn't put anything after lunch, but you and tosky are free to get together14:11
toskyI would say: let's set up Friday afternoon program based on the previous discussions14:12
toskydirectly on the place14:12
tellesnobregayes14:12
jeremyfreudbergagreed14:12
*** zhipeng has quit IRC14:12
tellesnobregathat is what I meant by you are free to do whatever you want with that time, sadly I won't be able to be there14:12
toskyso leaving immediately after lunch, right?14:13
tellesnobregaI think it is the safest option, since my flight is at 17:40 and I should be there at least 2h before14:13
*** zhipeng has joined #openstack-meeting-314:14
* jeremyfreudberg does mental math14:15
toskyleaving at most at 14.40, yes - I think that the hotel is on the way of the airport (for some bus)14:16
tellesnobregacool14:16
tellesnobregathat is the plan, I could participate a little after lunch14:16
tellesnobregabut not too much14:16
jeremyfreudbergyes, i was going to say you could probably make it till 2pm and still be comfortable14:17
jeremyfreudbergdepends what we have left for discussion after lunch14:17
tellesnobregaI will look up time and how to get from airport to hotel and figure out timing14:18
tellesnobregahopefully I'm able to be stay a little longer14:18
tellesnobregaare we ok to move on?14:19
jeremyfreudbergyep, next topic14:20
tellesnobrega#topic Open Discussion14:20
*** openstack changes topic to "Open Discussion (Meeting topic: sahara)"14:20
tellesnobregalets go ahead to open discussion, there is not too much that we need to talk today that can't wait until next week14:21
tellesnobregawe didn't cut a new RC, we are waiting to cut a final release stable/queens14:21
jeremyfreudbergthoughts on the 2.8.2 image stuff?14:21
tellesnobregayou said the patch isn't perfect, what is missing?14:22
jeremyfreudbergthat was just in reference to the tarballs stuff14:23
toskythe location14:23
tellesnobregahmm, that isn't really fixable right now14:24
toskytechnically we could release as it is, and use the PTG to understand if the patch which adds the variable for the path inside tarballs.o.o is needed or not14:24
tosky(namely https://review.openstack.org/#/c/546853/ )14:25
toskyfor sure we can use Monday (for Telles alone) and Tuesday (both) to crash into the infra/QA/etc help room and not go out until this is solved :)14:25
tellesnobregathat sounds good14:26
tellesnobregaare we ok with merging jeremy's patch as is and fix the location later?14:26
jeremyfreudbergmostly i am a fan of 546853 because if we do change the artifacts stuff again then the adjustments are not dependent on a new release (users can switch immediately)14:26
jeremyfreudbergon the other hand,14:27
jeremyfreudbergthe artifacts are fairly stable at this point14:27
*** baoli has joined #openstack-meeting-314:27
jeremyfreudbergso even if the url changes back to sahara the sahara-extra url will still be okay14:27
*** mjturek has joined #openstack-meeting-314:27
jeremyfreudbergbut i can't predict the future14:27
*** baoli has quit IRC14:28
*** baoli has joined #openstack-meeting-314:29
tellesnobregaSotK, what we have now, a patch that allows creation of vanilla 2.8.2 but publishes to sahara-extra14:29
toskyyep, and it works as it is14:29
tellesnobregaand a patch that fixes it but we don't know if it is necessary due to need to talk with QA/infra team and figure it out14:29
jeremyfreudberg"fixes it" meaning makes it future proof in case of emergency, without the need for followup-release14:30
tellesnobregasounds good14:31
tellesnobregaI'm ok with merging the vanilla patch14:31
toskyand backporting too14:31
jeremyfreudbergyes14:31
toskysorry, I need to bbl - just a quick note about final note:14:31
tellesnobregatrue14:31
toskyhttps://review.openstack.org/#/c/546691/14:31
toskywith that, I can tag a release for sahara-tests (so RDO people don't kill me :)14:32
jeremyfreudbergi think we can skip the follow-on patch, the use case in which it is absolutely necessary is narrow/unlikely14:32
tellesnobregatosky, reviewing :)14:32
tellesnobregaI'm approving the vanilla patch14:33
tellesnobregajeremyfreudberg, follow up with backport14:33
toskyso, if you are not on IRC later and/or tomorrow, see you at the PTG!14:34
tellesnobregasee you tosky14:34
* tosky disappears earlier14:34
tellesnobregajeremyfreudberg, anything else to talk about?14:34
*** alexchadin has joined #openstack-meeting-314:35
jeremyfreudbergnothing urgent, it's all better said at PTG14:35
tellesnobregasounds good, we can close it earlier14:36
jeremyfreudbergyep14:37
tellesnobregathanks jeremyfreudberg, see you next week14:37
tellesnobrega#endmeeting14:38
*** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/"14:38
openstackMeeting ended Thu Feb 22 14:38:11 2018 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)14:38
openstackMinutes:        http://eavesdrop.openstack.org/meetings/sahara/2018/sahara.2018-02-22-14.00.html14:38
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/sahara/2018/sahara.2018-02-22-14.00.txt14:38
openstackLog:            http://eavesdrop.openstack.org/meetings/sahara/2018/sahara.2018-02-22-14.00.log.html14:38
*** dklyle has quit IRC14:39
*** jeremyfreudberg has left #openstack-meeting-314:39
*** yamamoto has joined #openstack-meeting-314:45
*** tosky has left #openstack-meeting-314:47
*** psachin has quit IRC14:49
*** sgrasley1 has joined #openstack-meeting-314:49
*** yamamoto has quit IRC14:51
*** anilvenkata has joined #openstack-meeting-314:56
*** e0ne has quit IRC14:57
*** namnh has joined #openstack-meeting-314:57
*** cmurphy has quit IRC14:57
*** e0ne has joined #openstack-meeting-314:59
*** baoli has quit IRC15:05
*** baoli has joined #openstack-meeting-315:06
*** devananda has joined #openstack-meeting-315:23
*** alexchadin has quit IRC15:26
*** alexchadin has joined #openstack-meeting-315:27
*** alexchadin has quit IRC15:31
*** e0ne has quit IRC15:34
*** e0ne has joined #openstack-meeting-315:35
*** alexchadin has joined #openstack-meeting-315:37
*** krtaylor has quit IRC15:38
*** yamahata has joined #openstack-meeting-315:40
*** alexchadin has quit IRC15:42
*** zhipeng has quit IRC15:42
*** hongbin has joined #openstack-meeting-315:46
*** yamamoto has joined #openstack-meeting-315:47
*** david-lyle has joined #openstack-meeting-315:50
*** egallen has joined #openstack-meeting-315:51
*** rossella_s has quit IRC15:51
*** yamamoto has quit IRC15:51
*** VW has quit IRC15:56
*** VW has joined #openstack-meeting-315:57
*** anilvenkata is now known as anilvenkata|AFK15:57
edleafe#startmeeting api sig16:00
openstackMeeting started Thu Feb 22 16:00:02 2018 UTC and is due to finish in 60 minutes.  The chair is edleafe. Information about MeetBot at http://wiki.debian.org/MeetBot.16:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.16:00
*** openstack changes topic to " (Meeting topic: api sig)"16:00
openstackThe meeting name has been set to 'api_sig'16:00
edleafe#chair cdent elmiko edleafe dtantsur16:00
edleafe#link https://wiki.openstack.org/wiki/Meetings/API-SIG#Agenda16:00
openstackWarning: Nick not in channel: cdent16:00
openstackWarning: Nick not in channel: elmiko16:00
openstackWarning: Nick not in channel: dtantsur16:00
openstackCurrent chairs: cdent dtantsur edleafe elmiko16:00
edleafeAm I all alone? Anyone else around?16:00
edleafeelmiko said he would be late, so I'll hang around for a bit16:01
*** dtantsur has joined #openstack-meeting-316:02
dtantsuro/16:02
dtantsursorry, missed the start16:02
* edleafe doesn't feel so lonely16:03
dtantsurlol16:03
dtantsurI nearly feel like we should talk about the PTG schedule and wrap it up16:03
dtantsursince we hopefully meet in 4 days16:03
edleafeYeah16:04
edleafelet's skip over the action items16:04
edleafe#topic open mic and ongoing or new biz16:04
edleafePTG planning16:04
edleafe#link https://etherpad.openstack.org/p/api-sig-ptg-rocky16:04
edleafe#link Results of priority voting: https://ethercalc.openstack.org/xja22ghws13i16:04
*** openstack changes topic to "open mic and ongoing or new biz (Meeting topic: api sig)"16:04
*** elmiko has joined #openstack-meeting-316:04
elmikoo/16:04
edleafeI made an ethercalc of the results16:04
edleafeSince we have a 4-vote topic with "microversions" in the title, we should be in that room all day :)16:05
*** rossella_s has joined #openstack-meeting-316:05
elmikolol16:05
dtantsurhaha16:06
dtantsurmy bet is that we'll manage to cover the first 4 maximum16:06
edleafeprobably16:07
edleafeWe always add the "review X" topics in case we run out of things16:07
edleafeBut I'd be more inclined to wrap up early and get a pint together16:07
dtantsur++16:08
dtantsurI seem to clean up Monday evening a bit16:08
edleafeSince we're meeting in 4 days, as dtantsur pointed out, I propose skipping the rest of this meeting, unless there is something urgent on your minds16:10
elmikoedleafe: yeah, i just added the "review x" as a stop-gap16:11
elmikobut i agree about pints ;)16:11
* dtantsur has nothing16:11
elmikoi'm good with that16:11
edleafeOK, then, here's to safe travels for all, and we'll see each other in Dublin!16:11
elmikolikewise, safe travels!16:12
edleafe#endmeeting16:12
*** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/"16:12
openstackMeeting ended Thu Feb 22 16:12:17 2018 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:12
openstackMinutes:        http://eavesdrop.openstack.org/meetings/api_sig/2018/api_sig.2018-02-22-16.00.html16:12
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/api_sig/2018/api_sig.2018-02-22-16.00.txt16:12
openstackLog:            http://eavesdrop.openstack.org/meetings/api_sig/2018/api_sig.2018-02-22-16.00.log.html16:12
*** dtantsur has left #openstack-meeting-316:12
edleafeOh, I'll do the newsletter!16:12
*** namnh has quit IRC16:17
*** rossella_s has quit IRC16:20
*** rossella_s has joined #openstack-meeting-316:22
*** d0ugal has quit IRC16:35
*** d0ugal has joined #openstack-meeting-316:40
*** david-lyle is now known as dklyle16:42
*** e0ne has quit IRC16:46
*** yamamoto has joined #openstack-meeting-316:48
*** chyka has joined #openstack-meeting-316:48
*** cmurphy has joined #openstack-meeting-316:48
*** mab__ has quit IRC16:50
*** yamamoto has quit IRC16:52
*** dklyle has quit IRC16:56
*** rossella_s has quit IRC16:57
*** rossella_s has joined #openstack-meeting-316:59
*** VW_ has joined #openstack-meeting-317:00
*** VW has quit IRC17:00
*** david-lyle has joined #openstack-meeting-317:01
*** baoli has quit IRC17:24
*** baoli has joined #openstack-meeting-317:24
*** slaweq has joined #openstack-meeting-317:29
*** diman has quit IRC17:30
*** diman has joined #openstack-meeting-317:31
*** marios has quit IRC17:34
*** diman has quit IRC17:35
*** tssurya has quit IRC17:43
*** lpetrut has quit IRC17:46
*** slaweq has quit IRC17:48
*** yamamoto has joined #openstack-meeting-317:48
*** tbachman has joined #openstack-meeting-317:53
*** yamamoto has quit IRC17:54
*** rkukura has joined #openstack-meeting-318:01
rkukurahi18:02
tbachmanrkukura: hi!18:02
tbachmanrkukura: I guess if we don’t see SumitNaiksatam and/or annakk in the next 10 minutes, we should call it?18:06
rkukuratbachman: agreed18:07
*** SumitNaiksatam has joined #openstack-meeting-318:07
SumitNaiksatamrkukura: tbachman: hi18:08
tbachmanSumitNaiksatam: hi!18:08
rkukurahi SumitNaiksatam18:08
*** anilvenkata|AFK has quit IRC18:08
tbachmanSumitNaiksatam: I haven’t seen annakk yet18:08
SumitNaiksatamtbachman: thanks, yeah i was checking as well18:09
SumitNaiksatamsorry i am late again today18:09
tbachmanSumitNaiksatam: np!18:09
SumitNaiksatamunlike last time, i didnt forget today, just got late :-)18:09
SumitNaiksatam#startmeeting networking_policy18:09
openstackMeeting started Thu Feb 22 18:09:36 2018 UTC and is due to finish in 60 minutes.  The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot.18:09
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.18:09
*** openstack changes topic to " (Meeting topic: networking_policy)"18:09
openstackThe meeting name has been set to 'networking_policy'18:09
SumitNaiksatam#topic Pending patches18:09
*** openstack changes topic to "Pending patches (Meeting topic: networking_policy)"18:09
tbachmanSumitNaiksatam: we may also want to add “Reverting patches”18:10
tbachmantho we may just carry that discussion over gerrit18:10
SumitNaiksatamtbachman: ah18:10
SumitNaiksatamwhich ones?18:10
tbachmanthe one I merged :P18:10
SumitNaiksatamlets discuss here18:10
* tbachman goes to gerrit18:10
tbachman#link https://review.openstack.org/#/c/546269/18:11
SumitNaiksatamokay i see bob’s comments now18:11
SumitNaiksatamincidentally, amit had discussed this with me yesterday18:12
rkukuraI’m not arguing that we should revert the patch18:12
SumitNaiksatamand i had pointed out that we prefer to not use lockmode update18:12
rkukuraBut I do thing the locking is completely usefess18:12
SumitNaiksatamrkukura: is it useless in all cases?18:13
* tbachman wonders if we have an overarching set of DB access semantics for GBP18:13
tbachmans/semantics/idioms/18:13
rkukuraif the DB server is clustered, and locks are not global across the DB servers, they don’t do any good18:13
SumitNaiksatamfyi - #link https://wiki.openstack.org/wiki/OpenStack_and_SQLAlchemy#MySQLdb_.2B_eventlet_.3D_sad18:13
tbachmanlol18:14
SumitNaiksatamsee sections “MySQLdb + eventlet = sad” and also “Pessimistic Locking - SELECT FOR UPDATE"18:14
rkukuraand in cases like this, locking the set of rows returned from the query does not help  - what would be needed would be locking the whole table to prevent adding new rows that would have been returned by the query18:14
tbachmansounds heavy18:15
SumitNaiksatamrkukura: so i agress in the case of clustered (with Galera front ending) its not helpful18:15
rkukuraand in general, we would prefer optimistic approaches to pessimistic locking in order to scale an perform better18:15
SumitNaiksatam*agree18:15
*** slaweq has joined #openstack-meeting-318:15
rkukuraand I had forgotten about the potential deadlocking with threads18:16
tbachmanDo different clustered DB servers provide this as part of their clustering behavior?18:16
SumitNaiksatamtbachman: to your point, moving to an optimistic locking strategy seemed like an overarching project concern/goal18:16
SumitNaiksatamand it seemed too big a change for this patch18:16
tbachmanthis is way beyond the topic of this specific subject, but I find it challenging to define a set of DB access semantics/idioms when we can’t limit the set of what people use18:16
tbachmanIt seems like we’d have to go with a very low bar, in order to support any DB integration18:17
* tbachman gets off soapbox18:17
tbachmans/any/every possible/18:17
* tbachman taps mic18:18
SumitNaiksatamwe tried to think through ways of doing “optimistic” locking but per Amit, in this case the logic wasnt as straightforward18:18
rkukuraI don’t claim to know what the solution is, but my point on this patch is that adding the locking doesn’t really help anything, and we’ve got bigger issues of the same type18:18
* tbachman nods18:18
*** dklyle has joined #openstack-meeting-318:18
SumitNaiksatamrkukura: i think we are somewhat covered with the thread deadlocks because we will presumably retry18:18
rkukuraSumitNaiksatam: that may be the case18:19
SumitNaiksatamrkukura: it would help in the case the DB is not clustered18:19
rkukurabut how is locking the set of rows returned by the query prevent other sessions from adding new rows that would have been returned by the query?18:19
rkukuraAs I understand it, the rows in this DB are generally not even mutated, just added or removed. So locking existing rows doesn’t do much good, if any.18:20
SumitNaiksatamrkukura: so i did not study all the cases in which reference counting was being done in this patch, but it seemed to me that at least in the delete case it would help, no?18:21
rkukuraWhat is needed is some way for the commit to validate that the same rows would be returned at that point as when the query was made18:21
*** david-lyle has quit IRC18:22
*** yamahata has quit IRC18:22
tbachmanrkukura: in the above, you’re referring to MVCC18:23
tbachman?18:23
rkukuranot specifically, but maybe18:24
tbachmank18:24
*** diman has joined #openstack-meeting-318:24
tbachmanjust wasn’t sure what a lock means in that case18:24
SumitNaiksatamrkukura: in your review comments on the patch, you meant “live” not “leave” right?18:24
rkukurabut even that seems to mainly address existing rows, not rows that might get added concurrenty during a transaction18:24
*** slaweq has quit IRC18:25
rkukuraSumitNaiksatam: No, I meant “I’d prefer that we leave the … out for now”18:26
SumitNaiksatamah sorry, i was misreading, the sentence is fine as is18:26
*** lpetrut has joined #openstack-meeting-318:27
rkukuraI don’t think the locking does a whole lot of harm in this case, so I am not arguing to revert the patch. I just don’t want the illusion that we don’t have a problem.18:27
SumitNaiksatamrkukura: totally with you on the point about this creating the illusion18:28
* tbachman nods again18:28
SumitNaiksatamhence had requested Amit to add the comment18:28
tbachmansorry again folks for the quick merge on this — should have given it my due attention :(18:28
SumitNaiksatami had pointed this link to him yesterday so he is aware of the limitations18:28
*** diman has quit IRC18:29
SumitNaiksatamtbachman: hmmm, but i think we are converging to the same result18:29
SumitNaiksatamas for other patches18:29
SumitNaiksatamthey are building up18:29
tbachmanSumitNaiksatam: I put a comment in your DB patch — more of a question, really18:30
SumitNaiksatamtbachman: thanks, just noticed and responded18:30
SumitNaiksatami had considered that before18:30
tbachmank18:31
tbachmanI just couldn’t remember if this was an issue, since I last wrote a DB migration18:31
SumitNaiksatamwe add only what is needed in the migration18:32
tbachmanso, in the case where we doing things like adding columns, etc., that makes sense18:32
tbachmanbut if we need to move things around, etc., that would be an issue, right?18:32
SumitNaiksatamtbachman: sorry, i did not understand either of those18:33
SumitNaiksatamwe are doing this only in the case of data migration18:33
tbachmanbasically, anything that would affect the integrity of the row, if the row would include a relationship to another table row18:33
SumitNaiksatamnot schema migration18:34
tbachmanah18:34
rkukuraif you all would quit changing things on my, I could complete the validation/repair tool, and we could use that to handle the data migrations in some cases going forward18:35
SumitNaiksatamrkukura: “my” ?18:36
rkukuraI cannot type on this stupid keyboard18:36
rkukurame18:36
SumitNaiksatam:-)18:36
SumitNaiksatamgot it18:36
rkukura;)18:36
SumitNaiksatamdont mean to change the topic, but quick one - #link https://review.openstack.org/#/c/54601718:36
tbachmanJust added that it should have a commit message18:37
tbachmanexplanation/context18:37
SumitNaiksatamtbachman: agree :-)18:37
rkukurawhy is this only on stable/ocata?18:38
tbachmanI think this is to address the UT issue18:39
SumitNaiksatamrkukura: its relevant only to ocata18:39
SumitNaiksatamnot just a UT issue18:39
tbachmanthere’s a UT that’s failing sporadically, but only on ocata18:39
tbachmanah18:39
rkukuraok, wouldn’t hurt to mention that in the commit msg18:39
rkukurapresumably, the code is different on other branches?18:39
SumitNaiksatamit fixes the UT, but its a much bigger issue because the  new and old transaction patterns are getting mixed up18:40
SumitNaiksatamso it breaks our transaction integrity18:40
SumitNaiksatamthis is the part where neutron introduced this change selectively18:40
SumitNaiksatamand only in Pike it was done more comprehensively18:40
SumitNaiksatamso we adapted in Pike18:41
SumitNaiksatambut in Ocata we used the older pattern (and so does Neutron in most other cases)18:41
tbachmanSumitNaiksatam: thx for the background18:41
SumitNaiksatamand neutron doesnt care as much, most likely, because this plugin is either not used as much in Ocata, or not in the way that we do18:41
* tbachman definitely thinks it needs a commit message now ;)18:41
SumitNaiksatamtbachman: lol, no18:42
SumitNaiksatami cant type either18:42
SumitNaiksatam*np18:42
SumitNaiksatami spent some time debugging the UT issue and which led to this realization18:42
rkukuraSumitNaiksatam: That explains why its only on stavle/ocata. Did “it fixes the UT” but that it “breaks our transaction integrity” above both refer to the patch?18:42
SumitNaiksatamrkukura: yes18:42
rkukuraso the patch breaks transaction integrity?18:43
SumitNaiksatamso when i refer to the “plugin” i meant the “trunk” plugin18:43
SumitNaiksatamrkukura: yes, because it mixes up the old and new transaction creation pattern18:44
rkukuraSumitNaiksatam: Are you saying the the trunk plugin broke transaction integrity, or that this patch broke transaction integrity? I’m getting confused.18:45
SumitNaiksatamrkukura: the trunk plugin broke transactional integrity for our apic_aim mech driver18:46
SumitNaiksatamit might be working fine independently18:46
rkukuraok, and this patch fixes that issue?18:47
SumitNaiksatamrkukura: yes18:47
SumitNaiksatamfor example, use of #link https://github.com/openstack/neutron/blob/stable/ocata/neutron/services/trunk/plugin.py#L22618:47
SumitNaiksatamin Ocata, causes a problem for us18:48
SumitNaiksatamthis patch aims to change the use to the older pattern prevalent in ocata18:48
rkukuraOK, I’m fine with this patch, but wonder if its worth updating its commit message18:48
SumitNaiksatamrkukura: absolutely, i didnt think of it because i had the background, but i can see why it might not make sense without knowing it18:49
SumitNaiksatamof couse, there is not justification/motivation required for putting commit messages, it should always be required! :-)18:50
SumitNaiksatam*no18:50
tbachmanSumitNaiksatam: +118:50
tbachmanthis one would be particularly helpful18:50
*** yamamoto has joined #openstack-meeting-318:50
SumitNaiksatamas reviwers we should enforce it for all patches, even for trivial patches18:51
rkukura+118:51
SumitNaiksatamjust add one line! :-)18:51
SumitNaiksatambut it matters a lot when you are reading the git log18:51
tbachmanthis one might take more than one18:51
SumitNaiksatamtbachman: lol18:52
SumitNaiksatamof course18:52
SumitNaiksatamanyway18:52
tbachmanSumitNaiksatam: totally agree about when reading the log18:52
tbachmanI’ve run into that a bit lately18:52
SumitNaiksatami used to comment on the patches earlier asking for a commit message, but patches got merged in spite, so i gave up18:53
*** pbourke_ has quit IRC18:53
tbachmanheh18:53
SumitNaiksatamas for #link https://review.openstack.org/#/c/546427/18:53
SumitNaiksatamit fixes a show stopper, basically the installation upgrade fails18:54
SumitNaiksatam(sorry i changed context and circled back to this patch which we already discussed)18:54
SumitNaiksatamrkukura: i was responding to your earlier point about things like this getting fixed by your repair tool18:55
SumitNaiksatamdepending on when the tool is run, it might fix this issue, but at least for now, since the data migration is embeded in our migration chain, it runs and breaks the upgrade18:56
rkukurasorry18:56
tbachmanSumitNaiksatam: so, we don’t need to populate this: https://github.com/openstack/group-based-policy/blob/master/gbpservice/neutron/plugins/ml2plus/drivers/apic_aim/db.py#L31-L3418:56
*** yamamoto has quit IRC18:56
tbachmanor is it this:18:56
tbachmanhttps://github.com/openstack/group-based-policy/blob/master/gbpservice/neutron/plugins/ml2plus/drivers/apic_aim/db.py#L27-L2918:56
* tbachman gets confused about directions wrt ForeignKey and relationship18:57
SumitNaiksatamtbachman: i dont think the foreign key is required18:57
tbachmanSumitNaiksatam: k. thx18:58
SumitNaiksatamwe are not redefining the table18:58
SumitNaiksatamwe are just providing enough defintion to fill the columns that are required to migrate the data18:58
tbachmangot it18:58
SumitNaiksatamthere is a stack trace i can share with you offline that will provide more context on this18:59
SumitNaiksatamnote that the patch is doing two things18:59
tbachmaneven tho there’s the ondelete=‘CASCADE’?18:59
*** yamamoto has joined #openstack-meeting-318:59
SumitNaiksatamits also getting ready of the use of the mixin methods, which also breaks things18:59
SumitNaiksatamwe are out of time19:00
SumitNaiksatamtbachman: responding offline19:00
tbachmanSumitNaiksatam: ack. Thx!19:00
SumitNaiksatamrkukura: tbachman: thanks for joining!19:00
tbachmanSumitNaiksatam: bye!19:00
SumitNaiksatam#endmeeting19:00
*** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/"19:00
openstackMeeting ended Thu Feb 22 19:00:44 2018 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)19:00
openstackMinutes:        http://eavesdrop.openstack.org/meetings/networking_policy/2018/networking_policy.2018-02-22-18.09.html19:00
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/networking_policy/2018/networking_policy.2018-02-22-18.09.txt19:00
rkukurabye19:00
openstackLog:            http://eavesdrop.openstack.org/meetings/networking_policy/2018/networking_policy.2018-02-22-18.09.log.html19:00
*** tbachman has left #openstack-meeting-319:00
*** rkukura has left #openstack-meeting-319:02
*** yamamoto has quit IRC19:04
*** yamamoto has joined #openstack-meeting-319:14
*** yamamoto has quit IRC19:19
*** rossella_s has quit IRC19:20
*** rossella_s has joined #openstack-meeting-319:23
*** edleafe- has joined #openstack-meeting-319:24
*** edleafe has quit IRC19:25
*** edleafe- is now known as edleafe19:25
*** diman has joined #openstack-meeting-319:28
*** rossella_s has quit IRC19:28
*** yamamoto has joined #openstack-meeting-319:29
*** tssurya has joined #openstack-meeting-319:30
*** rossella_s has joined #openstack-meeting-319:32
*** diman has quit IRC19:33
*** yamamoto has quit IRC19:34
*** gugl has joined #openstack-meeting-319:36
*** VW_ has quit IRC19:37
*** VW has joined #openstack-meeting-319:38
*** egallen has quit IRC19:40
*** egallen has joined #openstack-meeting-319:41
*** yamamoto has joined #openstack-meeting-319:44
*** yamamoto has quit IRC19:52
*** rossella_s has quit IRC19:52
*** rossella_s has joined #openstack-meeting-319:53
*** yamahata has joined #openstack-meeting-320:03
*** slaweq_ has joined #openstack-meeting-320:03
*** raildo has quit IRC20:04
*** egallen_ has joined #openstack-meeting-320:06
*** egallen_ has quit IRC20:06
*** egallen_ has joined #openstack-meeting-320:07
*** egallen has quit IRC20:07
*** egallen_ is now known as egallen20:07
*** slaweq_ has quit IRC20:07
*** rossella_s has quit IRC20:08
*** rossella_s has joined #openstack-meeting-320:10
*** slaweq has joined #openstack-meeting-320:22
*** egallen has quit IRC20:24
*** fyxim has quit IRC20:28
*** fyx has joined #openstack-meeting-320:28
*** diman has joined #openstack-meeting-320:30
*** diman has quit IRC20:34
*** rossella_s has quit IRC20:36
*** SumitNaiksatam has quit IRC20:38
*** rossella_s has joined #openstack-meeting-320:38
*** yamamoto has joined #openstack-meeting-320:48
*** rossella_s has quit IRC20:48
*** rossella_s has joined #openstack-meeting-320:50
*** tellesnobrega has quit IRC20:52
*** yamamoto has quit IRC20:55
*** tellesnobrega has joined #openstack-meeting-321:17
*** iyamahat has joined #openstack-meeting-321:22
*** rossella_s has quit IRC21:24
*** diman has joined #openstack-meeting-321:26
*** rossella_s has joined #openstack-meeting-321:28
*** diman has quit IRC21:30
*** pcaruana has quit IRC21:37
*** iyamahat_ has joined #openstack-meeting-321:40
*** iyamahat has quit IRC21:41
*** tssurya has quit IRC21:49
*** devananda has quit IRC21:51
*** yamamoto has joined #openstack-meeting-321:53
*** yamamoto has quit IRC21:57
*** yamamoto has joined #openstack-meeting-322:01
*** tellesnobrega has quit IRC22:18
*** slaweq has quit IRC22:22
*** slaweq has joined #openstack-meeting-322:22
*** diman has joined #openstack-meeting-322:26
*** slaweq has quit IRC22:27
*** mjturek has quit IRC22:28
*** lyan has quit IRC22:28
*** diman has quit IRC22:30
*** tonyb has quit IRC22:35
*** tonyb has joined #openstack-meeting-322:36
*** diman has joined #openstack-meeting-322:39
*** devananda has joined #openstack-meeting-322:41
*** diman has quit IRC22:44
*** lpetrut has quit IRC22:50
*** gugl has quit IRC22:51
*** yamamoto has quit IRC23:02
*** VW has quit IRC23:05
*** VW has joined #openstack-meeting-323:05
*** rossella_s has quit IRC23:07
*** rossella_s has joined #openstack-meeting-323:09
*** VW has quit IRC23:10
*** VW has joined #openstack-meeting-323:15
*** VW has quit IRC23:20
*** rossella_s has quit IRC23:21
*** rossella_s has joined #openstack-meeting-323:22
*** apetrich has quit IRC23:25
*** rossella_s has quit IRC23:29
*** rossella_s has joined #openstack-meeting-323:30
*** hongbin has quit IRC23:34
*** VW has joined #openstack-meeting-323:36
*** VW has quit IRC23:41
*** VW has joined #openstack-meeting-323:42
*** slaweq has joined #openstack-meeting-323:43
*** VW has quit IRC23:46
*** slaweq has quit IRC23:48

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!