Wednesday, 2020-05-27

« Tuesday, 2020-05-26 Index Thursday, 2020-05-28 »
*** dave-mccowan has quit IRC00:19
*** dave-mccowan has joined #openstack-containers00:20
*** dave-mccowan has quit IRC00:30
openstackgerritFeilong Wang proposed openstack/magnum master: [WIP] Fix label fixed_network_cidr  https://review.opendev.org/73100000:54
*** ricolin has joined #openstack-containers00:57
*** flwang1 has quit IRC01:06
openstackgerritFeilong Wang proposed openstack/magnum master: Fix label fixed_network_cidr  https://review.opendev.org/73100001:16
*** dave-mccowan has joined #openstack-containers01:22
*** sapd1 has joined #openstack-containers02:21
*** rcernin has quit IRC02:47
*** rcernin has joined #openstack-containers02:50
openstackgerritMerged openstack/magnum master: [k8s] Use Helm v3 by default  https://review.opendev.org/72023403:05
*** mgariepy has quit IRC03:39
*** mgariepy has joined #openstack-containers03:42
*** dave-mccowan has quit IRC04:23
*** vishalmanchanda has joined #openstack-containers05:06
*** ianychoi has quit IRC05:34
*** udesale has joined #openstack-containers05:40
*** ykarel|away is now known as ykarel05:44
*** belmoreira has joined #openstack-containers07:01
openstackgerritSpyros Trigazis proposed openstack/magnum stable/ussuri: atomic: Do not install control-plane on minions  https://review.opendev.org/73109207:29
openstackgerritSpyros Trigazis proposed openstack/magnum stable/train: atomic: Do not install control-plane on minions  https://review.opendev.org/73109307:29
openstackgerritSpyros Trigazis proposed openstack/magnum stable/ussuri: atomic: Do not install control-plane on minions  https://review.opendev.org/73109207:31
openstackgerritSpyros Trigazis proposed openstack/magnum stable/ussuri: atomic: Do not install control-plane on minions  https://review.opendev.org/73109207:32
*** rcernin has quit IRC07:35
*** ttsiouts has joined #openstack-containers07:38
openstackgerritMerged openstack/magnum master: api: Do not guess based on name extension  https://review.opendev.org/73087707:57
*** ttsiouts has quit IRC08:01
*** ttsiouts has joined #openstack-containers08:10
*** born2bake has joined #openstack-containers08:11
*** guilhermesp_ has joined #openstack-containers08:12
*** mnaser_ has joined #openstack-containers08:13
*** brtknr has quit IRC08:20
*** mnaser has quit IRC08:20
*** guilhermesp has quit IRC08:20
*** mnaser_ is now known as mnaser08:20
*** guilhermesp_ is now known as guilhermesp08:20
openstackgerritMerged openstack/magnum master: Fix small issues rolling upgrade  https://review.opendev.org/72859208:27
*** brtknr has joined #openstack-containers08:28
*** ttsiouts has quit IRC08:51
*** flwang1 has joined #openstack-containers08:52
strigazio/08:57
flwang1#startmeeting magnum09:00
openstackMeeting started Wed May 27 09:00:03 2020 UTC and is due to finish in 60 minutes.  The chair is flwang1. Information about MeetBot at http://wiki.debian.org/MeetBot.09:00
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.09:00
*** openstack changes topic to " (Meeting topic: magnum)"09:00
openstackThe meeting name has been set to 'magnum'09:00
flwang1strigazi: hey, how are you09:00
strigaziall good09:00
*** ttsiouts has joined #openstack-containers09:01
flwang1brtknr: ping09:01
flwang1strigazi: let's wait brtknr a while09:01
strigazisure thing09:01
flwang1strigazi: i saw you proposed a topic about health status :)09:01
strigaziyeap, now that we can control it more we can enhance?09:03
flwang1i'm super happy to see you start to look into this09:03
flwang1strigazi: sure09:04
flwang1strigazi: should we start or wait a bit longer?09:04
strigaziWe can start from light things09:05
strigazieg the worklist bullet I added09:05
flwang1strigazi: no problem09:06
flwang1strigazi: which one you want to bring first?09:07
strigazi#topic using storyboard09:07
strigaziyou need to do the topic thing09:07
strigaziflwang1: ^^09:08
brtknrhello sorry was dealing with an emergency09:08
strigazibrtknr: all good now?09:08
flwang1#topic using storboard09:08
*** openstack changes topic to "using storboard (Meeting topic: magnum)"09:08
brtknrye kind of09:08
flwang1strigazi: i'm happy to use storyboard, but i'd like to understand if there is any rule/policy/process we need to follow09:09
strigaziTo be able to track what we do and what others ask us to do I started adding the magnum-victoria tag09:09
strigaziThe active stories at the moment are 7109:10
strigaziwe can easily review them now09:10
strigaziWe have two options IMO09:10
strigazieither let users add the tag (we can't control this)09:10
strigazior we add the stories be hand in a worklist (there are ACLs for this, I think)09:11
strigazifor example I created this worklist https://storyboard.openstack.org/#!/worklist/86509:11
strigazithat included all stories that that have the magnum-victoria tag09:11
strigaziand they are active09:12
brtknrstrigazi: flwang1 sorry guys i might have to drop out of the meeting today, i will catch up on the discussion later09:12
strigazibrtknr sure09:12
flwang1brtknr: no worries, take care09:12
brtknrhave a good day09:12
strigazibrtknr: cheers09:12
flwang1strigazi: can any user add story to that list?09:13
flwang1or only people got the permission?09:13
strigaziflwang1 if we use the tag, anyone09:14
strigaziflwang1: if we add stories manually to the worklist (only people with permission)09:14
flwang1ok, that's alright, i don't think much people will do that09:14
flwang1strigazi: and I think we can remove story by removing the tag from stories?09:15
strigaziflwang1: and here is a board with more lists https://storyboard.openstack.org/#!/board/21209:15
flwang1which one you prefer?09:15
strigaziflwang1 I think the tag makes sense at the moment09:15
flwang1agree09:15
flwang1it's easy to manage09:15
strigaziwe can review what we want to do and add the tag, then it will appear in the lists of the board09:16
flwang1at this stage, are you suggesting we use the magnum-victoria to track all the work for V?09:16
strigaziyes09:16
flwang1sounds good09:16
strigaziI have added you both in the board https://storyboard.openstack.org/#!/board/21209:17
flwang1can we do that individually and then we can go through on virtual PTG?09:17
strigaziwe can do it individually i think09:18
strigaziit is not much09:18
flwang1cool09:18
strigaziwe can present it at the PTF09:18
strigaziwe can present it at the PTG09:18
flwang1right, agree09:18
flwang1thanks for working on this09:18
strigaziideally we need go through the open stories here: https://storyboard.openstack.org/#!/project/openstack/magnum and add the tag if we want09:19
flwang1move on?09:19
flwang1anything else?09:19
strigazilet's move on09:20
flwang1yep, given you have closed a lot, so i assume that's not much09:20
strigaziyeah it is only recent things09:20
flwang1strigazi: which one you want to discuss next?  node/API  version in health_status_reason?09:20
strigazilet's do health first, it is trivial more or less09:21
flwang1#topic node/API  version in health_status_reason09:22
*** openstack changes topic to "node/API version in health_status_reason (Meeting topic: magnum)"09:22
flwang1strigazi: tell me more09:22
flwang1why do you want to add version? for NG09:22
strigazinot for NG09:22
strigaziWe want to have a view of alive clusters and which version are they running09:23
flwang1hmm... can't you get from the coe_version?09:23
strigaziwhat is what magnum expects to have in the cluster09:24
strigazior what it tried to have09:24
flwang1you mean coe_version or the health_status_reason?09:24
strigazicoe_version is the desired09:24
strigazihealth_status_reason? will be the "current"09:25
flwang1firstly, i think anything can help admin/user understand the health status can be put into the dict09:25
flwang1so i'm totally ok with that09:25
flwang1i'm just trying to understand the user case09:26
strigazithe biggest use case is old clusters that we don't know what is going on09:26
flwang1do you mean master and worker may run different version?09:26
flwang1ah, i see :D09:26
strigaziand clusters that the user sshed and did things09:27
flwang1because it has been upgraded  and we lost the versions?09:27
strigaziyeah, bith09:27
strigaziyeah, both09:27
flwang1right09:27
flwang1i'm ok with that09:27
flwang1throw a patch and i'm happy to review09:27
flwang1move on?09:28
strigaziwait09:28
strigaziI have the content in the etherpad09:28
strigaziand there is a follow up09:28
strigazithe dict we have09:28
strigaziit is not nested09:28
flwang1i'm ok with that format09:28
flwang1strigazi: we touched that topic before09:29
strigaziyeap09:29
flwang1we can do nested dict09:29
strigazibut we can't really do it09:29
flwang1why?09:29
strigaziit will be string, no?09:29
flwang1do you mean it will be saved in db as string?09:30
strigaziin db it will be string anyway09:30
strigaziI mean the type in he API09:30
flwang1right09:30
flwang1i see your point09:31
flwang1i need to do some test to double confirm09:31
flwang1but we probably can't do nested IIRC09:31
flwang1we can live with flat dict until we figure out a better way09:31
strigaziyeap, we can't do arbitrary depth or mix list and dict09:31
strigaziso what we do with the depth?09:32
strigazistrings that are escaped json?09:33
flwang1strigazi: hmm... that's ugly :(09:33
flwang1can we just use flat dict for now? i don't have a good answer tbh09:34
strigaziflwang1: and we have the issue for helm-config (there base64 makes some sense)09:34
strigaziflwang1 for health_status we can do flat09:34
flwang1ahhhhh09:34
strigazifor helm, we can not09:34
strigaziso to wrap the health_status subject09:35
*** ykarel is now known as ykarel|lunch09:35
strigaziyou are OK with having the version09:35
strigaziand have a flat dict for now, correct?09:35
flwang1yes09:35
strigazilet's switch to helm-config?09:35
flwang1for helm config, can we just read it as a escapsed string? and do the magic on server side?09:36
flwang1#topic helm-config09:36
*** openstack changes topic to "helm-config (Meeting topic: magnum)"09:36
strigazi#action striazi to propose a patch for adding node/API version in health_status_reason with a flat dict09:37
strigazi#undo09:37
strigazi#action strigazi to propose a patch for adding node/API version in health_status_reason with a flat dict09:37
flwang1#action strigazi to propose a patch for adding node/API version in health_status_reason with a flat dict09:37
strigaziflwang1: for helm-config, I don't know, some parsing will happen09:37
strigazieither to escape the JSON09:38
flwang1strigazi: can we use json.dumps and json.loads09:38
strigazior encode it ot base64 and pass it as is09:38
flwang1on the two sides, to make sure they're compatible09:38
strigaziflwang1: can we?09:38
flwang1don't know, just thinking aloud09:38
strigaziI don't know either09:39
*** threestrands has quit IRC09:39
flwang1for this piece, you guys probably need to a small PoC09:39
strigazithe relevant part is here: https://review.opendev.org/#/c/727756/4/specs/victoria/helm-config.rst@8709:39
flwang1i don't like the idea of base64, TBH09:40
strigazican you leave a comment with preference? I guess you propose to try escaped json09:41
flwang1sure, will do09:42
strigazimove on?09:43
flwang1comments added09:43
flwang1which next?09:43
strigazi#topic resize: Send only nodes_to_remove and node_count09:44
flwang1#topic resize: Send only nodes_to_remove and node_count09:44
*** openstack changes topic to "resize: Send only nodes_to_remove and node_count (Meeting topic: magnum)"09:44
strigazihttps://review.opendev.org/#/c/730868/09:44
flwang1i didn't get the issue09:44
strigaziresize need to do only resize09:45
strigaziso do only resize we need send only the node_count09:45
strigaziand which nodes to drop09:45
strigaziafter train, stein cluster were breaking09:45
strigazisomething similar can happen again (doesn't happen now)09:46
flwang1hmmm... that's the kind of patch i don't want to review :D09:46
flwang1it's too dangerous09:47
strigaziwhat is dangerous is what we have now09:47
strigazisee the commit message09:47
flwang1i understand.09:48
flwang1please give me some time to review it09:48
strigazinow we have this: https://review.opendev.org/#/c/642009/7/magnum/drivers/heat/driver.py@17609:48
flwang1strigazi: can you please add a steps how to reproduce the issue?09:48
strigaziflwang1: it now reproducible now (at least the same thing that I had issues with)09:49
strigaziflwang1: it not reproducible now (at least the same thing that I had issues with)09:50
strigazibut it was catastrophic i.e. it was replacing all nodes09:50
flwang1do you have mean i have to have a very old cluster to reproduce this?09:50
strigazistein09:50
strigazinot very old09:51
strigaziyou upgraded recently, no?09:51
flwang1yes09:51
strigaziso not that old09:51
flwang1let me ask more questions09:51
strigazicheck this part of the code you had:09:52
flwang1even if user only have the default-master and default-worker NGs, when doing resize,   all nodes will be replaced?09:52
strigazihttps://github.com/openstack/magnum/blob/stable/stein/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L70509:53
strigaziit really depends09:53
openstackgerritMerged openstack/magnum stable/ussuri: atomic: Do not install control-plane on minions  https://review.opendev.org/73109209:53
strigaziit depends on what is here: https://github.com/openstack/magnum/blob/stable/stein/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L51009:54
strigaziI don't know about your cherry-picks09:54
*** xinliang has joined #openstack-containers09:54
strigaziflwang1: take your time and review this method: https://review.opendev.org/#/c/730868/1/magnum/drivers/heat/driver.py@26409:55
strigaziit is very small and very very clear09:55
strigaziinstead of sending many things in this dict:09:56
flwang1i see the issue now09:56
strigaziheat_params.update(scale_params)09:56
strigazi'parameters': heat_params,09:56
flwang1you mean it will rebuild all the nodes of that NG?09:56
strigaziguarantee to end only two parameters09:56
strigazidepenfing on the the parameter both default NGS09:57
strigazimaybe it won't rebuild anything09:57
strigazibut maybe is not enough09:57
flwang1ok, i see. thanks for the heads up09:58
flwang1i will review it tomorrow09:59
strigaziit is not urgnet for master branch but we nee it avoid breaking things09:59
strigaziI think out time is up10:00
strigaziI think our time is up10:01
flwang1strigazi: right.10:01
flwang1it's a critical fix10:01
flwang1let's get it done asap10:01
flwang1#endmeeting10:01
*** openstack changes topic to "OpenStack Containers Team | Meeting: every Wednesday @ 9AM UTC | Agenda: https://etherpad.openstack.org/p/magnum-weekly-meeting"10:01
openstackMeeting ended Wed May 27 10:01:26 2020 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)10:01
openstackMinutes:        http://eavesdrop.openstack.org/meetings/magnum/2020/magnum.2020-05-27-09.00.html10:01
openstackMinutes (text): http://eavesdrop.openstack.org/meetings/magnum/2020/magnum.2020-05-27-09.00.txt10:01
openstackLog:            http://eavesdrop.openstack.org/meetings/magnum/2020/magnum.2020-05-27-09.00.log.html10:01
flwang1strigazi: pls help review https://review.opendev.org/#/c/724203/ when you have time10:01
strigaziflwang1:  will run it as soon as possible10:02
strigaziflwang1: let's get this in to ohttps://review.opendev.org/#/c/731093/10:03
strigazis/to/too/10:03
flwang1done10:03
flwang1i'm off for today, have a nice day10:04
strigazicheers10:05
*** flwang1 has quit IRC10:10
openstackgerritSpyros Trigazis proposed openstack/magnum stable/ussuri: k8s: Use the same kubectl version as API  https://review.opendev.org/73113610:12
openstackgerritSpyros Trigazis proposed openstack/magnum stable/train: k8s: Use the same kubectl version as API  https://review.opendev.org/73113710:13
openstackgerritSpyros Trigazis proposed openstack/magnum stable/ussuri: k8s: Add admin.conf kubeconfig  https://review.opendev.org/73113910:14
openstackgerritSpyros Trigazis proposed openstack/magnum stable/train: k8s: Add admin.conf kubeconfig  https://review.opendev.org/73114010:14
*** k_mouza has joined #openstack-containers10:16
*** xinliang has quit IRC10:29
*** ttsiouts_ has joined #openstack-containers10:31
*** ttsiouts has quit IRC10:33
*** udesale_ has joined #openstack-containers10:33
*** k_mouza has quit IRC10:34
*** udesale has quit IRC10:36
*** ykarel|lunch is now known as ykarel10:36
*** k_mouza has joined #openstack-containers10:47
openstackgerritMerged openstack/magnum stable/train: atomic: Do not install control-plane on minions  https://review.opendev.org/73109310:58
*** sapd1 has quit IRC11:01
*** ttsiouts_ has quit IRC11:02
*** ttsiouts has joined #openstack-containers11:10
openstackgerritDiogo Guerra proposed openstack/magnum stable/ussuri: [k8s] Expose autoscaler prometheus metrics  https://review.opendev.org/73116111:13
openstackgerritDiogo Guerra proposed openstack/magnum stable/ussuri: Scrape traefik and autoscaler metrics  https://review.opendev.org/73116211:14
openstackgerritDiogo Guerra proposed openstack/magnum stable/ussuri: Update prometheus monitoring chart and images  https://review.opendev.org/73116311:14
openstackgerritDiogo Guerra proposed openstack/magnum stable/ussuri: Scrape internal kubernetes components  https://review.opendev.org/73116411:14
openstackgerritDiogo Guerra proposed openstack/magnum stable/ussuri: [k8s] Use Helm v3 by default  https://review.opendev.org/73116511:14
*** k_mouza has quit IRC11:32
*** k_mouza has joined #openstack-containers11:32
*** k_mouza has quit IRC11:37
*** k_mouza has joined #openstack-containers11:45
openstackgerritSpyros Trigazis proposed openstack/magnum stable/ussuri: api: Do not guess based on name extension  https://review.opendev.org/73117411:51
openstackgerritSpyros Trigazis proposed openstack/magnum stable/train: api: Do not guess based on name extension  https://review.opendev.org/73117511:51
openstackgerritSpyros Trigazis proposed openstack/magnum stable/stein: api: Do not guess based on name extension  https://review.opendev.org/73117611:52
*** ykarel is now known as ykarel|afk11:59
*** ttsiouts has quit IRC12:00
*** ramishra has quit IRC12:10
*** k_mouza has quit IRC12:20
*** k_mouza has joined #openstack-containers12:22
*** ramishra has joined #openstack-containers12:24
*** ttsiouts has joined #openstack-containers12:28
*** k_mouza has quit IRC12:33
*** k_mouza has joined #openstack-containers12:34
*** ykarel|afk is now known as ykarel12:41
*** sapd1 has joined #openstack-containers12:53
*** ttsiouts has quit IRC12:56
*** ttsiouts has joined #openstack-containers13:01
*** jmlowe has quit IRC13:57
*** ttsiouts has quit IRC14:07
*** jmlowe has joined #openstack-containers14:31
*** ttsiouts has joined #openstack-containers14:43
*** ttsiouts has quit IRC14:48
*** jmlowe has quit IRC14:50
*** jmlowe has joined #openstack-containers14:55
*** jmlowe has quit IRC15:01
*** sapd1 has quit IRC15:10
*** jmlowe has joined #openstack-containers15:12
*** jmlowe has quit IRC15:18
*** ttsiouts has joined #openstack-containers15:19
*** jmlowe has joined #openstack-containers15:19
*** ykarel is now known as ykarel|away15:27
*** ttsiouts has quit IRC15:39
*** ttsiouts has joined #openstack-containers15:59
*** ttsiouts has quit IRC15:59
*** ttsiouts has joined #openstack-containers16:00
*** ttsiouts has quit IRC16:05
*** sapd1 has joined #openstack-containers16:17
*** udesale_ has quit IRC16:32
*** yolanda has quit IRC16:53
*** LuckyClover has joined #openstack-containers16:53
*** ricolin has quit IRC16:58
*** k_mouza has quit IRC16:59
*** LuckyClover85 has joined #openstack-containers17:00
*** LuckyClover85 has quit IRC17:01
*** LuckyClover has quit IRC17:04
*** LuckyClover has joined #openstack-containers17:06
*** ianychoi has joined #openstack-containers17:27
*** LuckyClover has quit IRC18:09
*** sapd1 has quit IRC18:43
*** k_mouza has joined #openstack-containers19:00
*** k_mouza has quit IRC19:05
*** k_mouza has joined #openstack-containers20:20
*** k_mouza has quit IRC20:24
*** belmoreira has quit IRC20:36
openstackgerritFeilong Wang proposed openstack/magnum stable/ussuri: Fix small issues rolling upgrade  https://review.opendev.org/73133920:52
openstackgerritFeilong Wang proposed openstack/magnum stable/ussuri: Support upgrade on behalf of user by admin  https://review.opendev.org/73134020:53
openstackgerritFeilong Wang proposed openstack/magnum stable/train: Fix small issues rolling upgrade  https://review.opendev.org/73135121:36
*** vishalmanchanda has quit IRC22:03
*** born2bake has quit IRC22:41
*** rcernin has joined #openstack-containers22:59
openstackgerritMerged openstack/magnum stable/ussuri: api: Do not guess based on name extension  https://review.opendev.org/73117423:21
openstackgerritMerged openstack/magnum stable/train: k8s: Add admin.conf kubeconfig  https://review.opendev.org/73114023:33
openstackgerritMerged openstack/magnum stable/ussuri: k8s: Use the same kubectl version as API  https://review.opendev.org/73113623:34
openstackgerritMerged openstack/magnum stable/train: k8s: Use the same kubectl version as API  https://review.opendev.org/73113723:34
« Tuesday, 2020-05-26 Index Thursday, 2020-05-28 »

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!