| *** markguz has quit IRC | 00:03 | |
| *** itlinux has joined #openstack-containers | 00:23 | |
| *** Nel1x has joined #openstack-containers | 00:24 | |
| *** spotz has quit IRC | 01:04 | |
| *** yolanda_ has joined #openstack-containers | 01:05 | |
| *** georgem1 has joined #openstack-containers | 01:07 | |
| *** yolanda has quit IRC | 01:08 | |
| *** itlinux has quit IRC | 01:08 | |
| *** yolanda has joined #openstack-containers | 01:09 | |
| *** yolanda_ has quit IRC | 01:11 | |
| *** itlinux has joined #openstack-containers | 01:12 | |
| *** yolanda_ has joined #openstack-containers | 01:13 | |
| *** yolanda has quit IRC | 01:16 | |
| *** yolanda__ has joined #openstack-containers | 01:19 | |
| *** spotz has joined #openstack-containers | 01:20 | |
| *** yolanda_ has quit IRC | 01:22 | |
| *** mdnadeem has joined #openstack-containers | 01:23 | |
| *** georgem1 has quit IRC | 01:39 | |
| *** hongbin has joined #openstack-containers | 02:06 | |
| *** AlexeyAbashkin has joined #openstack-containers | 02:10 | |
| *** ricolin has joined #openstack-containers | 02:17 | |
| *** Nel1x has quit IRC | 02:21 | |
| *** AlexeyAbashkin has quit IRC | 02:39 | |
| *** lpetrut has joined #openstack-containers | 03:49 | |
| *** udesale has joined #openstack-containers | 03:53 | |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: Make etcd loadbalancer optional https://review.openstack.org/574540 | 04:00 |
|---|---|---|
| *** hongbin has quit IRC | 04:27 | |
| *** nguyenhai has left #openstack-containers | 04:31 | |
| *** lpetrut has quit IRC | 04:32 | |
| *** AlexeyAbashkin has joined #openstack-containers | 04:32 | |
| *** ramishra has joined #openstack-containers | 04:34 | |
| *** ykarel has joined #openstack-containers | 05:00 | |
| *** janki has joined #openstack-containers | 05:00 | |
| *** AlexeyAbashkin has quit IRC | 05:25 | |
| *** openstackgerrit has quit IRC | 05:34 | |
| *** ozerovandrei has quit IRC | 05:34 | |
| *** iranzo has joined #openstack-containers | 05:44 | |
| *** openstackstatus has quit IRC | 05:51 | |
| *** openstackstatus has joined #openstack-containers | 05:51 | |
| *** ChanServ sets mode: +v openstackstatus | 05:51 | |
| *** yolanda__ is now known as yolanda | 05:59 | |
| *** flwang1 has quit IRC | 06:06 | |
| *** ramishra has quit IRC | 06:10 | |
| *** ramishra has joined #openstack-containers | 06:12 | |
| *** mjura has joined #openstack-containers | 06:16 | |
| *** lpetrut has joined #openstack-containers | 06:28 | |
| *** AlexeyAbashkin has joined #openstack-containers | 06:29 | |
| *** openstackgerrit has joined #openstack-containers | 06:30 | |
| openstackgerrit | Merged openstack/magnum stable/queens: Allow multimaster lb with no floating ip option https://review.openstack.org/578236 | 06:30 |
| *** ramishra has quit IRC | 06:35 | |
| *** ramishra has joined #openstack-containers | 06:37 | |
| *** AlexeyAbashkin has quit IRC | 06:41 | |
| openstackgerrit | Merged openstack/magnum master: Rename scripts https://review.openstack.org/562454 | 06:45 |
| *** ramishra has quit IRC | 06:55 | |
| *** ramishra has joined #openstack-containers | 07:00 | |
| *** serlex has joined #openstack-containers | 07:01 | |
| *** peereb has joined #openstack-containers | 07:09 | |
| *** pcaruana has joined #openstack-containers | 07:10 | |
| *** mago_ has joined #openstack-containers | 07:22 | |
| *** ramishra has quit IRC | 07:23 | |
| *** ramishra has joined #openstack-containers | 07:24 | |
| *** ramishra has quit IRC | 07:29 | |
| *** ramishra has joined #openstack-containers | 07:30 | |
| *** yolanda_ has joined #openstack-containers | 07:35 | |
| *** yolanda is now known as Guest63448 | 07:37 | |
| *** yolanda_ is now known as yolanda | 07:37 | |
| *** mgoddard has joined #openstack-containers | 07:37 | |
| *** Guest63448 has quit IRC | 07:38 | |
| *** ramishra has quit IRC | 07:39 | |
| *** yasemin has quit IRC | 07:44 | |
| openstackgerrit | Merged openstack/magnum master: Add option to specify Cgroup driver for Kubelet https://review.openstack.org/571583 | 07:49 |
| *** ispp has joined #openstack-containers | 07:54 | |
| strigazi | flwang: are you here? | 07:55 |
| *** ykarel is now known as ykarel|lunch | 08:03 | |
| *** threestrands has joined #openstack-containers | 08:06 | |
| *** threestrands has quit IRC | 08:06 | |
| *** threestrands has joined #openstack-containers | 08:06 | |
| *** threestrands has quit IRC | 08:13 | |
| openstackgerrit | Bharat Kunwar proposed openstack/magnum stable/queens: Add option to specify Cgroup driver for Kubelet https://review.openstack.org/578725 | 08:13 |
| *** mgoddard has quit IRC | 08:20 | |
| *** ramishra has joined #openstack-containers | 08:25 | |
| openstackgerrit | Bharat Kunwar proposed openstack/magnum master: Support disabling floating IPs in swarm mode https://review.openstack.org/571200 | 08:30 |
| openstackgerrit | Bharat Kunwar proposed openstack/magnum master: Support disabling floating IPs in swarm mode https://review.openstack.org/571200 | 08:32 |
| openstackgerrit | Bharat Kunwar proposed openstack/magnum master: Support disabling floating IPs in swarm mode https://review.openstack.org/571200 | 08:33 |
| *** flwang1 has joined #openstack-containers | 08:36 | |
| openstackgerrit | Bharat Kunwar proposed openstack/magnum stable/queens: Add option to specify Cgroup driver for Kubelet https://review.openstack.org/578725 | 08:40 |
| openstackgerrit | Bharat Kunwar proposed openstack/magnum stable/queens: Add option to specify Cgroup driver for Kubelet https://review.openstack.org/578725 | 08:43 |
| *** armaan has joined #openstack-containers | 09:17 | |
| *** salmankhan has joined #openstack-containers | 09:17 | |
| *** ykarel|lunch is now known as ykarel | 09:23 | |
| openstackgerrit | David Rabel proposed openstack/magnum master: Docs: Replace non-existing command https://review.openstack.org/578741 | 09:28 |
| *** pcichy has joined #openstack-containers | 09:30 | |
| *** pcichy has quit IRC | 09:35 | |
| *** pcichy has joined #openstack-containers | 09:35 | |
| *** rcernin has quit IRC | 09:56 | |
| *** armaan has quit IRC | 10:02 | |
| *** armaan has joined #openstack-containers | 10:02 | |
| *** armaan has quit IRC | 10:03 | |
| *** armaan has joined #openstack-containers | 10:04 | |
| *** ispp has quit IRC | 10:25 | |
| *** armaan has quit IRC | 10:26 | |
| *** rabel has joined #openstack-containers | 10:26 | |
| *** armaan has joined #openstack-containers | 10:27 | |
| rabel | hi. i get a "Unknown attributes for argument cluster: master_flavor_id, docker_volume_size, flavor_id (HTTP 400)" error from magnum (ocata) when openstack coe cluster create with the newest openstackclient and magnumclient from pypi. all three parameters are already specified in the cluster template. what am i doing wrong? | 10:29 |
| rabel | additionally specifying them in the openstack coe cluster create does not help either | 10:30 |
| strigazi | rabel: you use a newer version of the client than the server | 10:31 |
| rabel | strigazi: is that a problem? i thought that is why there are api versions and microversions | 10:31 |
| *** armaan has quit IRC | 10:31 | |
| *** udesale has quit IRC | 10:32 | |
| strigazi | rabel: i think so, you can try with the ocata client to veridy | 10:33 |
| strigazi | rabel: i think so, you can try with the ocata client to verify | 10:33 |
| openstackgerrit | David Rabel proposed openstack/magnum master: Remove wrong blank line in doc/source/user/index.rst https://review.openstack.org/578758 | 10:43 |
| *** peereb has quit IRC | 10:43 | |
| *** ramishra has quit IRC | 10:44 | |
| rabel | strigazi: seems that ocata version of magnumclient is noch integrated in openstackclient. so i work directly with magnumclient now. and the error message has changed, which is good :) | 10:44 |
| *** mdnadeem has quit IRC | 10:45 | |
| *** salmankhan has quit IRC | 10:45 | |
| *** ispp has joined #openstack-containers | 10:49 | |
| *** zul has joined #openstack-containers | 11:39 | |
| *** pcichy has quit IRC | 11:43 | |
| ftersin | strigazi: hi. are you around? | 11:43 |
| strigazi | ftersin: yes | 11:45 |
| ftersin | strigazi: could you please explain the usage of kube:kube in https://github.com/openstack/magnum/commit/faa9e90402bcf78acdd166198fff9612fa8be81c#diff-bcbe4fe100320ff5c4fd8d35ef9b068eR13 | 11:46 |
| ftersin | there is no such user in that container, so it does not work for me | 11:46 |
| ftersin | do i miss something? | 11:47 |
| *** zul has quit IRC | 11:49 | |
| *** serlex has quit IRC | 11:49 | |
| strigazi | ftersin: let me check, it seems that you are right. I was able to sign certs with kubernetes, give me a sec | 11:49 |
| *** ramishra has joined #openstack-containers | 11:50 | |
| strigazi | ftersin: the deployment is failing or just this line doesn't work? | 11:51 |
| *** ramishra has quit IRC | 11:51 | |
| *** ramishra has joined #openstack-containers | 11:51 | |
| ftersin | strigazi: my version quite differs from the upstream and it fails there. but i think upstream just doesn't work in this line | 11:56 |
| *** pcaruana has quit IRC | 11:56 | |
| ftersin | it writes: [INFO] {"deploy_stdout": "", "deploy_stderr": "chown: invalid user: 'kube:kube'\n", "deploy_status_code": 0} | 11:56 |
| ftersin | strigazi: http://paste.openstack.org/show/724517/ | 11:57 |
| openstackgerrit | Merged openstack/magnum stable/queens: Add option to specify Cgroup driver for Kubelet https://review.openstack.org/578725 | 12:05 |
| *** sheel has joined #openstack-containers | 12:06 | |
| strigazi | ftersin: you are right this shouldn't work. I wonder why it works in my env. I'll check later. Just delete these two lines chmod and chown to unblock yourself. I'll check my env | 12:13 |
| *** Nel1x has joined #openstack-containers | 12:14 | |
| ftersin | strigazi: ok. should i file a bug for that? | 12:15 |
| *** markguz has joined #openstack-containers | 12:17 | |
| *** markguz has quit IRC | 12:22 | |
| *** zul has joined #openstack-containers | 12:37 | |
| strigazi | ftersin yes please, a story in storyboard.openstack.org | 12:40 |
| *** pcaruana has joined #openstack-containers | 12:44 | |
| *** Nel1x has quit IRC | 12:48 | |
| ftersin | strigazi: thank you. the bug: https://storyboard.openstack.org/#!/story/2002748 | 12:51 |
| strigazi | ftersin: thank you | 12:51 |
| *** rcernin has joined #openstack-containers | 12:53 | |
| *** serlex has joined #openstack-containers | 12:56 | |
| *** chhagarw has joined #openstack-containers | 13:04 | |
| *** chhavi__ has joined #openstack-containers | 13:04 | |
| *** rcernin has quit IRC | 13:08 | |
| *** georgem1 has joined #openstack-containers | 13:09 | |
| *** markguz has joined #openstack-containers | 13:12 | |
| *** chhagarw has quit IRC | 13:18 | |
| *** udesale has joined #openstack-containers | 13:20 | |
| *** markguz has quit IRC | 13:24 | |
| *** hongbin has joined #openstack-containers | 13:50 | |
| *** markguz has joined #openstack-containers | 13:52 | |
| *** ykarel is now known as ykarel|afk | 13:59 | |
| *** lpetrut has quit IRC | 14:03 | |
| *** ykarel|afk has quit IRC | 14:04 | |
| *** mjura has quit IRC | 14:12 | |
| *** ispp has quit IRC | 14:14 | |
| rabel | is magnumclient integration into the openstackclient only in master branch? | 14:23 |
| strigazi | rabel: in pike http://git.openstack.org/cgit/openstack/python-magnumclient/tree/magnumclient/osc/v1?h=stable/pike | 14:24 |
| *** itlinux has quit IRC | 14:26 | |
| rabel | strigazi: thanks. so with magnumclient and openstackclient from pike something like openstack cluster list should work, right? | 14:28 |
| rabel | ah, sorry. my fault. yes it works :O) | 14:30 |
| *** lpetrut has joined #openstack-containers | 14:38 | |
| *** markguz has quit IRC | 14:40 | |
| *** markguz has joined #openstack-containers | 14:41 | |
| *** markguz has quit IRC | 14:45 | |
| *** markguz has joined #openstack-containers | 14:46 | |
| *** imdigitaljim has quit IRC | 14:52 | |
| *** mago_ has quit IRC | 14:58 | |
| rabel | strigazi: back to what we were talking about a few hours ago: so the problem seems to be that openstackclient passes parameters to the api that i do not specifiy, with value null. https://git.openstack.org/cgit/openstack/python-magnumclient/tree/magnumclient/osc/v1/clusters.py#n123 | 15:02 |
| rabel | could we fix that by only adding only the parameters to args that are not null? | 15:02 |
| strigazi | rabel: I think this was fixed | 15:03 |
| strigazi | rabel: no, it isn't in osc | 15:04 |
| strigazi | rabel: for labels it is https://github.com/openstack/python-magnumclient/blob/master/magnumclient/osc/v1/clusters.py#L136 | 15:04 |
| rabel | strigazi: yes, if we could do that for the other three parameters, too. it would be perfekt | 15:05 |
| rabel | strigazi: i can upload a change later, if you want me to | 15:06 |
| strigazi | rabel: go for it | 15:06 |
| rabel | :) | 15:07 |
| *** itlinux has joined #openstack-containers | 15:11 | |
| *** yamamoto has quit IRC | 15:12 | |
| *** yamamoto has joined #openstack-containers | 15:12 | |
| openstackgerrit | Jim Bach proposed openstack/magnum master: Masters are now given the role label for master https://review.openstack.org/576623 | 15:15 |
| openstackgerrit | David Rabel proposed openstack/python-magnumclient master: osc: Don't pass parameters with null value https://review.openstack.org/578843 | 15:18 |
| *** imdigitaljim has joined #openstack-containers | 15:18 | |
| *** digitaljim has joined #openstack-containers | 15:18 | |
| *** digitaljim has left #openstack-containers | 15:19 | |
| *** imdigitaljim has quit IRC | 15:19 | |
| *** imdigitaljim has joined #openstack-containers | 15:19 | |
| *** digitaljim has joined #openstack-containers | 15:19 | |
| *** ftersin has quit IRC | 15:19 | |
| *** digitaljim has quit IRC | 15:19 | |
| imdigitaljim | strigazi: you here still? | 15:20 |
| strigazi | imdigitaljim: yeap | 15:20 |
| imdigitaljim | thanks for checking with flannel i was curious with the behavior | 15:21 |
| imdigitaljim | like I was saying a bit ago, i understand what you meant with the previous discussion. i think until we restore kubelet ill have to use that conditional | 15:21 |
| imdigitaljim | im gonna work on converging all cni deployments here shortly to make it easier going forward | 15:22 |
| imdigitaljim | so the only variation is the cni | 15:22 |
| strigazi | imdigitaljim I was thinking the following | 15:22 |
| strigazi | imdigitaljim: we could run configure minion everywhere to have the kubelet everywhere | 15:23 |
| strigazi | imdigitaljim: then we wouldn't have this issue | 15:23 |
| strigazi | imdigitaljim: it is not a small a change it would clean things up | 15:24 |
| strigazi | imdigitaljim: it is not a small a change, but it would clean things up | 15:24 |
| imdigitaljim | hmmm yeah that could be a start until we do some refactor for naming/etc | 15:24 |
| imdigitaljim | but yeah that could be the quick fix | 15:24 |
| strigazi | imdigitaljim: Then I want to move flannel to the hosted deployment that uses the kubernetes api for storage | 15:25 |
| imdigitaljim | flannel still works with etcd right? | 15:26 |
| imdigitaljim | should we make it optional? | 15:26 |
| imdigitaljim | but that would be a neat feature option | 15:26 |
| strigazi | imdigitaljim: yes, it still works with etcd | 15:27 |
| strigazi | imdigitaljim: I think we kube-api only is better | 15:27 |
| imdigitaljim | fair enough, i think that would be a great option to allow :) | 15:28 |
| imdigitaljim | maybe we can make it available for all and not CNI dependent | 15:28 |
| imdigitaljim | calico can do the same | 15:28 |
| strigazi | calico docs are strange for this | 15:29 |
| imdigitaljim | https://docs.projectcalico.org/v3.0/getting-started/kubernetes/installation/hosted/kubernetes-datastore/ | 15:29 |
| strigazi | in canal they say go for kube-api | 15:29 |
| imdigitaljim | i think we should consider writing it in a generic way as much as possible | 15:29 |
| strigazi | imdigitaljim: IMO, every CNI should have its own deployment script. Isn't that generic enough? | 15:34 |
| *** lpetrut has quit IRC | 15:36 | |
| imdigitaljim | that sounds great for configuring the .yaml deployment but major control plane changes (etcd/kubelet/proxy) etc should all be left intact | 15:36 |
| *** lpetrut has joined #openstack-containers | 15:37 | |
| imdigitaljim | etcd vs kubeapi as well | 15:37 |
| strigazi | imdigitaljim: oh yes, of course | 15:37 |
| strigazi | imdigitaljim: but etcd kube-api is left in the deployment method of the cni | 15:37 |
| imdigitaljim | thats the one id say can be still chosen separate of the CNI | 15:38 |
| imdigitaljim | but for now id say using it with just flannel would be a great start | 15:38 |
| imdigitaljim | as in baking it into flannel now sounds good and we can revisit it later making it more generic | 15:39 |
| *** lxkong has quit IRC | 15:39 | |
| *** ispp has joined #openstack-containers | 15:40 | |
| imdigitaljim | its sounds like a good path forward :] | 15:40 |
| *** ykarel has joined #openstack-containers | 15:45 | |
| *** iranzo has quit IRC | 15:46 | |
| *** sheel has quit IRC | 15:55 | |
| *** udesale has quit IRC | 15:56 | |
| *** udesale has joined #openstack-containers | 15:56 | |
| *** lpetrut has quit IRC | 16:05 | |
| *** ispp has quit IRC | 16:20 | |
| *** janki has quit IRC | 16:25 | |
| *** raopajay has quit IRC | 16:26 | |
| *** udesale has quit IRC | 16:31 | |
| *** ramishra has quit IRC | 16:34 | |
| *** jmlowe has quit IRC | 16:41 | |
| openstackgerrit | David Rabel proposed openstack/python-magnumclient master: osc: Don't pass parameters with null value https://review.openstack.org/578843 | 16:53 |
| *** jmlowe has joined #openstack-containers | 16:56 | |
| strigazi | imdigitaljim: flwang are we going to have a meeting? | 16:58 |
| flwang1 | strigazi: i'm around | 16:59 |
| *** ramishra has joined #openstack-containers | 16:59 | |
| strigazi | imdigitaljim: ping | 16:59 |
| imdigitaljim | yeah im here | 16:59 |
| strigazi | #startmeeting containers | 17:00 |
| openstack | Meeting started Thu Jun 28 17:00:44 2018 UTC and is due to finish in 60 minutes. The chair is strigazi. Information about MeetBot at http://wiki.debian.org/MeetBot. | 17:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 17:00 |
| *** openstack changes topic to " (Meeting topic: containers)" | 17:00 | |
| openstack | The meeting name has been set to 'containers' | 17:00 |
| strigazi | #topic Roll Call | 17:00 |
| *** openstack changes topic to "Roll Call (Meeting topic: containers)" | 17:00 | |
| strigazi | o/ | 17:01 |
| imdigitaljim | o/ | 17:01 |
| flwang1 | o/ | 17:01 |
| colin- | hello | 17:01 |
| jslater | o/ | 17:01 |
| colin- | \o | 17:01 |
| strigazi | #topic Review Action Items | 17:02 |
| *** openstack changes topic to "Review Action Items (Meeting topic: containers)" | 17:02 | |
| strigazi | strigazi to summarize the meeting in a ML mail DONE | 17:02 |
| strigazi | #link http://lists.openstack.org/pipermail/openstack-dev/2018-June/131769.html | 17:04 |
| strigazi | I had two more, to create stories but I wanted to confirm before creating noise in the already heavy storyboard | 17:05 |
| strigazi | We can discuss them in: | 17:05 |
| strigazi | #topic Blueprints/Bugs/Ideas | 17:05 |
| *** openstack changes topic to "Blueprints/Bugs/Ideas (Meeting topic: containers)" | 17:05 | |
| strigazi | The first one is pretty clear and it is about moving kubelet/proxy on all nodes. | 17:06 |
| flwang1 | s/moving/having ? | 17:06 |
| strigazi | We had a quick chat with imdigitaljim. I propose to run the minion configuration on all nodes | 17:06 |
| strigazi | yes s/moving/having/ | 17:06 |
| imdigitaljim | im gonna post a refactor draft here soon on the k8s bootstrapping files, please discuss changes on it. =) if we like it ill continue forward and validate/enhance it. | 17:07 |
| imdigitaljim | quick fix minion configuration sounds good | 17:07 |
| imdigitaljim | longterm needs to be cleaner | 17:07 |
| strigazi | what does quickfix mean? | 17:08 |
| strigazi | Which changes? | 17:08 |
| flwang1 | i would like to postpone the changing to Stein | 17:08 |
| flwang1 | not in Rocky | 17:08 |
| imdigitaljim | thats fine, its just for visibility, doesnt have to be accepted | 17:09 |
| strigazi | flwang1: So only small patches? | 17:09 |
| flwang1 | it's milestone 3 now | 17:09 |
| flwang1 | i'm not sure if we have enough time to test every corner | 17:09 |
| strigazi | ok | 17:10 |
| imdigitaljim | adding the minion configuration to master should work though, lets try it out | 17:11 |
| flwang1 | i just wanna to see a stable/rocky | 17:12 |
| strigazi | The current situation that we configure kubelet in the master is a bit comlplicated. If the change is not big we can go for it | 17:12 |
| flwang1 | besides, k8s 1.11 just released, we may need to bump the k8s version in rocky | 17:13 |
| strigazi | flwang1: this is irrelevant | 17:13 |
| flwang1 | strigazi: if there is a patch in next 2 weeks, i'm happy to review | 17:13 |
| strigazi | changing the kube verision is easy and we can't make decisions based on that | 17:14 |
| flwang1 | yep, 1.11 is not releated, just thinking a loud to avoid i forgot | 17:14 |
| strigazi | we did that in the past, config should be generic enough | 17:14 |
| strigazi | ok, let's see a patch and we decide in the coming weeks. | 17:15 |
| flwang1 | Jul 23 - Jul 27R-5Rocky-3 milestone | 17:15 |
| strigazi | two weeks | 17:15 |
| flwang1 | we have about 4 weeks for R-3 | 17:15 |
| flwang1 | or for the worst case, if we can have a patch in good shape by the end of july, i'm happy | 17:16 |
| strigazi | The next change would be upgrading the heat templates to pike | 17:16 |
| strigazi | how we should make this decision? | 17:17 |
| strigazi | flwang1: imdigitaljim what heat version are you running? | 17:17 |
| flwang1 | strigazi: we will be on queens shortly | 17:18 |
| strigazi | should we fork to _v2 drivers for this change? | 17:18 |
| flwang1 | strigazi: that would be better to physically isolate the changes | 17:19 |
| flwang1 | so yes | 17:19 |
| imdigitaljim | we're still using the juno ones shipped with magnum | 17:19 |
| imdigitaljim | a v2 would be good for stability/going forward as well | 17:19 |
| *** lpetrut has joined #openstack-containers | 17:20 | |
| strigazi | I'm in favor of v2 but keep the scripts common, thoughts? | 17:20 |
| colin- | yeah that sounds resaonable | 17:20 |
| flwang1 | scripts is fine | 17:21 |
| strigazi | ok then | 17:21 |
| flwang1 | and if we do that, i would suggest us resondiering the folder structure | 17:21 |
| strigazi | flwang1: what do you mean? | 17:21 |
| strigazi | not k8s_fedora_atomic_v2? | 17:22 |
| flwang1 | i'm happy with k8s_fedora_atomic_v2 | 17:23 |
| strigazi | ok, one moment | 17:23 |
| cbrumm | would moving to a self hosted control plane be possible in a v2 driver? | 17:23 |
| imdigitaljim | ^ | 17:23 |
| flwang1 | i'm talking about magnum/drivers/common/templates/kubernetes/fragments | 17:24 |
| flwang1 | i'm a bit weird for me though i haven't got a better idea yet | 17:24 |
| flwang1 | s/i'm/it's | 17:24 |
| imdigitaljim | if you're not familiar with self-hosted | 17:24 |
| imdigitaljim | https://coreos.com/blog/self-hosted-kubernetes.html | 17:24 |
| imdigitaljim | heres a decent doc to explain | 17:24 |
| imdigitaljim | kubernetes is moving towards self-hosted model | 17:24 |
| imdigitaljim | but in other words your control plane containers run in k8s instead of as systemd | 17:25 |
| strigazi | #link https://storyboard.openstack.org/#!/story/2002750 | 17:25 |
| imdigitaljim | just the kubelet would be systemd | 17:25 |
| *** ricolin has quit IRC | 17:25 | |
| strigazi | we could do that, but from experience the benefit is not great | 17:25 |
| flwang1 | imdigitaljim: we did discussed that way | 17:26 |
| flwang1 | and we even want to drop master nodes from user's tenant, no mater it's in vm or container | 17:26 |
| *** ramishra has quit IRC | 17:26 | |
| colin- | for my part i've enjoyed being able to interact with all components of the control plane via the API, much in the same patern the project encourages elsewhere for self-managing resources | 17:27 |
| flwang1 | imdigitaljim: and i agree with strigazi, we need to understand the pros and cons | 17:27 |
| strigazi | colin-: I agree on it, it is much easier | 17:27 |
| strigazi | but | 17:27 |
| strigazi | streamlining the process we deliver kubelet is more important and if we do it for kubelet we can do it for all components | 17:28 |
| imdigitaljim | control plane updates in place, improved HA, better introspection on the cluster, | 17:28 |
| colin- | ok | 17:28 |
| imdigitaljim | control plane access through k8s API | 17:28 |
| cbrumm | I'm sure a doc could be put together to help weigh the idea | 17:29 |
| *** kbyrne has quit IRC | 17:29 | |
| imdigitaljim | #link https://coreos.com/blog/self-hosted-kubernetes.html | 17:29 |
| strigazi | imdigitaljim: colin- cbrumm we had this actually before. | 17:29 |
| *** kbyrne has joined #openstack-containers | 17:29 | |
| strigazi | The reason we shifted from was that the way to deliver the same software kubernetes whose being fragmented. | 17:30 |
| strigazi | The reason we shifted from was that the way to deliver the same software kubernetes was fragmented. | 17:30 |
| strigazi | I'm not against it in principle, but at the moment it would complicate things | 17:31 |
| cbrumm | ok, I just figured that if a v2 was being considered it might be a good time to bring the control plane in line with kubernetes current recommendations. Use kubeadm etc | 17:31 |
| cbrumm | but it's not small consideration for sure | 17:32 |
| flwang1 | those kinds of things really need specs | 17:32 |
| strigazi | self-hosted kubernetes would be certainly done with kubeamd. | 17:32 |
| imdigitaljim | i think it could cut our bootstrapping code into about 1/4 of what it is | 17:32 |
| flwang1 | we can't make decision here | 17:32 |
| *** ykarel has quit IRC | 17:32 | |
| cbrumm | flwang1 agreed | 17:32 |
| imdigitaljim | ^ | 17:32 |
| strigazi | so, we need a detailed proposal and target it to S | 17:34 |
| cbrumm | that sounds good | 17:34 |
| cbrumm | I can't commit that we'll get to that immediately, but if we can't then it's only our fault if it slips | 17:35 |
| strigazi | #link https://storyboard.openstack.org/#!/story/2002751 | 17:35 |
| cbrumm | thanks strigazi | 17:36 |
| openstackgerrit | Jim Bach proposed openstack/magnum master: Kubernetes Bootrapping IDEAS (not intended as PR) https://review.openstack.org/578945 | 17:36 |
| strigazi | next, I have one thing we need for Rocky and discovered this week | 17:38 |
| strigazi | While doing some benchmarks with spark and kubernetes. we fould out that the way magnum is interacting with heat, actually hurts heat a lot. | 17:39 |
| imdigitaljim | oh? | 17:39 |
| imdigitaljim | what does it do? | 17:39 |
| strigazi | In magnum, we don't have locking between the conductor | 17:39 |
| strigazi | and when the conductor(s) try to sync the status with heat they do a stack get. | 17:40 |
| strigazi | When a stack is CREATE|UDPATE_IN_PROGRESS we only need the status | 17:40 |
| strigazi | not the outputs of the stack or any metadata. | 17:41 |
| cbrumm | Yeah that sounds like overkill | 17:41 |
| strigazi | For large stacks, evaluating the stack outputs is expensive | 17:41 |
| colin- | have seen similar patterns in other projects | 17:41 |
| colin- | is there a better query? | 17:42 |
| strigazi | and when many processes hit the heat api and queue requestsm rabbit gets overloaded and eventually heat goes down | 17:42 |
| strigazi | The better query is heat list and filter of one id | 17:42 |
| imdigitaljim | that sounds like a great patch to get done | 17:42 |
| flwang1 | strigazi: can we use stack list instead of get? | 17:42 |
| flwang1 | and then matching the id in magnum | 17:42 |
| strigazi | filters={"id": some_stack_id} | 17:42 |
| colin- | nice | 17:43 |
| flwang1 | strigazi: can the fileters accept a list of id? | 17:43 |
| strigazi | a list takes a second, a get takes a minute for a cluster with 200 nodes | 17:43 |
| strigazi | yeap tested it | 17:43 |
| flwang1 | but anyway, we should be able to do that | 17:43 |
| flwang1 | strigazi: cool, have a patch already? | 17:43 |
| flwang1 | i'm keen to review it | 17:43 |
| strigazi | I'll push a patch I have it almost ready | 17:44 |
| imdigitaljim | another issue with heat/magnum cleanup. if the cloud-controller-manager allocates something like a loadbalancer and the cluster is deleted.. the LB is dangling and not cleaned up because heat nor magnum know about it as it stands | 17:44 |
| strigazi | There is a patch in review | 17:44 |
| strigazi | Can you dig it? I can send it you too | 17:44 |
| strigazi | Can you dig it? I can send it to you too | 17:44 |
| flwang1 | strigazi: the heat issue? sure | 17:45 |
| strigazi | flwang1: the lbaas clean up | 17:45 |
| strigazi | for the heat issue I'll push it | 17:45 |
| flwang1 | in our testing, we can see the lb is deleted successfully | 17:45 |
| imdigitaljim | is this allocated through cloud-controller-manager? | 17:46 |
| strigazi | for lbass (created by the cloud provider) | 17:46 |
| flwang1 | without cloud-controller-manager | 17:46 |
| imdigitaljim | using kubernetes | 17:46 |
| strigazi | there is a patch | 17:46 |
| imdigitaljim | yeah the normal LB's allocated by heat are fine | 17:46 |
| *** jmlowe has quit IRC | 17:46 | |
| strigazi | give me a sec | 17:46 |
| flwang1 | we haven't tried that way | 17:46 |
| flwang1 | i'm happy to test and fix it | 17:46 |
| imdigitaljim | in other words, kubernetes users need to use load balancers | 17:46 |
| imdigitaljim | kubernetes talks to neutron to get a LB | 17:46 |
| flwang1 | imdigitaljim: i see | 17:46 |
| flwang1 | it's for service LB not masters LB | 17:47 |
| strigazi | #link https://review.openstack.org/#/c/497144/ | 17:47 |
| imdigitaljim | neutron/octavia | 17:47 |
| imdigitaljim | correct | 17:47 |
| imdigitaljim | im not sure what all we can/should do to rectify it, we might want to setup a BP for that | 17:47 |
| flwang1 | strigazi: thanks, will dig it | 17:47 |
| strigazi | it's the linked one, have a look | 17:48 |
| imdigitaljim | oh yes | 17:48 |
| imdigitaljim | strigazi: ill review this https://review.openstack.org/#/c/497144/3/magnum/drivers/heat/driver.py | 17:48 |
| imdigitaljim | err https://review.openstack.org/#/c/497144/ | 17:48 |
| strigazi | imdigitaljim: thanks | 17:48 |
| strigazi | and quickly one more | 17:48 |
| imdigitaljim | great callout for it =) | 17:48 |
| strigazi | The keypait issue: | 17:49 |
| strigazi | The keypair issue: | 17:49 |
| strigazi | #link https://storyboard.openstack.org/#!/story/2002648 | 17:49 |
| strigazi | if you offer magnum to others and they want you to do things for them it's pretty important | 17:50 |
| strigazi | The idea is to create a keypait per stack | 17:50 |
| strigazi | The idea is to create a keypair per stack | 17:50 |
| strigazi | So when a client wants you to scale or fix their stack with stack update you should be able to do it. | 17:51 |
| strigazi | I'll push a patch for it as well. | 17:52 |
| strigazi | Apologies for this meeting folks, we can try to make the next one more structured. | 17:53 |
| colin- | no complaints here :) | 17:54 |
| imdigitaljim | its complaints here as well | 17:54 |
| imdigitaljim | s/its/no | 17:54 |
| *** lpetrut has quit IRC | 17:54 | |
| strigazi | :) | 17:54 |
| strigazi | Do tou need to bring up anything else? | 17:55 |
| flwang1 | i need testing from imdigitaljim for https://review.openstack.org/576029 | 17:56 |
| flwang1 | or any other bless for this one | 17:56 |
| cbrumm | Nope, I think we've covered everything | 17:56 |
| *** markguz has quit IRC | 17:56 | |
| imdigitaljim | flwang: im on it by the end of this week | 17:56 |
| imdigitaljim | i should post about it early next week | 17:57 |
| flwang1 | imdigitaljim: cool, thanks | 17:57 |
| *** markguz has joined #openstack-containers | 17:57 | |
| cbrumm | Thank you guys for hosting this again. It's great to get some solid time with you both | 17:57 |
| strigazi | Let's meet in 6 days and 23 hours again :) | 17:57 |
| cbrumm | sounds like a plan | 17:57 |
| flwang1 | good to discuss with you guys, very productive | 17:58 |
| strigazi | cbrumm: imdigitaljim colin- flwang1 jslater thanks for joining, see you soon | 17:58 |
| colin- | ttyl | 17:59 |
| flwang1 | ttyl | 17:59 |
| cbrumm | Have a good evening/day | 17:59 |
| strigazi | cheers! | 17:59 |
| strigazi | #endmeeting | 17:59 |
| *** openstack changes topic to "OpenStack Containers Team" | 17:59 | |
| openstack | Meeting ended Thu Jun 28 17:59:32 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:59 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/containers/2018/containers.2018-06-28-17.00.html | 17:59 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/containers/2018/containers.2018-06-28-17.00.txt | 17:59 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/containers/2018/containers.2018-06-28-17.00.log.html | 17:59 |
| *** markguz has quit IRC | 18:01 | |
| *** jmlowe has joined #openstack-containers | 18:08 | |
| *** pcaruana has quit IRC | 18:14 | |
| *** yolanda_ has joined #openstack-containers | 18:25 | |
| *** yolanda has quit IRC | 18:27 | |
| *** markguz has joined #openstack-containers | 18:28 | |
| *** markguz has quit IRC | 18:38 | |
| *** markguz has joined #openstack-containers | 18:39 | |
| *** armaan has joined #openstack-containers | 18:40 | |
| *** markguz has quit IRC | 18:43 | |
| *** armaan has quit IRC | 18:45 | |
| *** armaan has joined #openstack-containers | 18:45 | |
| openstackgerrit | David Rabel proposed openstack/python-magnumclient master: osc: Don't pass parameters with null value https://review.openstack.org/578843 | 18:47 |
| *** markguz_ is now known as markguz | 18:47 | |
| *** armaan has quit IRC | 18:50 | |
| *** chhavi__ has quit IRC | 18:58 | |
| *** itlinux has quit IRC | 19:29 | |
| *** itlinux has joined #openstack-containers | 19:31 | |
| *** georgem1 has quit IRC | 19:31 | |
| *** georgem1 has joined #openstack-containers | 19:31 | |
| *** georgem1 has quit IRC | 19:36 | |
| *** flwang1 has quit IRC | 19:58 | |
| *** georgem1 has joined #openstack-containers | 20:02 | |
| *** georgem1 has quit IRC | 20:04 | |
| *** georgem1 has joined #openstack-containers | 20:04 | |
| *** georgem1 has quit IRC | 20:04 | |
| *** zul has quit IRC | 20:10 | |
| *** itlinux has quit IRC | 20:28 | |
| *** itlinux has joined #openstack-containers | 20:42 | |
| openstackgerrit | Jim Bach proposed openstack/magnum master: Enable magnum to allow load balancers to be in their own subnet https://review.openstack.org/577570 | 21:00 |
| *** flwang1 has joined #openstack-containers | 21:04 | |
| imdigitaljim | flwang1: just tested https://review.openstack.org/#/c/576029/4 | 21:16 |
| imdigitaljim | works great | 21:16 |
| imdigitaljim | good job! | 21:16 |
| flwang | imdigitaljim: fantastic, thanks a lot | 21:17 |
| flwang | please update it as a comment to the patch, so that strigazi can be confident to merge it | 21:17 |
| flwang | as for the issue he mentioned in the review comments, i have already opened a new task to track it, i will add a comment too | 21:18 |
| flwang | cheers | 21:18 |
| itlinux | Hi Flwang: I have checked looks like the script on step 5 is failing to get the cert | 21:25 |
| itlinux | any tips on that.. | 21:25 |
| flwang | firstly, manually run the script to check the connection b/w your vm and your keystone/magnum | 21:26 |
| flwang | secondly, let's see the result of step #1 :) | 21:27 |
| *** serlex has quit IRC | 21:33 | |
| itlinux | ok sounds good thanks | 21:37 |
| *** rcernin has joined #openstack-containers | 21:42 | |
| flwang | cool | 21:42 |
| *** itlinux has quit IRC | 21:50 | |
| openstackgerrit | Rick Cano proposed openstack/magnum master: Fixing coreos driver - WIP https://review.openstack.org/579026 | 21:58 |
| *** lxkong has joined #openstack-containers | 22:04 | |
| *** hongbin has quit IRC | 22:16 | |
| *** linkmark has quit IRC | 22:38 | |
| *** threestrands has joined #openstack-containers | 23:05 | |
| *** threestrands has quit IRC | 23:05 | |
| *** threestrands has joined #openstack-containers | 23:05 | |
| *** yolanda__ has joined #openstack-containers | 23:07 | |
| *** yolanda_ has quit IRC | 23:11 | |
| *** yolanda__ has quit IRC | 23:18 | |
| *** threestrands has quit IRC | 23:38 | |
| *** masber has quit IRC | 23:48 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!