| *** rcernin_ has joined #openstack-containers | 00:02 | |
| *** rcernin has quit IRC | 00:03 | |
| *** rcernin has joined #openstack-containers | 00:29 | |
| *** rcernin has quit IRC | 00:29 | |
| *** rcernin has joined #openstack-containers | 00:30 | |
| *** rcernin_ has quit IRC | 00:32 | |
| *** flwang1 has quit IRC | 00:57 | |
| *** v1k0d3n has quit IRC | 00:57 | |
| *** kgz has quit IRC | 00:57 | |
| *** yankcrime has quit IRC | 00:57 | |
| *** flwang1 has joined #openstack-containers | 01:04 | |
| *** v1k0d3n has joined #openstack-containers | 01:04 | |
| *** kgz has joined #openstack-containers | 01:04 | |
| *** yankcrime has joined #openstack-containers | 01:04 | |
| *** openstackgerrit has quit IRC | 01:06 | |
| *** Nel1x has joined #openstack-containers | 01:13 | |
| *** hongbin has joined #openstack-containers | 02:01 | |
| *** markguz_ has joined #openstack-containers | 02:19 | |
| *** jaewook_oh has joined #openstack-containers | 02:25 | |
| *** ricolin has joined #openstack-containers | 02:38 | |
| *** ykarel|away has joined #openstack-containers | 03:02 | |
| *** Bhujay has joined #openstack-containers | 03:04 | |
| *** markguz_ has quit IRC | 03:05 | |
| *** Nel1x has quit IRC | 03:12 | |
| *** Nisha_Agarwal has joined #openstack-containers | 03:31 | |
| *** ykarel|away has quit IRC | 03:34 | |
| *** hongbin has quit IRC | 03:39 | |
| *** Bhujay has quit IRC | 03:40 | |
| *** udesale has joined #openstack-containers | 03:53 | |
| *** Nisha_ has joined #openstack-containers | 03:59 | |
| *** Nisha_Agarwal has quit IRC | 04:01 | |
| *** dave-mccowan has quit IRC | 04:12 | |
| *** Bhujay has joined #openstack-containers | 04:21 | |
| *** janki has joined #openstack-containers | 04:28 | |
| *** flwang1 has quit IRC | 04:53 | |
| *** ykarel|away has joined #openstack-containers | 05:05 | |
| *** ykarel|away is now known as ykarel | 05:11 | |
| *** Nisha_ has quit IRC | 05:28 | |
| *** Nisha_away has joined #openstack-containers | 05:29 | |
| *** ricolin has quit IRC | 06:43 | |
| *** pcaruana has joined #openstack-containers | 06:44 | |
| *** adrianc has joined #openstack-containers | 06:44 | |
| *** mattgo has joined #openstack-containers | 07:01 | |
| *** rcernin has quit IRC | 07:02 | |
| *** jaewook_oh has quit IRC | 07:28 | |
| *** salmankhan has joined #openstack-containers | 07:29 | |
| *** salmankhan has quit IRC | 07:33 | |
| *** jaewook_oh has joined #openstack-containers | 07:35 | |
| *** openstackgerrit has joined #openstack-containers | 07:37 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/magnum master: Imported Translations from Zanata https://review.openstack.org/590127 | 07:37 |
|---|---|---|
| *** janki is now known as janki|lunch | 07:52 | |
| *** jaewook_oh has quit IRC | 07:55 | |
| *** ykarel is now known as ykarel|lunch | 07:56 | |
| *** Bhujay has quit IRC | 08:10 | |
| *** openstackstatus has quit IRC | 08:12 | |
| *** ktibi has joined #openstack-containers | 08:13 | |
| *** serlex has joined #openstack-containers | 08:17 | |
| *** Bhujay has joined #openstack-containers | 08:34 | |
| *** ykarel|lunch is now known as ykarel | 08:36 | |
| *** janki|lunch is now known as janki | 08:42 | |
| *** jaewook_oh has joined #openstack-containers | 08:50 | |
| *** jaewook_oh has joined #openstack-containers | 08:51 | |
| openstackgerrit | Spyros Trigazis proposed openstack/magnum master: [k8s] Set order in kubemaster software deployments https://review.openstack.org/591592 | 09:09 |
| *** salmankhan has joined #openstack-containers | 09:10 | |
| *** salmankhan1 has joined #openstack-containers | 09:15 | |
| *** salmankhan has quit IRC | 09:16 | |
| *** salmankhan1 is now known as salmankhan | 09:16 | |
| *** openstackstatus has joined #openstack-containers | 09:40 | |
| *** ChanServ sets mode: +v openstackstatus | 09:40 | |
| *** mvpnitesh has joined #openstack-containers | 09:41 | |
| *** Nisha_away has quit IRC | 09:44 | |
| *** ykarel is now known as ykarel|afk | 09:58 | |
| mvpnitesh | strigazi: Hi Strigazi. I'm trying to create a cluster, when the cloud init is executing "atomic install --storage ostree --system --system-package=no --name=kube-apiserver docker.io/openstackmagnum/kubernetes-apiserver:v1.9.3" it is pulling the image and throwing this error " g-io-error-quark: fallocate: No space left on device (12)".yesterday as you said I've tried to add this information in template --docker-storage-driver overlay2 | 10:22 |
| mvpnitesh | and tried with it. No help, it is still failing at the same step | 10:22 |
| *** flwang1 has joined #openstack-containers | 10:25 | |
| flwang1 | strigazi: around? | 10:25 |
| *** adrianc has quit IRC | 10:25 | |
| *** adrianc has joined #openstack-containers | 10:26 | |
| mvpnitesh | flwang1: Hi, i'm creating a cluster with Fedora-26, in cluster creation, "atomic install --storage ostree --system --system-package=no --name=kube-apiserver docker.io/openstackmagnum/kubernetes-apiserver:v1.9.3" it is pulling the image and throwing this error " g-io-error-quark: fallocate: No space left on device (12)" | 10:28 |
| flwang1 | mvpnitesh: which version of magnum you're using? | 10:29 |
| mvpnitesh | master | 10:29 |
| mvpnitesh | flwang1: i've tried to create a cluster with Fedora-27 the default, it is not getting created | 10:30 |
| flwang1 | mvpnitesh: with master, just use fedora 27 | 10:30 |
| flwang1 | are you using devstack? | 10:30 |
| mvpnitesh | flwang1: yes, i'm using devstack | 10:31 |
| flwang1 | then, just use the built in 27 image | 10:31 |
| flwang1 | it should work | 10:31 |
| strigazi | flwang1: hi | 10:42 |
| flwang1 | strigazi: today i had a chat with hogepodge about the functional testing and certificated k8s | 10:43 |
| flwang1 | hogepodge: will help find resource for the functional test and i will work on the certified k8s based on magnum | 10:43 |
| flwang1 | just FYI | 10:43 |
| strigazi | flwang1: thanks, I hope that it will move forward | 10:44 |
| flwang1 | strigazi: let's see ;) | 10:44 |
| flwang1 | anything else you want to discuss | 10:44 |
| strigazi | 591592 | 10:45 |
| strigazi | 572897 this is with the native client? | 10:45 |
| flwang1 | yes | 10:46 |
| strigazi | the one we don't want, right? | 10:47 |
| strigazi | do you get what 591592 does? | 10:47 |
| flwang1 | not sure i understand that | 10:48 |
| flwang1 | we need 572897 to update the health status | 10:49 |
| strigazi | we do? | 10:49 |
| strigazi | why? | 10:49 |
| strigazi | I mean, we want to use requests right? | 10:49 |
| flwang1 | yep, i haven't finish the patch | 10:50 |
| flwang1 | personally, i'd like to reuse the k8s python client as much as possbile, if it doesn't work,i will just use requests | 10:51 |
| strigazi | it doesn't | 10:51 |
| flwang1 | that part should be ok, i need your comments for the overall desing | 10:51 |
| flwang1 | design | 10:51 |
| strigazi | in 572897 ? | 10:52 |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: [k8s] Update cluster health status by native API https://review.openstack.org/572897 | 10:52 |
| flwang1 | yes, still need testing, but as i said above, either using requests or k8s python client, should be OK, we can manage | 10:53 |
| flwang1 | i just upload a new patch set | 10:53 |
| flwang1 | need your comments for the overall design | 10:53 |
| flwang1 | currently, i'm reusing the existing k8s monitor to pull nodes info from k8s api and if there is any node is not ready, then we think the cluster is unhealthy | 10:54 |
| flwang1 | and i'm returning all the conditions of all nodes, so that the auto healing algorithm can make a reasonable decision about how to heal | 10:55 |
| flwang1 | does that make sense for you? | 10:55 |
| strigazi | kind of | 10:57 |
| strigazi | give me 5' to go through again | 10:57 |
| flwang1 | sure | 10:57 |
| flwang1 | we will improve the healthy calculating algorithm later, for now, i think we can make it a little bit strict, any node is not ready, then the cluster is not healthy, means it needs to be repaired | 10:59 |
| strigazi | ^^ make sense | 10:59 |
| *** udesale has quit IRC | 10:59 | |
| strigazi | I think we need one field | 11:00 |
| strigazi | the status of the API | 11:00 |
| flwang1 | it's on my list | 11:00 |
| strigazi | first check if the API is up | 11:00 |
| strigazi | then node status | 11:00 |
| strigazi | makes sense?> | 11:00 |
| strigazi | makes sense? | 11:00 |
| flwang1 | but for api, we can't only reuse the /healthz result | 11:00 |
| flwang1 | because there is no way to get the status for each master node | 11:01 |
| strigazi | you are asking me or telling ? | 11:01 |
| strigazi | you are asking me or telling me ? :) | 11:01 |
| strigazi | well, that is true and makes sense | 11:01 |
| flwang1 | but for api, we cat only reuse the /healthz result | 11:01 |
| flwang1 | can | 11:01 |
| flwang1 | sorry for fat finger | 11:01 |
| flwang1 | in the future, we may need better way to monitor the master status | 11:02 |
| strigazi | ok is enough | 11:02 |
| flwang1 | because for a 3 master nodes cluster, 1 master dead, the cluster is still O | 11:02 |
| flwang1 | OK | 11:02 |
| flwang1 | but if there is another master node dead, the cluster is dead | 11:03 |
| strigazi | if master is down etcd will not return ok | 11:03 |
| strigazi | if 1 master is down etcd will not return ok | 11:03 |
| flwang1 | hmm... that's a good point | 11:03 |
| strigazi | plus in case if calico | 11:03 |
| strigazi | where we already have the kubelet in the master | 11:03 |
| strigazi | you will have the status of the node there too | 11:04 |
| flwang1 | i like the point, thanks for lighting me | 11:04 |
| strigazi | we are working with imdigitaljim to add kubelet everywhere | 11:04 |
| flwang1 | i will add the master status/conditions to the health_status_reason dict | 11:04 |
| strigazi | I think: | 11:04 |
| flwang1 | it would make the code more clean, re kubelet everywhere | 11:05 |
| strigazi | {api_status: {}, nodes: {}} | 11:05 |
| flwang1 | i'm OK with that | 11:05 |
| flwang1 | api_status will be the result from /healthz? | 11:06 |
| strigazi | yes | 11:06 |
| flwang1 | deal | 11:06 |
| *** mvpnitesh has quit IRC | 11:15 | |
| strigazi | sorry, distractions, working in a building with 200 people | 11:15 |
| flwang1 | haha | 11:18 |
| flwang1 | generally, there are only 10 people around me | 11:18 |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: [k8s] Update cluster health status by native API https://review.openstack.org/572897 | 11:19 |
| strigazi | flwang1: so, with pythoi-reqs this part will be the same?: https://review.openstack.org/#/c/572897/3/magnum/service/periodic.py@90 | 11:19 |
| flwang1 | yes | 11:19 |
| flwang1 | only this part https://review.openstack.org/#/c/572897/3/magnum/conductor/k8s_api.py could be changed if we have to use 'reqeuests' | 11:20 |
| flwang1 | hence why i'd like to only focus on the overall design instead of the api accessing | 11:21 |
| strigazi | flwang1 + | 11:21 |
| strigazi | flwang1 +1 | 11:21 |
| strigazi | flwang1: maybe s/pull_data/get_health_data/ | 11:22 |
| flwang1 | strigazi: i thought that tool | 11:22 |
| flwang1 | too | 11:22 |
| flwang1 | in Stein, i will totally drop the 'pod' stuff in the monitor | 11:22 |
| flwang1 | as I discussed with you before, it doesn't make any sense | 11:23 |
| flwang1 | i'd like to make the k8s monitor only focus on general status update and health | 11:23 |
| *** serlex has quit IRC | 11:24 | |
| strigazi | so what we do? I think we can do s/pull_data/get_health_data/ | 11:25 |
| strigazi | or add get_health | 11:25 |
| strigazi | or just health | 11:25 |
| flwang1 | the name doesn't matter, i agree to create a new function to get the health data | 11:27 |
| flwang1 | i'm going to offline, anything else you want to discuss? | 11:30 |
| strigazi | about the patch I pushed? | 11:30 |
| strigazi | the order of software deployments | 11:31 |
| flwang1 | strigazi: what happened without the patch? | 11:31 |
| strigazi | since we added the check for the api | 11:32 |
| strigazi | the api must be up the deployments to move | 11:33 |
| flwang1 | ok | 11:33 |
| strigazi | if the ca.key is not in place the api doesn start: deadlock | 11:33 |
| flwang1 | got it | 11:33 |
| strigazi | one last thing | 11:36 |
| strigazi | about moving kubelet, we are very close, I think we can add it back | 11:36 |
| flwang1 | strigazi: in rocky? | 11:36 |
| strigazi | yes | 11:37 |
| flwang1 | hmm... ok | 11:37 |
| strigazi | it is in for calico anyway | 11:37 |
| flwang1 | yes | 11:37 |
| strigazi | we can also move flannel, that doesn't affect at all | 11:37 |
| strigazi | we can also move flannel, that doesn't affect you at all | 11:37 |
| flwang1 | but i assume it will introduce much change if we don't just simply add it back | 11:37 |
| flwang1 | what do yo mean 'move flannel'? | 11:38 |
| strigazi | to hosted on k8s | 11:38 |
| flwang1 | ah | 11:39 |
| flwang1 | i don't really care about flannel TBH, so i don't mind hold it until stein | 11:39 |
| strigazi | https://github.com/coreos/flannel/blob/master/Documentation/kube-flannel.yml | 11:40 |
| flwang1 | we're going to use Rocky, so i'm conservative to introduce big changes ;) | 11:41 |
| flwang1 | and btw, where did we get the prometheus config? | 11:41 |
| strigazi | I guess you will have a tests end for some time right? | 11:41 |
| flwang1 | strigazi: yes, full test and sonobuoy | 11:42 |
| strigazi | we usually need a couple of days to test for prod | 11:42 |
| flwang1 | i mean this one https://review.openstack.org/#/c/508172/ | 11:42 |
| flwang1 | it doesn't work for me | 11:42 |
| strigazi | sonobuoy doesn't test openstack and CERN filesystems, sonobuoy conformance is given | 11:43 |
| flwang1 | since now magnum can support lb without floating ip, so it would be nice if we can drop the node port of prometheus | 11:43 |
| flwang1 | which is the only addon of magnum still using node port | 11:43 |
| strigazi | what doesn't work? | 11:44 |
| *** mvpnitesh has joined #openstack-containers | 11:44 | |
| flwang1 | the data resource doesn't work | 11:45 |
| flwang1 | my env has been destroyed | 11:45 |
| strigazi | what? | 11:45 |
| flwang1 | i will setup it again and give you more details | 11:45 |
| strigazi | by prometheus? | 11:45 |
| flwang1 | yes | 11:45 |
| strigazi | how can this even happen? | 11:46 |
| flwang1 | hence why i asked where did we get the original config | 11:46 |
| flwang1 | i'd like to do some comparision | 11:46 |
| flwang1 | pls revisit this corner if you have time recently | 11:47 |
| *** ykarel|afk is now known as ykarel | 11:47 | |
| flwang1 | strigazi: i have to go, sorry | 11:49 |
| flwang1 | it's late here | 11:49 |
| strigazi | flwang1: thanks Feilong, good night | 11:51 |
| *** flwang1 has quit IRC | 11:52 | |
| *** slagle has joined #openstack-containers | 11:53 | |
| *** serlex has joined #openstack-containers | 12:04 | |
| mvpnitesh | strigazi: I'm trying to create a cluster with Desvtack master setup. THe master node is getting create and non of the cloud init scripts are running | 12:10 |
| *** udesale has joined #openstack-containers | 12:33 | |
| *** dave-mccowan has joined #openstack-containers | 12:43 | |
| *** mvpnitesh has quit IRC | 12:45 | |
| *** Bhujay has quit IRC | 13:07 | |
| *** Bhujay has joined #openstack-containers | 13:07 | |
| *** jaewook_oh has quit IRC | 13:11 | |
| *** savvas has joined #openstack-containers | 13:14 | |
| savvas | anyone has any idea on why cluster creation could time out? http://paste.openstack.org/show/728012/ . Master node gets created and initial connectivity to it can be established, but at some point during the tasks it executes I lose connectivity and eventually the stack fails | 13:15 |
| canori01 | savvas: Looks like it's not getting to the part where it notifies heat. You would need to check the logs on the master to see where it is erroring out | 13:27 |
| savvas | ok ye I can no longer access it because it loses connectivity at some point during the install. The password on those atom images is random so I can't access it via console | 13:38 |
| *** Bhujay has quit IRC | 13:40 | |
| *** pbourke has quit IRC | 13:54 | |
| *** pbourke has joined #openstack-containers | 13:56 | |
| *** brtknr has quit IRC | 14:05 | |
| *** ricolin has joined #openstack-containers | 14:15 | |
| *** hongbin has joined #openstack-containers | 14:21 | |
| *** brtknr has joined #openstack-containers | 14:22 | |
| *** brtknr has quit IRC | 14:24 | |
| *** Bhujay has joined #openstack-containers | 14:52 | |
| *** ktibi has quit IRC | 14:57 | |
| *** adrianc has quit IRC | 15:29 | |
| *** adrianc has joined #openstack-containers | 15:29 | |
| *** janki has quit IRC | 15:36 | |
| *** savvas has quit IRC | 15:36 | |
| *** ykarel is now known as ykarel|away | 15:41 | |
| *** itlinux has joined #openstack-containers | 15:54 | |
| *** ykarel|away has quit IRC | 15:59 | |
| *** pcaruana has quit IRC | 16:02 | |
| *** mattgo has quit IRC | 16:09 | |
| openstackgerrit | Jim Bach proposed openstack/magnum master: cleanup config-k8s-masters.sh, added roles to nodes on startup https://review.openstack.org/589214 | 16:47 |
| *** udesale has quit IRC | 16:54 | |
| *** serlex has quit IRC | 16:56 | |
| *** ricolin has quit IRC | 17:05 | |
| *** adrianc has quit IRC | 17:06 | |
| *** mattgo has joined #openstack-containers | 17:09 | |
| *** Bhujay has quit IRC | 17:22 | |
| *** hongbin has quit IRC | 17:35 | |
| *** hongbin has joined #openstack-containers | 17:35 | |
| *** hongbin has quit IRC | 17:35 | |
| *** hongbin has joined #openstack-containers | 17:36 | |
| *** janki has joined #openstack-containers | 17:45 | |
| *** salmankhan has quit IRC | 18:07 | |
| *** salmankhan has joined #openstack-containers | 20:20 | |
| imdigitaljim | meeting today? | 20:54 |
| strigazi | imdigitaljim: yes | 20:55 |
| imdigitaljim | great | 20:55 |
| strigazi | flwang: ping | 20:58 |
| strigazi | #startmeeting containers | 20:59 |
| openstack | Meeting started Tue Aug 14 20:59:04 2018 UTC and is due to finish in 60 minutes. The chair is strigazi. Information about MeetBot at http://wiki.debian.org/MeetBot. | 20:59 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 20:59 |
| openstack | The meeting name has been set to 'containers' | 20:59 |
| strigazi | #topic Roll Call | 20:59 |
| imdigitaljim | o/ | 20:59 |
| strigazi | o/ | 20:59 |
| colin- | \o | 20:59 |
| strigazi | I guess flwang prefers working at midnight not midday :) | 21:00 |
| strigazi | #topic Announcements | 21:00 |
| imdigitaljim | haha | 21:00 |
| strigazi | we have stable/rocky and magnum 7.0.0 | 21:01 |
| imdigitaljim | \o/ | 21:01 |
| strigazi | we have still patches to add so 7.0.1 would be the release | 21:01 |
| strigazi | \o/ indeed :) | 21:02 |
| strigazi | #topic Blueprints/Bugs/Ideas | 21:02 |
| strigazi | I promised canori01 to tests coreos and I did | 21:02 |
| strigazi | and apparently it is me that kind of broke it | 21:03 |
| strigazi | I mean the coreos driver | 21:03 |
| strigazi | details here: | 21:03 |
| strigazi | #link https://review.openstack.org/#/c/579026 | 21:03 |
| imdigitaljim | didnt you go to the fedora convention? | 21:03 |
| imdigitaljim | what are the atomic/coreos plans that you migth know of | 21:04 |
| strigazi | The issue is passing certs with heat and using string replacement | 21:04 |
| imdigitaljim | #link https://review.openstack.org/#/c/590443/ | 21:04 |
| imdigitaljim | #link https://review.openstack.org/#/c/590346/ | 21:04 |
| imdigitaljim | #link https://review.openstack.org/#/c/589214/ | 21:04 |
| strigazi | cloud-config in coreos != cloud-init | 21:05 |
| imdigitaljim | #link https://review.openstack.org/#/c/577570/ | 21:05 |
| imdigitaljim | are all ready for review | 21:05 |
| strigazi | thanks | 21:05 |
| colin- | this reminds me, i was wondering strigazi or others if anyone has used hashicorp's Vault as a sidecar certificate authority in the cluster? | 21:05 |
| colin- | to ease some of these certificate management/distribution tasks? | 21:06 |
| strigazi | colin-: no, at least at CERN we have barbican | 21:06 |
| strigazi | there is work to integrate barbican and k8s | 21:07 |
| colin- | ok | 21:07 |
| strigazi | eg barbican sdk was added to gophercloud | 21:07 |
| *** canori02 has joined #openstack-containers | 21:07 | |
| colin- | that's great, the interest is less in Vault in more in a more graceful way to manage all the TLS files | 21:07 |
| imdigitaljim | were working on org approval to work on kubernetes org repo to contribute several PRs for the standalone openstack controller | 21:07 |
| imdigitaljim | which will also translate to some magnum PRs | 21:08 |
| strigazi | cool :) | 21:08 |
| canori02 | o/ | 21:08 |
| strigazi | canori02: o/ | 21:09 |
| *** mattgo has quit IRC | 21:09 | |
| strigazi | imdigitaljim: have you pushed a patch to add kube-proxy already? | 21:09 |
| imdigitaljim | id need the two files merged to apply the rest | 21:10 |
| imdigitaljim | make-cert and configure-master | 21:10 |
| imdigitaljim | because it uses elements of those to complete it | 21:10 |
| strigazi | imdigitaljim: ok | 21:10 |
| imdigitaljim | however the work has been already done | 21:11 |
| imdigitaljim | just a matter of sequence | 21:11 |
| strigazi | you can push with depencies in gerrit, I guess you know this | 21:11 |
| strigazi | *dependencies | 21:11 |
| *** canori02 has quit IRC | 21:12 | |
| imdigitaljim | i did not actually know, im not super knowledgeable with gerrit | 21:12 |
| imdigitaljim | ill check it out | 21:12 |
| imdigitaljim | and push it | 21:12 |
| strigazi | https://docs.openstack.org/infra/manual/developers.html#adding-a-dependency | 21:12 |
| imdigitaljim | oh awesome | 21:12 |
| imdigitaljim | thanks | 21:12 |
| imdigitaljim | that will make it easy | 21:13 |
| strigazi | three more items from me: | 21:13 |
| *** canori02 has joined #openstack-containers | 21:13 | |
| strigazi | 1. i'm working on the unit tests for the upgrade API, of course I love writing unit tests | 21:13 |
| strigazi | 2. with flwang we made some progress in : | 21:14 |
| strigazi | #link https://review.openstack.org/#/c/572897/ | 21:14 |
| strigazi | pulling the k8s cluster for health status | 21:14 |
| strigazi | and return something like {api_health: {}, nodes: []} | 21:15 |
| strigazi | 3. k8s 1.11.x works without any issue, all patches for the certificates are merged and the contianer images are updated | 21:16 |
| imdigitaljim | yep! | 21:16 |
| imdigitaljim | we're also on 1.11.2 | 21:16 |
| strigazi | really easy now :) | 21:17 |
| strigazi | the patch is 6 character, it would be 1 if I used a variable :) | 21:17 |
| imdigitaljim | we recently updated calico to 3.1.3 | 21:18 |
| imdigitaljim | we're like 17 releases behind in magnum | 21:18 |
| strigazi | args in dockerfiles require a newer docker version | 21:18 |
| imdigitaljim | probably more relevant to flwang: | 21:18 |
| strigazi | imdigitaljim: what we have in magnum atm | 21:18 |
| strigazi | ? | 21:18 |
| imdigitaljim | 2.6.7 | 21:18 |
| strigazi | I wonder why flwang pushed for 2.6.x | 21:19 |
| imdigitaljim | there were some 3.x issues at the time | 21:19 |
| strigazi | I think 3.x.y was out at that time | 21:19 |
| imdigitaljim | but to us they appear to be resolved | 21:19 |
| strigazi | oh, ok | 21:19 |
| strigazi | makes sense | 21:19 |
| imdigitaljim | yeah he was probably doing what was necessary | 21:19 |
| strigazi | Finally, a comment for Fedora CoreOS | 21:19 |
| strigazi | we need at least 6 months fedora having something solid to use | 21:20 |
| strigazi | All the builds they have now are experimental and none of them are public | 21:20 |
| strigazi | Fedora Atomic 30 will be the last one | 21:20 |
| strigazi | Fedora CoreOS will be based on rpms | 21:21 |
| strigazi | and will rpm-ostree | 21:21 |
| strigazi | and will use rpm-ostree | 21:21 |
| imdigitaljim | will we still be able to use the same files that you know of? "atomic install etc" | 21:21 |
| strigazi | atomic cli no, but something similar | 21:21 |
| imdigitaljim | ok so we'll need to make some changes | 21:22 |
| imdigitaljim | maybe we can get a good way to prebuild the new images as well quickly | 21:22 |
| strigazi | they said since there are users like us they will take it into account | 21:22 |
| imdigitaljim | so we can inject extra stuff on the base fedora coreos image | 21:22 |
| strigazi | regaring that, we will need to work with ignition | 21:23 |
| *** slagle has quit IRC | 21:23 | |
| strigazi | we can start investigating this with coreos and be ready | 21:23 |
| strigazi | we need a way to compile the ignition json | 21:23 |
| *** salmankhan has quit IRC | 21:23 | |
| strigazi | we can join the fedora coreos meeting on the 21st | 21:24 |
| *** salmankhan has joined #openstack-containers | 21:24 | |
| imdigitaljim | https://coreos.com/ignition/docs/latest/ | 21:24 |
| imdigitaljim | got it | 21:24 |
| strigazi | imdigitaljim: are you interested? | 21:24 |
| imdigitaljim | that would be good actually | 21:24 |
| strigazi | #link https://apps.fedoraproject.org/calendar/workstation/2018/8/20/#m9315 | 21:25 |
| strigazi | 7:30 for me, great... | 21:25 |
| strigazi | imdigitaljim: for you? | 21:25 |
| strigazi | I can ask for the other one, if it is better, I think they will alternate | 21:26 |
| imdigitaljim | 1030PM | 21:26 |
| imdigitaljim | i should be able to make it if i remember :p | 21:27 |
| strigazi | imdigitaljim: I can ping you :) | 21:27 |
| strigazi | speaking of atomic, I tested F28AH, works fine with magnum 7.0.0 | 21:27 |
| strigazi | and queens actually, I'll push a patch | 21:28 |
| strigazi | I think that was all from me, imdigitaljim colin- canori02 do you want to add anything? | 21:29 |
| canori02 | How far away is fedora coreos? | 21:30 |
| strigazi | at least 6 months | 21:30 |
| canori02 | I ask since the work I've been doing makes the coreos driver work through cloud-init and not ignition | 21:31 |
| canori02 | Should we just deprecate that? | 21:31 |
| strigazi | I would help | 21:31 |
| strigazi | It would help | 21:32 |
| strigazi | I think we can add your patch in rocky and then work on ignition | 21:32 |
| strigazi | canori02: makes sense? | 21:32 |
| canori02 | Yeah, makes sense | 21:33 |
| imdigitaljim | nothing else here | 21:33 |
| canori02 | Also, had another question. Sorry if you covered this already since I was late | 21:33 |
| canori02 | What were your guy's thougts on the discovery.etcd.io issues they were having? | 21:34 |
| strigazi | discovery.etcd.io is back, but they plan to deprecate it, but without a clear timeline | 21:34 |
| imdigitaljim | canori02: if you want to put the effort with it, this can be run internally/locally | 21:35 |
| strigazi | magnum has the option to use a local discovery so we can use that | 21:35 |
| flwang | strigazi: re calico | 21:35 |
| strigazi | I posted a link in the channel last week on how to do it | 21:36 |
| flwang | when I worked on calico, GKE is also using 2.6.7 | 21:36 |
| canori02 | Yeah, I did deploy one internally. Wasn't too bad | 21:36 |
| flwang | so I trust google so I assume 2.6.7 is stable at least | 21:36 |
| strigazi | sounds good ^^ | 21:36 |
| flwang | and given we have calico node tag, so user can easily upgrade if they want | 21:37 |
| strigazi | flwang: we could change the default? | 21:37 |
| flwang | new version is cool, but i'm always a old man style, so.... | 21:37 |
| flwang | strigazi: we can | 21:37 |
| flwang | i can test with 3.x | 21:37 |
| flwang | and propose the version upgrade if it can pass the sonobuoy testing | 21:38 |
| colin- | do you remember what they were version locking on 2.6.x to support flwang? i seem to recall it being related to IPVS use for kube-proxy but could be wrong | 21:39 |
| strigazi | canori02: we can start an issue in etcd repo to ask what they recommend | 21:39 |
| strigazi | I imagine they will say, run your own discovery | 21:39 |
| strigazi | bootstraping etcd without knowing the ips beforehand is tedious | 21:40 |
| canori02 | I saw they were asking for feedback on how we use the service. So we can give them that | 21:42 |
| flwang | colin-: i can't remember, sorry | 21:44 |
| strigazi | yeap, you can reply, I can follow it up too | 21:44 |
| flwang | for etcd discovery issue, at least, we can add a retry for the function | 21:44 |
| imdigitaljim | also the calico_tag wont work entirely | 21:45 |
| imdigitaljim | the yaml format changed | 21:45 |
| imdigitaljim | specifically plugins | 21:45 |
| imdigitaljim | minor changes though | 21:45 |
| imdigitaljim | flwang:^ | 21:45 |
| imdigitaljim | flwang: we've also gotten sonobuoy set up as well | 21:46 |
| strigazi | imdigitaljim: flwang it always takes one hour to run? | 21:48 |
| imdigitaljim | yeah | 21:48 |
| imdigitaljim | its like 67 minutes for us | 21:48 |
| imdigitaljim | although sonobuoy make some more assumptions on a fwe things but overall its pretty good | 21:49 |
| strigazi | what assumptions? example? | 21:49 |
| imdigitaljim | 1 sec | 21:51 |
| *** itlinux has quit IRC | 21:54 | |
| flwang | sorry, i was in a meeting and going into another one | 21:55 |
| flwang | strigazi: yes, 1 hour | 21:55 |
| strigazi | flwang: enjoy :) | 21:55 |
| flwang | but i'm going to dig to see if we can have a smoke test set | 21:55 |
| imdigitaljim | i was gonna try to link the codeline but i cant find it right this second | 21:56 |
| strigazi | imdigitaljim: are you still there? | 21:56 |
| flwang | imdigitaljim: yep, that's a good point, i will check if we should upgrade to 3.x | 21:56 |
| imdigitaljim | but basically its assuming a master node is labeled in a specific way | 21:57 |
| imdigitaljim | and its not even a good way | 21:57 |
| imdigitaljim | i think sonobuoy pulls from the kk e2e anyways so it might actually just be a bad kk test | 21:57 |
| imdigitaljim | so it skips like 100s of tests | 21:57 |
| imdigitaljim | based solely on that | 21:57 |
| strigazi | imdigitaljim: do you have the name of the test? I didn't see anything in the logs about the label | 21:58 |
| imdigitaljim | yeah ill dig it up again | 21:58 |
| strigazi | maybe I missed ti | 21:58 |
| flwang | imdigitaljim: i'm interested in too | 21:58 |
| strigazi | thanks, when I run again I'll look closer | 21:59 |
| imdigitaljim | i could have totally missed something too so it would be good to find out | 21:59 |
| strigazi | let's end the meeting then | 22:00 |
| strigazi | see you next week everyone | 22:00 |
| strigazi | #endmeeting | 22:00 |
| openstack | Meeting ended Tue Aug 14 22:00:47 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 22:00 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/containers/2018/containers.2018-08-14-20.59.html | 22:00 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/containers/2018/containers.2018-08-14-20.59.txt | 22:00 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/containers/2018/containers.2018-08-14-20.59.log.html | 22:00 |
| *** janki has quit IRC | 22:01 | |
| imdigitaljim | https://github.com/kubernetes/kubernetes/blob/master/test/e2e/framework/metrics/metrics_grabber.go#L79 | 22:01 |
| imdigitaljim | strigazi:flwang: here this was part of it | 22:01 |
| imdigitaljim | https://github.com/kubernetes/kubernetes/blob/master/pkg/util/system/system_utils.go#L29 | 22:02 |
| imdigitaljim | thats the assumption | 22:02 |
| imdigitaljim | hwev5mpebvv4-master-0 is the end of the master node name | 22:04 |
| imdigitaljim | not ending in master | 22:04 |
| *** salmankhan has quit IRC | 22:04 | |
| imdigitaljim | u remember what they were version locking on 2.6. | 22:04 |
| imdigitaljim | https://golang.org/pkg/strings/#HasSuffix | 22:04 |
| imdigitaljim | ignore copy/paste fial | 22:04 |
| imdigitaljim | fail* | 22:04 |
| *** canori02 has quit IRC | 22:28 | |
| flwang | imdigitaljim: thanks, that's helpful | 22:51 |
| *** hongbin has quit IRC | 22:53 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!