| *** rcernin has joined #openstack-containers | 00:31 | |
| *** eyalb has joined #openstack-containers | 05:07 | |
| *** eyalb has quit IRC | 05:25 | |
| *** ykarel|away has joined #openstack-containers | 05:41 | |
| *** eyalb has joined #openstack-containers | 06:07 | |
| *** ykarel|away has quit IRC | 06:09 | |
| *** rcernin has quit IRC | 08:17 | |
| *** ykarel|away has joined #openstack-containers | 08:22 | |
| *** ykarel|away has quit IRC | 08:54 | |
| *** ykarel|away has joined #openstack-containers | 08:55 | |
| *** ykarel|away has quit IRC | 09:10 | |
| *** slagle has quit IRC | 12:44 | |
| *** slagle has joined #openstack-containers | 13:52 | |
| *** eyalb has quit IRC | 14:42 | |
| *** ykarel has joined #openstack-containers | 15:05 | |
| openstackgerrit | Spyros Trigazis proposed openstack/magnum master: Use existing templates for cluster-update command https://review.openstack.org/600806 | 16:20 |
|---|---|---|
| *** ykarel has quit IRC | 17:33 | |
| *** pcaruana has joined #openstack-containers | 18:48 | |
| *** pcaruana has quit IRC | 19:07 | |
| *** hongbin has joined #openstack-containers | 20:40 | |
| *** hongbin has quit IRC | 21:53 | |
| flwang | strigazi: how can i help? | 21:58 |
| strigazi | flwang: hello | 22:37 |
| strigazi | flwang: you there? | 22:38 |
| flwang | strigazi: yes | 22:38 |
| strigazi | flwang: Can you have a look into Use existing templates for cluster-update command https://review.openstack.org/600806 ? | 22:38 |
| flwang | strigazi: what do you mean 'existing templates'? | 22:39 |
| strigazi | flwang: Also the patch to use the kubernetes client for the health status is not python 3.6 compatible | 22:39 |
| strigazi | flwang: stack update --existing | 22:39 |
| flwang | strigazi: ok, will do. we need it for rolling update feature, right? | 22:39 |
| strigazi | flwang: not only, the biggest issue is that if you have running clusters. Say a k8s cluster built with magnum version x.y.z, then you update magnum to x.y.(z+1) where the z+1 has a change in the heat templates that go into vm user-data, and if the user tries to scale the cluster when magnum is z+1, the stack update is desstructive, it replaces vms. | 22:42 |
| flwang | strigazi: ah, i see. will take a look then. | 22:43 |
| flwang | strigazi: as for the health status patch, if it's a really compatible issue, then we may have to drop the k8s client totally | 22:43 |
| flwang | though i don't want to | 22:44 |
| *** rcernin has joined #openstack-containers | 22:44 | |
| strigazi | flwang: for the patch to sync the cluster health with the k8s client, I don't know if there is a solution. We can investigate | 22:44 |
| strigazi | I think when use a sigle process it is fine. | 22:45 |
| strigazi | flwang: in my multicore laptop breaks, in a single vm it works | 22:45 |
| strigazi | flwang: in my multicore laptop breaks, in a single-core vm it works | 22:45 |
| flwang | strigazi: interesting, in my laptop it just works | 22:54 |
| openstackgerrit | Feilong Wang proposed openstack/magnum master: Keep the cover job non-voting https://review.openstack.org/603001 | 23:12 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!