Friday, 2020-02-21

*** jmlowe has joined #openstack-containers01:01
*** jmlowe has quit IRC01:30
*** vishalmanchanda has joined #openstack-containers01:55
brtknrflwang: on and off with the babies02:01
brtknrhows it going?02:02
flwanggood, i'm sorry the patch doesn't work for you, i will test it again today02:05
openstackgerritMerged openstack/magnum master: [hca] Live log for SoftwareDeployment scripts  https://review.opendev.org/70769402:05
openstackgerritMerged openstack/magnum stable/train: Add selector in monitoring deployments  https://review.opendev.org/70546502:05
openstackgerritMerged openstack/magnum stable/train: Execute traefik systemd unit over ssh  https://review.opendev.org/70546402:05
openstackgerritMerged openstack/magnum master: core-podman: Mount os-release properly  https://review.opendev.org/70474902:05
flwangbrtknr: ^02:09
brtknrflwang: yes I do not understand why because previously it accepted 0 as argument but simply didn’t create a volume but when AZ is given, it needs the size to be non zero02:09
flwangbrtknr: do you think the idea we discussed on Wed sounds good to you?02:09
flwangbrtknr: maybe i missed push the latest changes, i will check02:10
brtknrflwang: Different volume AZ label?02:15
brtknrflwang: axtualy with your ps, even without AZ label, cluster creation fails if etcd volume size is 0 or unspecified02:17
flwangok, shame on me, i may missed some cases02:22
flwangyes, another label for volume AZ and a new config02:23
brtknrMagnum config for volume AZ?02:27
brtknrflwang: what if not specified?02:28
flwangif both label and config are not specified, then cluster will fail to create02:28
*** ykarel|away is now known as ykarel03:51
brtknrBut only if volume size is non zero rigjt05:38
brtknrflwang:05:38
*** sapd1 has joined #openstack-containers05:52
*** sapd1_ has quit IRC05:52
*** elenalindq has joined #openstack-containers06:08
*** ricolin has quit IRC06:15
*** ricolin has joined #openstack-containers07:21
*** belmoreira has quit IRC07:29
*** lpetrut has joined #openstack-containers07:39
*** ivve has quit IRC07:39
*** ramishra has quit IRC07:49
*** ramishra has joined #openstack-containers08:03
*** ivve has joined #openstack-containers08:44
*** ykarel is now known as ykarel|lunch08:47
*** ykarel|lunch is now known as ykarel08:55
cosmicsoundbrtknr , using latest stable train09:07
*** rcernin has quit IRC09:11
*** sapd1 has quit IRC09:11
cosmicsoundGood day09:13
cosmicsoundbrtknr , here it is http://paste.openstack.org/show/789838/09:13
brtknrcosmicsound: has docker_storage_driver: devicemapper worked for you before? i always use overlay209:37
brtknrcosmicsound: can you be explicit when you say latest stable/train? so you are not using a release?09:38
brtknrcosmicsound: do the servers get created?09:38
brtknrcan you log into the master node and look at the logs?09:38
*** rcernin has joined #openstack-containers09:48
openstackgerritBharat Kunwar proposed openstack/magnum master: [k8s] Make metrics-server work without DNS  https://review.opendev.org/70598409:53
openstackgerritBharat Kunwar proposed openstack/magnum master: [k8s] Make metrics-server work without DNS  https://review.opendev.org/70598409:55
ionidefault  kubernetes-dashboard is deployed but whenever i try to access the dashboard after running kubectl proxy, the browser returnes that service doesn't exists10:07
ioniis there any other way to acces it?10:07
ioniok, got it, the documentation from kubernetes uses other namespace10:10
*** pcaruana has joined #openstack-containers10:19
cosmicsoundbrtknr , we use Train with Kolla10:27
cosmicsoundWe use latest stable relese from December10:27
cosmicsoundI have tried both devicemapper and overlay10:27
cosmicsoundNothing helped10:27
cosmicsoundIl make a new one now with overlay10:27
cosmicsoundbrtknr , http://paste.openstack.org/show/789844/ this uses overlay10:31
cosmicsoundI started a new cluster now, will paste results when done10:31
cosmicsoundusually how long it takes to create a cluster 1 m 2 nodes 40 gb overlay10:35
cosmicsoundMost confusing part is that the k9s master machine is up and running  https://mdb.uhlhost.net/uploads/dcaee9bd86322597/image.png10:36
*** pcaruana has quit IRC10:41
brtknrcosmicsound: so this has never worked since you upgrated to train?10:50
brtknrcosmicsound: can you ssh into the master and look at the logs10:51
brtknrjournalctl -u heat-container-agent10:51
ionii'm using latest from branch train and works fine11:11
ionibut i noticed is that if i use other kube_tag like 1.15.10 i have to use etcd_tag 3.2.26, otherwise it will fail to deploy11:12
ionisince etcd 3.2.something is not in google repo11:12
cosmicsoundhttp://paste.openstack.org/show/789846/ latest log same as with devicemapper11:13
cosmicsoundLet me check those logs11:13
ioni'etcd_volume_size': '20', 'kube_tag': 'v1.15.10', 'etcd_tag': '3.2.26', 'heat_container_agent_tag': 'train-stable', 'boot_volume_type': 'ssd', 'use_podman': 'true', 'etcd_volume_type': 'ssd', 'boot_volume_size': '20', 'flannel_backend': 'vxlan'11:14
cosmicsoundroot@sc:~# journalctl -u heat-container-agent11:14
cosmicsound-- Logs begin at Fri 2019-12-20 11:59:48 CET, end at Fri 2020-02-21 12:14:10 CET. --11:14
cosmicsound-- No entries --11:14
cosmicsoundroot@sc:~#11:14
ionithis are by labels11:14
ionimy11:14
cosmicsoundexcuse my paste11:14
cosmicsoundFunny ha that no logs...11:14
ioniyou can drop the volume labels if you don't boot it from volume11:15
*** ykarel is now known as ykarel|afk11:16
openstackgerritMerged openstack/magnum master: [k8s] Fix instance ID issue with podman and autoscaler  https://review.opendev.org/70733611:39
*** ykarel|afk is now known as ykarel11:42
*** vishalmanchanda has quit IRC12:03
*** pcaruana has joined #openstack-containers12:03
*** ricolin has quit IRC12:14
openstackgerritBharat Kunwar proposed openstack/magnum master: Add cinder_csi_enabled label  https://review.opendev.org/70035712:17
openstackgerritBharat Kunwar proposed openstack/magnum master: [bug] Fix regression when use_podman=false  https://review.opendev.org/70912312:17
*** rcernin has quit IRC12:28
*** pcaruana has quit IRC12:29
*** iokiwi has quit IRC12:43
*** iokiwi has joined #openstack-containers12:43
*** pcaruana has joined #openstack-containers12:48
*** iokiwi has quit IRC12:48
*** iokiwi has joined #openstack-containers12:48
*** ykarel is now known as ykarel|afk13:22
*** mgariepy has joined #openstack-containers13:30
ionidoes anyone uses nfs provisioner?13:34
*** pcaruana has quit IRC13:34
*** pcaruana has joined #openstack-containers13:36
brtknrioni: have used it once14:08
ionibrtknr, i think i got it, i had to deploy using mountoptions 4.114:10
ioniit was failing to create containers due to mount version 3 and statd14:10
*** jmlowe has joined #openstack-containers14:16
*** elenalindq has left #openstack-containers14:49
*** elenalindq has joined #openstack-containers14:50
*** ivve has quit IRC15:12
*** ykarel|afk is now known as ykarel15:34
*** jmlowe has quit IRC15:34
*** lpetrut has quit IRC16:09
*** elenalindq has quit IRC16:12
*** elenalindq has joined #openstack-containers16:25
*** pcaruana has quit IRC17:23
*** jmlowe has joined #openstack-containers17:59
*** jmlowe has quit IRC18:12
*** ykarel is now known as ykarel|away18:24
*** openstackstatus has quit IRC18:33
*** jmlowe has joined #openstack-containers18:35
openstackgerritMerged openstack/magnum master: Fix ingress traefik systemd unit  https://review.opendev.org/70544318:50
*** ianychoi_ has joined #openstack-containers19:15
*** ianychoi has quit IRC19:18
*** jmlowe has quit IRC20:25
*** jmlowe has joined #openstack-containers20:28
*** jmlowe has quit IRC20:35
*** jmlowe has joined #openstack-containers20:42
*** jmlowe has quit IRC20:47
*** jmlowe has joined #openstack-containers21:08
*** dave-mccowan has quit IRC23:11
*** elenalindq has quit IRC23:13
*** dave-mccowan has joined #openstack-containers23:30
*** jmlowe has quit IRC23:31
*** jmlowe has joined #openstack-containers23:31
*** jmlowe has quit IRC23:53

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