Friday, 2018-06-08

*** mdnadeem has joined #openstack-containers00:20
*** yamamoto has joined #openstack-containers00:34
*** yamamoto has quit IRC00:39
*** hongbin has joined #openstack-containers00:43
*** fragatina has quit IRC00:44
*** gsimondon has joined #openstack-containers01:01
*** gsimondon has quit IRC01:05
*** dardelean_ has joined #openstack-containers01:05
*** dardelean_ has quit IRC01:10
*** Nel1x has joined #openstack-containers01:22
*** georgem11 has quit IRC01:39
*** zhubingb_ has joined #openstack-containers01:49
*** zhubingb_ has quit IRC01:54
*** gsimondon has joined #openstack-containers01:55
*** gsimondon has quit IRC01:59
*** pengdake has joined #openstack-containers02:19
*** hongbin has quit IRC02:26
*** gsimondon has joined #openstack-containers02:49
*** gsimondon has quit IRC02:54
*** Nel1x has quit IRC03:02
*** fragatina has joined #openstack-containers03:04
*** fragatina has quit IRC03:05
*** fragatin_ has joined #openstack-containers03:05
*** ramishra has joined #openstack-containers03:06
*** fragatin_ has quit IRC03:17
*** fragatina has joined #openstack-containers03:17
*** yamamoto has joined #openstack-containers03:26
*** yamamoto has quit IRC03:26
*** yamamoto has joined #openstack-containers03:31
*** Roybent has joined #openstack-containers03:36
*** gsimondon has joined #openstack-containers03:43
*** dave-mcc_ has quit IRC03:44
*** lpetrut has joined #openstack-containers03:45
*** gsimondon has quit IRC03:48
*** udesale has joined #openstack-containers03:48
*** ricolin has joined #openstack-containers03:52
*** pengdake has quit IRC04:22
*** lpetrut has quit IRC04:23
*** chhagarw has joined #openstack-containers04:25
*** chhagarw has quit IRC04:32
*** gsimondon has joined #openstack-containers04:37
*** gsimondon has quit IRC04:42
*** ianychoi has quit IRC04:44
*** zhubingb_ has joined #openstack-containers04:51
*** flwang1 has quit IRC04:54
*** zhubingb_ has quit IRC04:56
*** gsimondon has joined #openstack-containers05:31
*** gsimondon has quit IRC05:35
*** AlexeyAbashkin has joined #openstack-containers05:48
*** pengdake has joined #openstack-containers05:51
*** Alexey_Abashkin has joined #openstack-containers05:54
*** AlexeyAbashkin has quit IRC05:55
*** Alexey_Abashkin is now known as AlexeyAbashkin05:55
*** serlex has joined #openstack-containers05:56
*** iranzo has joined #openstack-containers05:57
*** pcaruana has joined #openstack-containers05:57
-openstackstatus- NOTICE: Zuul stopped receiving gerrit events around 04:00UTC; any changes submitted between then and now will probably require a "recheck" comment to be requeued. Thanks!06:00
*** armaan has joined #openstack-containers06:04
*** pcaruana has quit IRC06:14
*** pcaruana has joined #openstack-containers06:15
*** armaan has quit IRC06:20
*** armaan has joined #openstack-containers06:20
*** lpetrut has joined #openstack-containers06:22
*** gsimondon has joined #openstack-containers06:30
*** gsimondo1 has joined #openstack-containers06:36
*** gsimondon has quit IRC06:38
*** armaan has quit IRC06:44
*** armaan has joined #openstack-containers06:45
*** zhubingb_ has joined #openstack-containers06:53
*** zhubingb_ has quit IRC06:57
*** b_bezak has joined #openstack-containers07:02
*** rcernin has quit IRC07:13
*** ramishra has quit IRC07:14
*** dardelean_ has joined #openstack-containers07:15
*** Alexey_Abashkin has joined #openstack-containers07:23
*** AlexeyAbashkin has quit IRC07:26
*** Alexey_Abashkin is now known as AlexeyAbashkin07:26
*** dardelean_ has quit IRC07:32
*** hongbin has joined #openstack-containers07:32
*** dardelean_ has joined #openstack-containers07:32
*** hongbin has quit IRC07:35
*** dardelean_ has quit IRC07:36
*** ramishra has joined #openstack-containers07:44
*** rcernin has joined #openstack-containers07:48
*** ktibi has joined #openstack-containers07:57
*** AlexeyAbashkin has quit IRC08:06
*** yamamoto has quit IRC08:08
*** ramishra has quit IRC08:11
*** ramishra has joined #openstack-containers08:13
*** mgoddard has joined #openstack-containers08:15
*** ricolin has quit IRC08:19
*** yamamoto has joined #openstack-containers08:22
*** AlexeyAbashkin has joined #openstack-containers08:28
*** chhagarw has joined #openstack-containers08:30
*** pengdake has quit IRC08:46
*** dardelean_ has joined #openstack-containers08:48
*** pengdake has joined #openstack-containers08:54
*** Alexey_Abashkin has joined #openstack-containers09:00
*** AlexeyAbashkin has quit IRC09:01
*** Alexey_Abashkin is now known as AlexeyAbashkin09:01
*** flwang1 has joined #openstack-containers09:16
*** ramishra has quit IRC09:17
*** ramishra has joined #openstack-containers09:19
*** openstackgerrit has joined #openstack-containers09:21
openstackgerritFeilong Wang proposed openstack/magnum master: Fix race condition issue for k8s  multi masters  https://review.openstack.org/57363909:21
flwang1strigazi: ping09:22
strigazithe idea looks ok, we need to modify the scripts to not run on the secondary masters09:23
openstackgerritFeilong Wang proposed openstack/magnum master: Fix race condition issue for k8s  multi masters  https://review.openstack.org/57363909:24
flwang1which scripts?09:24
flwang1strigazi:  see https://github.com/kubernetes/kubernetes/issues/4416509:25
flwang1it's a known issue in kubectl09:25
flwang1and it's still open unfortunately09:25
openstackgerritBharat Kunwar proposed openstack/magnum master: Add option to specify Cgroup driver for Kubelet  https://review.openstack.org/57158309:25
strigaziall scripts that try to create something on the api server09:26
strigaziif is_not_primary_master : exit 009:26
flwang1how the script know it's not on primary_master?09:28
strigazihttps://github.com/openstack/magnum/blob/master/magnum/drivers/swarm_fedora_atomic_v2/templates/fragments/write-swarm-master-service.sh#L1309:36
*** Roybent has quit IRC09:40
flwang1strigazi: cool, thanks09:51
*** mdnadeem has quit IRC09:55
*** zhubingb_ has joined #openstack-containers09:55
*** zhubingb_ has quit IRC10:00
*** pengdake has quit IRC10:05
*** Roybent has joined #openstack-containers10:06
*** Roybent has quit IRC10:14
*** ricolin has joined #openstack-containers10:16
*** udesale_ has joined #openstack-containers10:27
*** udesale has quit IRC10:29
*** udesale_ has quit IRC10:31
*** rcernin has quit IRC10:37
brtknrflwang1: how does creating a secondary master resolve race condition issue? is this becuase you've defined dependence on primary master for kube minion?10:40
strigazibrtknr: yes, the secondary masters and the minions depend on the primary master11:15
*** pengdake has joined #openstack-containers11:20
*** pengdake has quit IRC11:26
*** AlexeyAbashkin has quit IRC11:53
*** AlexeyAbashkin has joined #openstack-containers11:54
*** vabada has joined #openstack-containers12:05
*** yamamoto has quit IRC12:11
*** AlexeyAbashkin has quit IRC12:12
*** AlexeyAbashkin has joined #openstack-containers12:12
*** yamamoto has joined #openstack-containers12:15
*** yamamoto has quit IRC12:16
*** armaan has quit IRC12:25
*** yamamoto has joined #openstack-containers12:27
*** yamamoto has quit IRC12:28
*** AlexeyAbashkin has quit IRC12:30
*** AlexeyAbashkin has joined #openstack-containers12:33
*** AlexeyAbashkin has quit IRC12:41
*** armaan has joined #openstack-containers12:41
*** armaan has quit IRC12:41
*** AlexeyAbashkin has joined #openstack-containers12:41
*** armaan has joined #openstack-containers12:41
*** georgem1 has joined #openstack-containers12:45
*** armaan has quit IRC12:46
*** georgem1 has left #openstack-containers12:56
*** AlexeyAbashkin has quit IRC12:56
*** AlexeyAbashkin has joined #openstack-containers12:57
*** zhubingb_ has joined #openstack-containers12:58
*** zhubingb_ has quit IRC13:02
*** vijaykc4 has joined #openstack-containers13:04
*** Nel1x has joined #openstack-containers13:09
*** Nel1x has quit IRC13:16
*** vabada has quit IRC13:16
*** yamamoto has joined #openstack-containers13:29
*** yamamoto has quit IRC13:31
*** yamamoto has joined #openstack-containers13:32
*** jmlowe has quit IRC13:35
*** ramishra has quit IRC13:38
-openstackstatus- NOTICE: A misapplied distro security package update caused many jobs to fail with a MERGER_FAILURE error between ~06:30-12:30 UTC; these can be safely rechecked now that the problem has been addressed13:46
*** markguz_ has joined #openstack-containers13:57
*** lpetrut has quit IRC13:58
*** ianychoi has joined #openstack-containers14:04
*** hongbin has joined #openstack-containers14:05
*** jmlowe has joined #openstack-containers14:07
*** markguz_ has quit IRC14:11
*** gsimondo1 has quit IRC14:15
*** jmlowe has quit IRC14:15
*** ricolin has quit IRC14:18
*** gsimondon has joined #openstack-containers14:20
*** AlexeyAbashkin has quit IRC14:21
*** serlex has quit IRC14:22
*** lpetrut has joined #openstack-containers14:22
*** mgoddard has quit IRC14:31
*** georgem1 has joined #openstack-containers14:37
*** b_bezak has quit IRC14:39
*** jmlowe has joined #openstack-containers14:39
*** mgoddard has joined #openstack-containers14:47
*** lpetrut has quit IRC14:54
*** zhubingb_ has joined #openstack-containers14:59
georgem1Any idea what can be done about https://bugs.launchpad.net/magnum/+bug/1775460 ?15:00
openstackLaunchpad bug 1775460 in Magnum "Cluster creation fails because of password complexity" [Undecided,New]15:00
*** zhubingb_ has quit IRC15:04
*** armaan has joined #openstack-containers15:05
*** zhubingb_ has joined #openstack-containers15:10
*** gsimondon has quit IRC15:14
*** ricolin has joined #openstack-containers15:25
*** dave-mccowan has joined #openstack-containers15:25
imdigitaljimgeorge1: just curious roughly at what point of cluster creation does this failure get thrown for you?15:27
georgem1imdigitaljim: stack resource: "kube_masters" and Stack Resource Type: OS::Heat::ResourceGroup15:28
georgem1imdigitaljim: https://photos.app.goo.gl/6bONrIZsI7ZdfNzS215:30
imdigitaljimthe vms got created?15:31
imdigitaljimget*15:32
*** dave-mccowan has quit IRC15:32
*** iranzo has quit IRC15:32
georgem1nope15:32
imdigitaljimhmm so not that far15:33
imdigitaljimok, ill see if i can spot something15:33
imdigitaljimthanks!15:33
*** zhubingb_ has quit IRC15:40
georgem1thank you too15:45
imdigitaljimflwang1: https://review.openstack.org/#/c/573639/ I was thinking about how to resolve this issue as well and I believe this approach will cause other complexity-related issues such as if you ever want to scale masters. What do you all think about leaving it as one resource group but adding logic on the k8s/fragments side for bootstrapping that 'primary master'.  Make a parameter out of the %index% value and write it to the heat-params15:53
imdigitaljimfile and then during the software configuration/cloud-init have a set of common master scripts but also some that run for a bootstrapping master vs a regular master.  The host can just look at the heat-param and if it is master 0 you could execute one thing and if not exit 0 and continue to the common scripts (if any). Alternatively, if the host is master 1, it can wait for some indication that master 0 has completed its work and then15:53
imdigitaljimcontinue. There could be retry logic, maybe wait a few seconds between each check for the okay to continue, and maximum retries before returning a failure to heat through the wait handler so it wont have to wait until the timeout rather than reasonable time passed. The only thing here is if new hosts come up as master 0 (e.g. master 0 failed and a new one brought up) later they would need to check if the cluster is already active, it15:53
imdigitaljimshould just then take the master 1 route.15:53
*** mago_ has quit IRC15:53
*** ktibi_ has joined #openstack-containers15:57
*** ktibi has quit IRC16:00
imdigitaljimhttps://review.openstack.org/#/c/571583/ +1 for this feature but would someone enlighten me on what would be the rationale for whether I'd use one or the other (in terms of Magnum usage)? I'm doing Magnum/Queens Fedora-k8s on Fedora27 but I've only been getting it functional with cgroupfs (against the system default of both Fedora and magnum)16:06
*** dardelean_ has quit IRC16:12
*** dardelean_ has joined #openstack-containers16:12
*** armaan has quit IRC16:16
*** armaan has joined #openstack-containers16:16
*** vijaykc4 has quit IRC16:23
*** lpetrut has joined #openstack-containers16:29
*** gsimondon has joined #openstack-containers16:29
*** mgoddard has quit IRC16:32
*** fragatina has quit IRC16:33
*** fragatina has joined #openstack-containers16:34
*** vijaykc4 has joined #openstack-containers16:34
*** zhubingb_ has joined #openstack-containers16:41
*** Guest14735 is now known as sdake16:41
*** zhubingb_ has quit IRC16:46
*** lpetrut has quit IRC16:59
*** pcaruana has quit IRC17:02
*** chhagarw has quit IRC17:10
*** markguz_ has joined #openstack-containers17:12
*** markguz_ has quit IRC17:16
*** dardelean_ has quit IRC17:20
*** dardelean_ has joined #openstack-containers17:20
*** iranzo has joined #openstack-containers17:21
*** iranzo has joined #openstack-containers17:21
-openstackstatus- NOTICE: The Zuul scheduler was offline briefly to clean up from debugging a nodepool issue, so changes uploaded or approved between 16:50 and 17:15 UTC may need to be rechecked or reapproved (all already queued changes are in the process of being reenqueued now)17:23
*** dardelean_ has quit IRC17:25
*** lpetrut has joined #openstack-containers17:41
*** dardelean_ has joined #openstack-containers17:54
*** dardelean_ has quit IRC17:58
*** georgem1 has left #openstack-containers18:00
*** markguz_ has joined #openstack-containers18:10
*** markguz_ has quit IRC18:25
*** yamamoto has quit IRC18:25
*** yamamoto has joined #openstack-containers18:26
*** yamamoto has quit IRC18:31
*** ricolin has quit IRC18:36
*** zhubingb_ has joined #openstack-containers18:43
*** zhubingb_ has quit IRC18:48
*** fragatina has quit IRC18:49
flwang1imdigitaljim: still around?18:58
imdigitaljimyeah19:00
flwang1imdigitaljim: thanks for your comments, as I mentioned in the launchpad bug, I know  https://github.com/kubernetes/kubernetes/issues/44165 is the root cause19:03
flwang1however, it has been there more than 1 year, so we may have to fix it in magnum19:04
flwang1imdigitaljim: and i like your idea in the comments #119:04
flwang1not sure if we can pass the %index% as a heat param, if we can, that could work19:05
imdigitaljimyeah i can confirm we can (i did it on a basic centos7 driver i was doing)19:05
imdigitaljimnode_id: '%index%'19:06
imdigitaljimsomething like that would be sufficient to pass the index on the OS::Heat::ResourceGroup19:07
imdigitaljimthen if [ "$NODE_ID" == "0" ]; then (do singleton operations) fi19:08
imdigitaljimthis method works as expected19:09
imdigitaljimi can setup a PR with it over next week if you19:10
imdigitaljim'd like19:10
flwang1thank you, i'm keen to test it asap because it will block our release. it's kind a good idea19:11
flwang1thank you for the valuable comments19:11
flwang1are you using magnum in your prod?19:12
flwang1imdigitaljim: i will propose a patch set today19:16
flwang1here is Sat now ;)19:16
imdigitaljimWe want to get Magnum to production soon but we've got some patching we've had to do (upstreaming soon) and some tuning for environment but we hope to have it in prod ASAP19:28
imdigitaljimbut only in a dev capacity right now19:28
*** yamamoto has joined #openstack-containers19:29
flwang1imdigitaljim: cool, pls pls upstream and i'm happy to review them19:32
*** jmlowe has quit IRC19:33
flwang1as for the multi masters bug, did you see it in your test env?19:33
flwang1imdigitaljim: ^19:34
*** armaan has quit IRC19:34
*** armaan has joined #openstack-containers19:34
*** yamamoto has quit IRC19:37
imdigitaljimflwang1: yes it has intermittently come up20:01
flwang1imdigitaljim: cool20:01
flwang1looking forward to reviewing20:01
*** armaan has quit IRC20:12
*** armaan has joined #openstack-containers20:13
*** mgoddard has joined #openstack-containers20:22
*** armaan has quit IRC20:31
*** armaan has joined #openstack-containers20:31
*** iranzo has quit IRC20:35
*** vijaykc4 has quit IRC20:44
*** ktibi_ has quit IRC20:47
*** gsimondon has quit IRC21:01
*** fragatina has joined #openstack-containers21:02
*** fragatina has quit IRC21:03
*** fragatina has joined #openstack-containers21:04
*** fragatina has quit IRC21:05
*** fragatina has joined #openstack-containers21:05
*** vijaykc4 has joined #openstack-containers21:05
*** mgoddard has quit IRC21:07
*** lpetrut has quit IRC21:15
*** jmlowe has joined #openstack-containers21:19
*** markguz_ has joined #openstack-containers21:26
*** markguz_ has quit IRC21:31
*** armaan has quit IRC21:41
*** jmlowe has quit IRC21:43
*** jmlowe has joined #openstack-containers21:48
*** jmlowe has quit IRC22:00
*** jmlowe has joined #openstack-containers22:01
*** ykarel has joined #openstack-containers22:09
*** ykarel_ has joined #openstack-containers22:14
*** jmlowe has quit IRC22:15
*** ykarel has quit IRC22:16
*** jmlowe has joined #openstack-containers22:20
*** harlowja has joined #openstack-containers22:30
*** vijaykc4 has quit IRC22:40
*** hongbin has quit IRC22:43
*** ykarel_ has quit IRC22:49
*** ykarel_ has joined #openstack-containers23:04
*** ykarel_ has quit IRC23:09
*** markguz_ has joined #openstack-containers23:22
*** vijaykc4 has joined #openstack-containers23:43
*** markguz_ has quit IRC23:44
*** markguz_ has joined #openstack-containers23:58

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