| *** spsurya has joined #openstack-monasca | 02:32 | |
| openstackgerrit | Li Yingjun proposed openstack/monasca-events-api master: Fixes policy for admin https://review.opendev.org/689677 | 06:33 |
|---|---|---|
| *** pcaruana has joined #openstack-monasca | 07:16 | |
| openstackgerrit | Vishal Manchanda proposed openstack/monasca-ui master: Use Horizon project template for django jobs https://review.opendev.org/687621 | 07:18 |
| *** witek has joined #openstack-monasca | 07:33 | |
| *** dougsz has joined #openstack-monasca | 08:00 | |
| *** openstackgerrit has quit IRC | 09:37 | |
| chaconpiza | Hi, we have found a bug in editing an Alarm-Definition when removing the option Deterministic. More info and how to reproduce it: https://storyboard.openstack.org/#!/story/2006750 | 10:15 |
| witek | chaconpiza: thanks for reporting, I assume `alarm-list` returns also not updated expressions? | 10:26 |
| chaconpiza | Right | 10:27 |
| witek | seems to me to be a bug in thresholding engine | 10:28 |
| chaconpiza | I think, the issue is in Monasca-api, since I was able to reproduce by Patching the alarm-definition using a direct call to the API. | 10:28 |
| chaconpiza | maybe | 10:28 |
| witek | alarm definition gets updated, alarm itself not | 10:29 |
| chaconpiza | In general it is not permitted to edit the expression or to change/add/remove the match_by. | 10:29 |
| chaconpiza | https://github.com/openstack/monasca-api/blob/master/docs/monasca-api-spec.md#patch-alarm-definition | 10:30 |
| chaconpiza | However is possible to remove/add the deterministic token from the expression: max(cpu.percent,deterministic)>99 | 10:31 |
| chaconpiza | into max(cpu.percent)>99 | 10:31 |
| *** dougsz has quit IRC | 10:38 | |
| *** dougsz has joined #openstack-monasca | 10:40 | |
| *** openstackgerrit has joined #openstack-monasca | 11:01 | |
| openstackgerrit | Bharat Kunwar proposed openstack/monasca-persister master: Support default_retention_policy for influxdb https://review.opendev.org/689712 | 11:01 |
| *** trident has quit IRC | 11:11 | |
| *** trident has joined #openstack-monasca | 11:15 | |
| openstackgerrit | Witold Bedyk proposed openstack/monasca-persister master: Move data_points to repository class https://review.opendev.org/689513 | 11:42 |
| *** dougsz has quit IRC | 11:46 | |
| brtknr | witek: what do you think about supporting default_retention_policy in config? As before, if default_retention_policy is not specified, nothing happens | 11:56 |
| *** dougsz has joined #openstack-monasca | 12:04 | |
| *** jgrassler has joined #openstack-monasca | 12:06 | |
| *** adriancz has joined #openstack-monasca | 12:16 | |
| openstackgerrit | Witold Bedyk proposed openstack/monasca-persister master: Move data_points to repository class https://review.opendev.org/689513 | 13:04 |
| openstackgerrit | Witold Bedyk proposed openstack/monasca-persister master: Move data_points to repository class https://review.opendev.org/689513 | 13:08 |
| openstackgerrit | Witold Bedyk proposed openstack/monasca-events-api master: Bump the openstackdocstheme extension to 1.20.0 https://review.opendev.org/674204 | 13:12 |
| witek | brtknr: good change, left you comments in review | 14:23 |
| brtknr | witek: thanks, what do you mean by "for consistency we should also cover the case when db_per_tenant is disabled" | 14:23 |
| brtknr | should we also create a retention policy for a case when db_per_tenant is not allowed? | 14:24 |
| brtknr | should we always create a database if it doesnt exist? | 14:24 |
| witek | I think, I'd create/alter the retention policy for the case when db_per_tenant=false | 14:26 |
| brtknr | witek: I am not sure how to avoid making a call to check if the retention policy is correct every single time | 14:28 |
| witek | we just need to make it once, at the start | 14:30 |
| brtknr | witek: at present, we dont create a database if db_per_tenant is disabled | 14:36 |
| brtknr | should we change this behaviour to creating a database/rp in all cases? | 14:36 |
| witek | I think these are two different topics; I see also two other points we should consider: | 14:40 |
| witek | * what should happen if `default_retention_policy` gets updated after DB creation? | 14:41 |
| witek | * should we consider cases when different persister instances have inconsistent retention period values? | 14:42 |
| brtknr | witek: I think default_retention_policy is only applicable during database creation | 14:48 |
| brtknr | If we start applying retention policy to existing databse, it will be dangerous | 14:48 |
| brtknr | Since administrators might want to specify different retentention policy for different tenants | 14:49 |
| brtknr | We do not want to override this behaviour | 14:49 |
| openstackgerrit | Bharat Kunwar proposed openstack/monasca-persister master: Support default_retention_policy for influxdb https://review.opendev.org/689712 | 15:22 |
| *** trident has quit IRC | 15:29 | |
| openstackgerrit | Bharat Kunwar proposed openstack/monasca-persister master: Support default_retention_policy for influxdb https://review.opendev.org/689712 | 15:31 |
| *** trident has joined #openstack-monasca | 15:32 | |
| openstackgerrit | Bharat Kunwar proposed openstack/monasca-persister master: Support default_retention_policy for influxdb https://review.opendev.org/689712 | 16:07 |
| *** witek has quit IRC | 16:11 | |
| *** spsurya has quit IRC | 16:40 | |
| *** dougsz has quit IRC | 17:51 | |
| *** pcaruana has quit IRC | 20:16 | |
| *** openstackgerrit has quit IRC | 22:07 | |
| joadavis | New story for a very useful certificate expiration monitoring feature - https://storyboard.openstack.org/#!/story/2006753 | 22:08 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!