| opendevreview | chandan kumar proposed openstack/watcher stable/2024.2: [host_maintenance] Pass des hostname in add_action solution https://review.opendev.org/c/openstack/watcher/+/949442 | 06:14 |
|---|---|---|
| opendevreview | chandan kumar proposed openstack/watcher-tempest-plugin master: Add tests for host_maintenance strategy with backup node https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/947944 | 06:22 |
| amoralej | fyi, i sent https://review.opendev.org/c/openstack/watcher/+/949225 , please, review when you have a chance, thanks! | 09:25 |
| amoralej | Hi, while preparing the spec for the SKIPPED state in Actions, I found this https://github.com/openstack/watcher/blob/f38ab70ba46756b2c3ae74b1a2fafdb39ac58cc7/watcher/api/controllers/v1/action.py#L49-L51 so, apparently, someone thought on this behavior, althoug it is not implemented | 10:26 |
| amoralej | the only two cases where CANCELLED status is used so far is for actions which were ONGOING when starting the applier (so the applier was stopped right in the middle of an action execution) and when an action plan is cancelled, all the nested actions are set to CANCELLED | 10:29 |
| amoralej | but I understand that our two new use cases, cloud admin explicitely excluding an action execution from an action pland and pre-condition to find some specific conditions, would be also valid for CANCELLED state | 10:31 |
| amoralej | wdyt? ^ , use CANCELLED for those or implement new SKIPPED one as we discussed? | 10:31 |
| jgilaber | using CANCELLED makes sense, but I think I would still clearly differentiate in the logs or elsewhere the two cases (i.e cancelled because of some pre-condition or cancelled because the operator actually executed a cancel command) | 10:44 |
| amoralej | adding a new filed cancel_reason would be good | 10:57 |
| amoralej | user visible | 10:57 |
| jgilaber | yes, that would solve my concern | 11:06 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!