Compute services not getting deleted after scale down
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
tripleo |
Fix Released
|
Undecided
|
David Vallee Delisle |
Bug Description
When we scaledown, we normally source the undercloud's credentials (stackrc).
The openstack command that calls for the compute service deletion is simply setting the environment variable OS_CLOUD to the overcloud's stack name (or RootStackName), but this is not enough to hit the overcloud's nova API. Scale down fails when running on novaless deployment because of that.
(undercloud) [stack@undercloud-0 ~]$ export OS_CLOUD=overcloud
(undercloud) [stack@undercloud-0 ~]$ openstack compute service list
+----+-
| ID | Binary | Host | Zone | Status | State | Updated At |
+----+-
| 1 | nova-conductor | undercloud-
| 2 | nova-scheduler | undercloud-
| 6 | nova-compute | undercloud-
+----+-
(undercloud) [stack@undercloud-0 ~]$ for key in $( set | awk -F= '/^OS_/ {print $1}' ); do unset "${key}" ; done
[stack@
Missing value auth-url required for auth plugin password
[stack@
[stack@
+----+-
| ID | Binary | Host | Zone | Status | State | Updated At |
+----+-
| 1 | nova-conductor | controller-
| 3 | nova-scheduler | controller-
| 6 | nova-compute | compute-
+----+-
Changed in tripleo: | |
assignee: | nobody → David Vallee Delisle (valleedelisle) |
status: | New → In Progress |
Fix proposed to branch: master /review. opendev. org/c/openstack /python- tripleoclient/ +/795198
Review: https:/