nova handler "Restart nova-libvirt container" fails
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
kolla-ansible |
Invalid
|
Undecided
|
Unassigned | ||
Stein |
Fix Committed
|
Medium
|
Mark Goddard |
Bug Description
Kolla version: kolla-ansible=
Ansible version: 2.9.0.dev0 894d19b1088d8d4
Execution host runs Debian 10, target hosts are CentOS 7.
I have a recently created multinode setup with 1 pure controller node and 1 controller+compute node. Now I wanted to enable support for provider networks (enable_
For my pure controller node the playbook completes (probably because it needs no libvirt), however, for my compute node it doesn't. There is one VM running on this compute node.
Something in the handler's bool chain goes wrong:
RUNNING HANDLER [nova : Restart nova-libvirt container] *******
Thursday 05 September 2019 12:34:50 +0200 (0:00:00.180) 0:05:11.410 ****
fatal: [redacted]: FAILED! => {
"msg": "The conditional check 'config_
}
Hard to figure out which of these is the culprit. Any hints how to further troubleshoot/
description: | updated |
description: | updated |
Changed in kolla-ansible: | |
status: | New → Incomplete |
information type: | Public → Public Security |
information type: | Public Security → Public |
no longer affects: | kolla-ansible/rocky |
I worked around it by destroying everything and re-deploying from scratch so I can't try possible suggestions anymore. Would still be cool to avoid this for others.