rescue of boot from volume image fails when --image is not specified
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Triaged
|
Medium
|
Lee Yarwood |
Bug Description
From an environment where all storage is on ceph, it was not possible to omit the --image parameter when rescuing a boot-from-volume instance as suggested near the end of this doc:
https:/
The doc implies that omitting --image will use the image of the rescued instance as the rescue image.
Works:
openstack server rescue --image d0d00858-
Does not work:
openstack server rescue 81057513-
Not specifying --image appears to succeed with the command line client:
REQ: curl -g -i --insecure -X POST http://
http://
However in the nova compute log we see this stacktrace http://
Environment
===========
1. OpenStack Victoria / OpenStack-Ansible 22.0.0
2. Which hypervisor did you use?
Libvirt + KVM libvirt 6.0.0-0ubuntu8.
2. Which storage type did you use? Ceph
3. Which networking type did you use?
Neutron / LinuxBridge
Changed in nova: | |
status: | New → Triaged |
importance: | Undecided → Medium |
assignee: | nobody → Lee Yarwood (lyarwood) |
TO follow up on this - we subsequently found that it was not possible to use the original OS image as the FSID(?) matched and even though the rescue image was attached correctly the wrong device was used for booting.