Feature #1500
fails gracefully when hypervisor has gone away
| Status: | New | Start: | 02/22/2012 | |
|---|---|---|---|---|
| Priority: | Low | Due date: | ||
| Assigned to: | - | % Done: | 0% |
|
| Category: | VM management | |||
| Target version: | - | |||
| Backlog: | No | Difficulity: | ||
| Votes: | 0 |
Description
(Even just turning off a hypervisor seems to cause issues).
In this case, I've rebuilt the hypervisor. I'd like to destroy the old VMs from foreman. Clicking "host -> delete" ends with this stack track. Unmanaging the node doesn't seem to help.
NoMethodError
undefined method `connect' for nil:NilClass
/usr/share/foreman/app/models/orchestration/libvirt.rb:51:in `queue_libvirt_destroy'
/usr/share/foreman/app/models/orchestration.rb:63:in `destroy'
/usr/share/foreman/app/controllers/hosts_controller.rb:116:in `destroy'
Foreman v0.4.1
History
Updated by Adrian Bridgett over 1 year ago
Hmm, bizarely it's now working. I've trying from the "all hosts" screen rather than the "host foo -> delete" screen this time.
Updated by Benjamin Papillon 6 months ago
- Tracker changed from Bug to Feature
- Subject changed from fails when hypervisor has gone away to fails gracefully when hypervisor has gone away
Aside from using computer ressources instead of Hypervisor code, try to get a graceful error when hypervisor/computer ressource is not available
