Feature #75
Include Error state
| Status: | Closed | Start date: | ||
|---|---|---|---|---|
| Priority: | High | Due date: | ||
| Assignee: |  Ruben S. Montero | % Done: | 0% | |
| Category: | Core & System | |||
| Target version: | Release 1.4 | |||
| Resolution: | fixed | Pull request: |
Description
OpenNebula 1.2 periodically polls a VM, in case the VM could not be found or if it failed OpenNebula assumes that it is done and silently puts it in DONE state. It would be better to mark this VM as in an ERROR state, so the admin can manually remove it.
Additionally, we can still poll the VM in case the hypervisor recovers from the error. During the Error state the capacity allocated to the VM can not be freed
Related issues
History
#1
Updated by Ruben S. Montero about 12 years ago
This has been implemented in the OpenNebula_XML branch. The changes include:
- A VM not found is put in UNKNOWN state
- OpenNebula core also polls VMs in UNKNOWN state
- If the VM is recovered, e.g. by rebooting it or restoring the connection with the physical node, the VM is placed in RUNNING state
- A VM in the UNKNOWN state can be restarted (
onevm restart). Note that VM disk images are in place and we can try to boot the VM.
- New state machine
- New commands for onevm
#2
Updated by Ruben S. Montero almost 12 years ago
- Status changed from New to Closed
- Resolution set to fixed