Bug #3151

VM stops running and can't be restarted (UUID already exists

Added by Christian Hüning about 5 years ago. Updated about 5 years ago.

Status:ClosedStart date:08/13/2014
Priority:HighDue date:
Assignee:-% Done:

0%

Category:-
Target version:-
Resolution:worksforme Pull request:
Affected Versions:OpenNebula 4.8

Description

Since the uprgade to 4.8 I am sometimes getting a deployment error : error: operation failed: domain 'one-160' already exists with uuid d0b479a7-86e9-49fd-a8f6-7fd3be56b201

Here's a complete log of a machine that I created yesterday, which ran successfully for a very brief amount of time, then shutdown. When I noticed today, it couldn't be restarted caused by the uuid error as shown below. What's going on here?

Tue Aug 12 17:29:23 2014 [Z0][DiM][I]: New VM state is ACTIVE.
Tue Aug 12 17:29:23 2014 [Z0][LCM][I]: New VM state is PROLOG.
Tue Aug 12 17:29:59 2014 [Z0][LCM][I]: New VM state is BOOT
Tue Aug 12 17:29:59 2014 [Z0][VMM][I]: Generating deployment file: /var/lib/one/vms/160/deployment.0
Tue Aug 12 17:30:00 2014 [Z0][VMM][I]: ExitCode: 0
Tue Aug 12 17:30:00 2014 [Z0][VMM][I]: Successfully execute network driver operation: pre.
Tue Aug 12 17:30:00 2014 [Z0][VMM][I]: ExitCode: 0
Tue Aug 12 17:30:00 2014 [Z0][VMM][I]: Successfully execute virtualization driver operation: deploy.
Tue Aug 12 17:30:00 2014 [Z0][VMM][I]: ExitCode: 0
Tue Aug 12 17:30:00 2014 [Z0][VMM][I]: Successfully execute network driver operation: post.
Tue Aug 12 17:30:00 2014 [Z0][LCM][I]: New VM state is RUNNING
Tue Aug 12 17:30:04 2014 [Z0][DiM][I]: New VM state is POWEROFF
Wed Aug 13 12:39:20 2014 [Z0][DiM][I]: New VM state is ACTIVE.
Wed Aug 13 12:39:20 2014 [Z0][LCM][I]: New VM state is BOOT_POWEROFF
Wed Aug 13 12:39:20 2014 [Z0][VMM][I]: Generating deployment file: /var/lib/one/vms/160/deployment.1
Wed Aug 13 12:39:21 2014 [Z0][VMM][I]: ExitCode: 0
Wed Aug 13 12:39:21 2014 [Z0][VMM][I]: Successfully execute network driver operation: pre.
Wed Aug 13 12:39:21 2014 [Z0][VMM][I]: Command execution fail: cat << EOT | /var/tmp/one/vmm/kvm/deploy '/var/lib/one//datastores/103/160/deployment.1' '141.22.29.3' 160 141.22.29.3
Wed Aug 13 12:39:21 2014 [Z0][VMM][I]: error: Failed to create domain from /var/lib/one//datastores/103/160/deployment.1
Wed Aug 13 12:39:21 2014 [Z0][VMM][I]: error: operation failed: domain 'one-160' already exists with uuid d0b479a7-86e9-49fd-a8f6-7fd3be56b201
Wed Aug 13 12:39:21 2014 [Z0][VMM][E]: Could not create domain from /var/lib/one//datastores/103/160/deployment.1
Wed Aug 13 12:39:21 2014 [Z0][VMM][I]: ExitCode: 255
Wed Aug 13 12:39:21 2014 [Z0][VMM][I]: Failed to execute virtualization driver operation: deploy.
Wed Aug 13 12:39:21 2014 [Z0][VMM][E]: Error deploying virtual machine: Could not create domain from /var/lib/one//datastores/103/160/deployment.1
Wed Aug 13 12:39:22 2014 [Z0][LCM][I]: Fail to boot VM. New VM state is POWEROFF

History

#1 Updated by Ruben S. Montero about 5 years ago

Hi Christian

Did you triggered the power off action, or was it automatically moved to power off? Are you using the last packages, those release yesterday? When the VM is not found in the hypervisor (e.g. by virsh list) OpenNebula assumes that it is power off, sometimes there may be a race condition between the boot and monitor operations, that may lead to a poweroff VM but running at the hypervisor. In that case any attempt to reboot the VM will fail becasue it is running.

The last version of the packages are supposed to fix this....

Cheers

Ruben

#2 Updated by Christian Hüning about 5 years ago

It was automatically moved to poweroff. Nope, did not install the new patch. I've done that right now, and until now things are looking good. Will monitor the behaviour.

Thanks for the response!

#3 Updated by Ruben S. Montero about 5 years ago

  • Status changed from Pending to Closed
  • Resolution set to worksforme

Ok thanks for the update!, close and reopen if needed

Also available in: Atom PDF