Hi all,
I am trying to reboot my OpenNebula cluster. I am using onehost flush $nodename
for live-migrating all the running VMs off the node to be rebooted. However, one of my VMs refuse to migrate, with the following error message visible in Sunstone:
Scheduler Error
Mon May 19 09:07:34 2025: Cannot dispatch VM to any Host. Possible reasons: Not enough capacity in Host or System DS, dispatch limit reached, or limit of free leases reached.
From the command line, the error message is different, and somewhat confusing:
# onevm migrate --live $vmid $nodeid
[one.vm.migrate] Migrate action is not available for state RUNNING
The same error message is also in sched.log
:
Mon May 19 09:09:51 2025 [Z0][VM][E]: Error deploying virtual machine XXXX to HID: 62. Reason: [one.vm.migrate] Migrate action is not available for state RUNNING
Mon May 19 09:09:51 2025 [Z0][VM][E]: Error deploying virtual machine XXXX to HID: 61. Reason: [one.vm.migrate] Migrate action is not available for state RUNNING
I tried to shutdown OpenNebula processes and run onedb fsck
just in case there was some kind of inconsistency in the database, but even after that the VM cannot be live-migrated away from the host to be rebooted, with the same error message.
This is ONe 6.10.0 CE on AlmaLinux 9 hosts.
How can I tell ONe to migrate this VM? Thanks for any hint,
-Yenya