[Resolved] Live Migration failing KVM - 4.14Beta2

Community Support
1

Hi Guys,

Tried Live Migration on kvm version - 4.14Beta2 (Ubuntu 14.04) and it is failing with below error :-

VM Log Error

Fri Sep 25 19:54:03 2015 [Z0][VMM][I]: Command execution fail: /var/tmp/one/vmm/kvm/migrate ‘one-2’ ‘192.168.4.201’ ‘192.168.4.200’ 2 192.168.4.200
Fri Sep 25 19:54:03 2015 [Z0][VMM][E]: migrate: Command “virsh --connect qemu:///system migrate --live --unsafe one-2 qemu+ssh://192.168.4.201/system” failed: error: internal error: Attempt to migrate guest to the same host 00020003-0004-0005-0006-000700080009
Fri Sep 25 19:54:03 2015 [Z0][VMM][E]: Could not migrate one-2 to 192.168.4.201
Fri Sep 25 19:54:03 2015 [Z0][VMM][I]: ExitCode: 1
Fri Sep 25 19:54:03 2015 [Z0][VMM][I]: Successfully execute transfer manager driver operation: tm_failmigrate.
Fri Sep 25 19:54:03 2015 [Z0][VMM][I]: Failed to execute virtualization driver operation: migrate.
Fri Sep 25 19:54:03 2015 [Z0][VMM][E]: Error live migrating VM: Could not migrate one-2 to 192.168.4.201
Fri Sep 25 19:54:03 2015 [Z0][VM][I]: New LCM state is RUNNING
Fri Sep 25 19:54:03 2015 [Z0][LCM][I]: Fail to live migrate VM. Assuming that the VM is still RUNNING (will poll VM).

ONED LOG

Fri Sep 25 19:56:15 2015 [Z0][ReM][D]: Req:2480 UID:0 VirtualMachineInfo result SUCCESS, “20…”
Fri Sep 25 19:56:15 2015 [Z0][VMM][D]: Message received: LOG I 2 Successfully execute transfer manager driver operation: tm_premigrate.

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG I 2 ExitCode: 0

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG I 2 Successfully execute network driver operation: pre.

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG I 2 Command execution fail: /var/tmp/one/vmm/kvm/migrate ‘one-2’ ‘192.168.4.201’ ‘192.168.4.200’ 2 192.168.
4.200

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG E 2 migrate: Command “virsh --connect qemu:///system migrate --live --unsafe one-2 qemu+ssh://192.168.4.201
/system” failed: error: internal error: Attempt to migrate guest to the same host 00020003-0004-0005-0006-000700080009

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG E 2 Could not migrate one-2 to 192.168.4.201

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG I 2 ExitCode: 1

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG I 2 Successfully execute transfer manager driver operation: tm_failmigrate.

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: LOG I 2 Failed to execute virtualization driver operation: migrate.

Fri Sep 25 19:56:16 2015 [Z0][VMM][D]: Message received: MIGRATE FAILURE 2 Could not migrate one-2 to 192.168.4.201

Fri Sep 25 19:56:19 2015 [Z0][VMM][D]: Message received: POLL SUCCESS 2 STATE=a CPU=2.5 MEMORY=131072

Fri Sep 25 19:56:19 2015 [Z0][VMM][D]: VM 2 successfully monitored: STATE=a CPU=2.5 MEMORY=131072
Fri Sep 25 19:56:19 2015 [Z0][ImM][D]: Datastore system (0) successfully monitored.
Fri Sep 25 19:56:20 2015 [Z0][InM][D]: Host 192.168.4.200 (0) successfully monitored.

Any suggestions how to fix this?

Regards
Kiran Ranjane

2

It seems that both your nodes have the same UUID. This can be caused because the hardware reports the same UUID or the node disk was cloned but the UUID was not updated. You can see the same problem in this email:

https://www.redhat.com/archives/libvirt-users/2011-August/msg00085.html

One way to fix this is setting the UUID manually in libvirt configuration as described here:

https://access.redhat.com/solutions/461133

3

Hi Javi,

Changing the UUID helped in resolving the issue, also I had to add below line in /etc/apparmor.d/abstractions/libvirt-qemu

owner /var/lib/one/** rw,

Thanks for you help.

Regards
Kiran Ranjane