Unable to connect to server at 'cloudadmin:16514': Connection refused

Community Support
1

Hi,

I have setup Opennebula for research purpose and i am trying to do Live migration( Migrate VM from one host to another) with shared transfer manager. I wasnt successful and end up with following error.

VM.log

Wed Mar 11 15:39:11 2015 [LCM][I]: New VM state is MIGRATE
Wed Mar 11 15:39:11 2015 [VMM][I]: Successfully execute transfer manager driver operation: tm_premigrate.
Wed Mar 11 15:39:11 2015 [VMM][I]: ExitCode: 0
Wed Mar 11 15:39:11 2015 [VMM][I]: Successfully execute network driver operation: pre.
Wed
Mar 11 15:39:11 2015 [VMM][I]: Command execution fail:
/var/tmp/one/vmm/kvm/migrate ‘one-17’ ‘cloudadmin’ ‘cloudadmin3’ 17
cloudadmin3
Wed Mar 11 15:39:11 2015 [VMM][E]: migrate: Command
"virsh --connect qemu:///system migrate --live one-17
qemu://cloudadmin/system" failed: error: unable to connect to server at
’cloudadmin:16514’: Connection refused
Wed Mar 11 15:39:11 2015 [VMM][E]: Could not migrate one-17 to cloudadmin
Wed Mar 11 15:39:11 2015 [VMM][I]: ExitCode: 1
Wed Mar 11 15:39:11 2015 [VMM][I]: Failed to execute virtualization driver operation: migrate.
Wed Mar 11 15:39:11 2015 [VMM][E]: Error live migrating VM: Could not migrate one-17 to cloudadmin
Wed Mar 11 15:39:12 2015 [LCM][I]: Fail to live migrate VM. Assuming that the VM is still RUNNING (will poll VM).

I am stuck
 Pls help.

Thanks in advance

Sindhu

2

It seems a problem with “cloudadmin” host, take a look at the relevant line
in the logs. Also check qemu.logs at the host to debug your configuration:

Wed Mar 11 15:39:11 2015 [VMM][E]: migrate: Command
"virsh --connect qemu:///system migrate --live one-17
qemu://cloudadmin/system" failed: error: unable to connect to server at
’cloudadmin:16514’: Connection refused

3

Hey Ruben,

I have now succeded and able to do live migration. Thanks !!!

I dont know if this is a resolver, but I added the SSH public key of front end in the “add SSH contextualization” tab in the template and it worked
:slight_smile:

1 Like
4

Just ran into this
The problem is the tcp port selected

’cloudadmin:16514’: Connection refused

the correct port range for qemu live migration is 49152 to 49215.
If you have a firewall on the host it’s likely not open, and especially not the port 16514 that libvirt chose in the above example. For mine it was 16509. It should use a hypervisor specific range. IDK yet why it instead did that, but just for the next person running into this, this is where you can start looking deeper.
edit:
comes from
#tcp_port = "16509" in libvirtd.conf