vmotion not working: A general system error occurred: Source detected that destination failed to resume.

[gpalmer1981]

Been having this problem all day. I am trying to evacuate a host to do Host update but have one vm that wont vmotion. I cannot power this vm down and would like to vmotion it live. I have checked the vmware.log file and have the following in there:

Nov 26 15:45:20.716: vcpu-0| Migrate: Failed to resume on the destination: Lock was not free. Exiting...

Nov 26 15:45:20.716: vcpu-0| MigrateSetStateFinished: type=2 new state=11

Nov 26 15:45:20.716: vcpu-0| MigrateStateUpdate: Transitioning from state 10 to 11.

Nov 26 15:45:20.717: vcpu-0| Migrate_SetFailure: Failed to resume destination VM: Lock was not free.

Nov 26 15:45:20.717: vcpu-0|

Nov 26 15:45:20.718: vcpu-0| Msg_Post: Error

Nov 26 15:45:20.718: vcpu-0| http://msg.checkpoint.destination.resume.fail Failed to resume destination VM: Lock was not free.

Has anyone got any ideas?

Please

[idle-jam]

Can you check of the current VMFS which VMFS seats on have like 20% free space? also does the destination Host has access to the VMFS which the current VM seats on?

iDLE-jAM | VCP 2, VCP 3 & VCP 4

If you found this or any other answer useful please consider the use of the Helpful or correct buttons to award points

[gpalmer1981]

The VMFS datastore has more than 20% free space. Also all hosts can see all datastores (I have 3 hosts and 4 datastores in my infrastructure). The fact that I was able to vmotion other vm's off that host to the destination host tells me it cant be a problem with the host but more an isolated issue with the VM itself.

Anyway remember I am not trying to do a storage vmotion, just a normal vmotion. In the log it mentioned about "Lock was not free. Exiting." Maybe this has something to do with it.

I also read somewhere that maybe it is failing because that particular VM is somehow already registered on the other hosts. I am having trouble vmotioning to either of my 2 hosts.

[vickey0rana]

it seems that you VM is register to multiple esx hosts that we called dual registration ... check on each ESX host by running command ps -ef | grep <vm_config_file.vmx> if it find the results on more than 2 hosts it means you have dual registration of your VM. Check this first and let me know than we think ahead...

Thanks!

[vickey0rana]

Dude!!!! my bad ... I didn't saw 2 more post listed below:

if you have dual registration than there is a downtime

here is steps to remove the dual resigteration:

>Note location of VM .vmx file and output of vmware-cmd-l

>shutdown the VM

>after shutdown, you will need to logon to the ESX host servers and use the command line find out process ID of the VM name and kill -9 the pid for that particular VM

>run ps -ef | grep <vm_config_file.vmx>

>kill PID of VM

>make sure it is not on any other ESX hosts of your cluster if it is than kill PID on all ESX for this VM

after that unregister the VM from the hosts

>register it from VC by browsing the lcation of VMx

NOTE:

May need to restart VM management services on the hosts after killing and unregistering the VMs per the instructions

service mgmt-vmware restart

service vmware-vpxa restart

Cheers!!!!!!!

[gpalmer1981]

Thanks vickey0rana your fix worked! I was able to find a process running on another Host i kill -9 ed it restarted the management services and was able to vmotion.

[vickey0rana]

Great news. .

Any reward points for me.? 😛