system.service.vmware.vsla

[barnette08]

When attempting to clone vApps within the cloud or even publish from cloud to catalog I see the following error:

vCD UI Error:

"A general system error occurred: The virtual machine requires the vcdFQDN/system.service.vmware.vsla.vcenterFQDN service, which is not available on the host."

vcloud-container-debug.log shows that the error is happening within vCenter, but vCenter references the same error shown in the vCD UI.

vcloud-container-debug.log:

2014-01-06vcloud-container-debug.log 13:04:57,390 | DEBUG	| ActiveMQ Session Task	| TaskManager	| Handling completion update from MessageBusAdapter for task [vcId=1c495c36-1f5b-4322-bc19-b77b8526b898, moref=task-40207] with state ERROR |
2014-01-06 13:04:57,390 | DEBUG	| ActiveMQ Session Task	| WaitHandle	| Task updated. task = [moref=task-40207, state=ERROR, taskName=CloneVM_Task, progress=null, entityName=vnxfile01 (a6632beb-77e6-4749-8577-6a7ef876ab07)], polled from vc = false |

I have rebooted the host in question, as well as failed cell01 over to cell02 in hopes to isolate the issue, but this didn't seem to work either.  Has anyone else run into this recently?

[cfor]

If you check the host in vCloud administration does it show any errors on the host (or any hosts)?  (System->Administration->Manage->Hosts)

If you are able to isolate the host having the issue might be able to unprepared and prepare the host and see if that helps.  Also reboot of the host might help if it really is an agent issue.

[barnette08]

cfor, I've checked the host in manage & monitor and it doesn't show any errors.  I also had already rebooted the host and migrated the VMs to different hosts but the error message still shows even on different hosts. 

[_morpheus_]

The error means the host is either not prepared with VCD agent, or the host is disabled in VCD (in system, manage & monitor, hosts)

[barnette08]

morpheus, In this particular case both are true on all hosts on the cluster in question.  I will do a unprepare/prepare though so see if it will workout any of the issues.  These hosts have been prepared for over a year, but only recently have been moved to a different cluster.

[barnette08]

Since this is a newer cluster with very few VMs, I was able to un-prepare and then re-prepare all the hosts in the cluster.  Something must have been hung up in the vCD agent?

[_morpheus_]

Was VCD or any of the hosts recently upgraded?

[barnette08]

Nope, they belonged to other PvDCs.  I only moved them to this new cluster for to have an isolated PvDC for a new project we are working on.

[QuincySanders]

Hi barnette08, did you ever find a resolution to this ?

[IamTHEvilONE · Accepted solution]

when a host is prepared, we set a flag on it to ensure that the host is associated to a particular vcloud instance.

If the host is disabled, then the flag is turned off and VMs will fail vMotion compatibility (preventing VMs from running on the host).

Ideally an unprepare/prepare should do the trick, or just enable the host if it was disabled.

[QuincySanders]

IamTHEvilONE, thanks for your response; We meet again. Thank you. In my case, I'd already discovered that one of my 4 vCD hosts was disabled. I discovered this when I attempted to do a vMO and when I click on that particular server it displayed that error message. That prompted me to further look at this particular host, only to determine it was disabled. Once enabled all is well. Of course, all of this stems from vCC flat out not working and BCS not being able to provide me with a solution or alternative to the problem I'm facing, which is how to migrate my vApps from one vCD org to another vCD org with different vCD's and different vCenters. fml

[barnette08]

Quincy, I never marked this as correct - but if you take a look at one of my replies, as unprepare/prepare worked.

[QuincySanders]

Hi barnette08, thanks for the reply. I usually only attempt the responses marked as correct. In my case, one of my hosts was disabled and I only need to re-enable it.