Orphaned templates in VirtualCenter

[fcapizzo]

I logged in to the VIC this morning and I noticed that 2 of my templates are greyed out and listed as "orphaned". I looked at the Events within VirtualCenter and noticed the following:

info 11/10/2007 1:21:19 AM Win2003 R2 SP2 STD x64 10GB on host esx102.something.com in Datacenter is disconnected

warning 11/10/2007 1:21:32 AM Win2003 R2 SP2 STD x64 10GB does not exist on esx102.something.com in Datacenter

So from there I did some digging with the logs on that particular ESX host and this is what I came up with:

From /var/log/messages:

Nov 10 01:21:10 esx102 watchdog-vpxa: '/usr/sbin/vpxa' exited after 1536438 seconds

Nov 10 01:21:10 esx102 watchdog-vpxa: Executing '/usr/sbin/vpxa'

Nov 10 01:21:21 esx102 /usr/lib/vmware/hostd/vmware-hostd[1438]: Accepted password for user vpxuser from 127.0.0.1

From /var/log/secure:

Nov 10 01:21:20 esx102 xinetd[1326]: START: vmware-authd pid=27739 from=172.21.2.132

Nov 10 01:21:20 esx102 vmware-authd[27739]: login from 172.21.2.132 as vpxuser

From /var/log/vmware/vpx/vpxa.log:

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

Unexpected return result. Expect 1 sample, receive 2

So it looks like vpxa failed if I'm reading these logs correctly, but I'm not entirely sure. The same exact thing (4 templates showed up as "orphaned") happened earlier in the week with a different ESX host in the DataCenter (it was part of a different Cluster as well), so I'm trying to nail this issue down. Lastly, while I'm certain that I could simply remove the Orphaned templates from the VIC and re-add them by browsing the Datastore, I am unable to make changes at this company due to a freeze in IT changes. So the more investigative "look but don't touch" work I can do, the better.

[TCronin]

Can you restart the management agents on the host?

service mgmt-vmware restart

[fcapizzo]

TCronin, I can't try any changes at the moment. I'm trying to find out why this happened first before doing any changes.

[wtreutz]

the both deamons mgmt-vmware and vmware-vpxa make the connection between ESX-Server and VCMS-Server.

you can restart both without reboot the whole ESX-Server.

I had seen orphanded templates after connection lost between ESX-Server and VCMS-Server.

After restart the VCMS the connection was "rebuild" and the state "orphaned" gone.

regards

Werner

Message was edited by: wtreutz

look at the following --> http://communities.vmware.com/thread/51923?tstart=0&start=45

[Sorwyn]

We used to see this often. Restarting the Virtual Centre service normally fixes it.

[Rajeev_S]

Yes, I have faced this issue. I will re-add it to the inventory. That worked for me. Yet to find the root cause. Do anyone know that?

[VirtualNoitall]

Hello,

There is a known issue with tmeplates going orphaned. Call tech support, and if your issue has the same cause, you can get attached to a PR to get this fixed. Right now there are a half dozen or so companies on it and the more there are the more likely it is to get fixed ina timely manor.

To correct the issue for us we either restart the VC server service or remove and add the tmeplates back to inventory.