VC bombing during VEEAM backup

[spf62]

Hello all...
We are running ESX 4.0 with 4.0 VC. We have 28 ESX hosts, with approx 380 VM's
We have been using VEEAM to do all of our VM backups for about 3 months.

One issue we have had is....every week at least once a week, the VC will stop and restart it's service. This happens at different times and there is no real pattern.
Doing this, chokes our VEEAM backup, and we do not get any backups from there on.
We have to re-run in the AM, which works fine.

We have gone to VMWare with this first and they had us:
1st - removed .NET and re-install
2nd - remove and re-install the VC server
None of which has helped
Has anyone had anything like this happen to them?

any help is appreciated

thanks

[Vitaliy S.]

Hello Steve,

Could this issue be connected to the hardware on which your vCenter Server is currently running? What is the current load on the vCenter machine, also could you please clarify if it is a physical or virtual machine? Besides, have you had a chance to look through vCenter Server log files for more information on the reason why the service is being stopped?

Thanks!

[spf62]

Currently my VC is physical.
The load is not bad at all.
Of the 28 ESX hosts, 9 are shut down by DPM.
Nothing has exceeded 50%

Yes I did review the logs with VMware Engineer, which led to the removal and re-install of .NET and removal and re-installation of the VC itself.

The odd part is it seems to occur mostly on Saturdays, no specific time, but only during backups.

We looked into other tasks running at the same time, but nothing is scheduled to run anywhere near the times it has failed

[Gostev]

Do you use any monitoring or reporting solution? These can create very high load on vCenter server sometimes.

[spf62]

We are testing VFoglite by Quest software...

[jgremillion]

I had this same issue about a month ago. Turned out if was a problem with the SQL server that was hosting our vCenter DB. I ended up recreating our vCenter DB on a dedicated SQL server for vCenter only.

[spf62]

Currently our SQL server DB is on it's own physical dedicated server.
I turned off vFoglight and the problem still exists.

We had our backups bomb last evening after 33 successful backups...6 backups failed because the VC service restarted.
We get this type of error after:
Validating task Unable to connect to the remote server No connection could be made because the target machine actively refused it 149.158.35.63:443

[Gostev]

It may help to have VMware support research vCenter logs... from the error above it looks like Veeam is unable to connect to vCenter, and without having connection we definitely cannot be the one "bombing" it with API calls.