Hyper-V 2012 Backup Problems

[cevas2001]

I have been a customer of Veeam for a few years now using the product to backup VMWare guests.
We have recently begun migration to a Hyper-V 2012 cluster and have started a 30 day trial of V6.5 with the latest patch to backup this infrastructure. I am having two major problems:

1) I get the following logged on several VMs every backup:

23/02/2013 14:42:52 :: Failed to flush change tracking data before snapshot. Using full scan incremental run.

This makes the backups slow but is not a show stopper.

2) The show stopper. During backup any VM guest that is restarted hangs on (starting 5%..) until we go over to Veeam and cancel the currently running job. Suddenly the guest begins to start up as normal.
Also during backup some of the clustered guests hang and the cluster fails to restart them. When the backup job completes they start up without problem (it doesn't happen to the guest being backed up)

We are using CSV volumes and have a few guests using Hyper-Vs built in replication to replicate to another host.

I have re-installed the Veeam components on the Host servers and have recreated the backup jobs without success.

Any ideas?

[foggy]

Chris, we have an existing topic regarding your first issue, please review. It is recommended to install the latest patch, could help (depending on the cause of the issue).

The second problem could be caused by issues with the redirected I/O mode on the cluster in case you are using software VSS provider. I would suggest to contact support with that as logs investigation is definitely required for troubleshooting.

Thanks.

[cevas2001]

Thanks for your reply. As stated in the original query I am running the latest patch and show a build version of 6.5.0.128. It still occurs on random VMs.

With regards to the second issue I will contact support next time this occurs. I have made some changes to the backup job including enabling throttling and splitting the large job in to several small ones and it hasn't occurred in the last round of backups.

Thanks

[cevas2001]

Did some further testing and this is what occurred tonight.

New Veeam backup job created to backup a VM on its only CSV volume.
Current owner of this CSV volume was HYPERV2 host.
10 minutes in to the backup a different VM fails on HYPERV2 (different CSV volume) and hangs on starting at 5% when the cluster tries to restart the failed resource.
I cancel the job and the VM continues to start within a minutes of the job stopping.

Event log on host shows the VM as a failed resource followed by : 'Virtual Machine DC3' failed to shutdown the virtual machine during the resource termination. The virtual machine will be forcefully stopped.

Any more suggestions? We have some Hyper-V rather than Veeam replicas do these cause an issue with Veeam?

[Gostev]

Please include support case IDs for both issues here, as requested when you click New Topic.
#2 could be caused by the known Microsoft issue covered in today's weekly forum digest.

[vota]

i had the same issue with "cannot flush CBT"

i opened a support case. The L2 told me to reset the execute 2 commands in cmd on all my hyper-v hosts:
sc stop veeamfct
sc start veeamfct

this solved the problem for me. Hopefully it can do it for you.

BR
Stefan

*EDIT*: Casenumber: 00185066