Request: Direct SAN Restore failover to network

[Threonine]

This week we had a customer that attempted to perform a full VM restore on Veeam 8.x Patch 1 and was surprised to see that the restore failed due to the SAN LUNs being presented to the Veeam proxy server in read-only mode -- KB2005 (http://www.veeam.com/kb2005). I was also surprised that this occurred and could not find any way to change the transport mode in the restore wizard. Since this is a very common configuration, I would recommend that the network failover functionality be updated to handle this scenario (the option is enabled in the proxy server settings ("Failover to network mode if primary mode fails, or is unavailable").

The job details looked like this:
Error Rolling back Hard disk 1 (40.0 GB):
Error: VDDK error: 17592452332404352.One of the parameters supplied is invalid
Failed to download disk.
Agent failed to process method {DataTransfer.SyncDisk}.

The only workaround now appears to be to manually change the transport mode before and after a full VM restore.

Thanks!

[Gostev]

Network failover should handle this scenario, so please have the customer open a support case for investigation, and let us know the support case ID. From the log snippet, it does not look like the issue has anything to deal with the read-only LUN mode (and rather with VDDK initialization).

Transport mode is a backup proxy setting, and is set in the proxy properties - as opposed to the restore wizard. Restore wizard does allow selecting the desired proxy to perform the restore through though.

Thanks!

[Threonine]

Thanks for the quick reply! Interesting... Perhaps it is a different issue... This did occur with a VM that was deployed with an OVF and we were suspecting it might be related to the OVF environment settings... The restore was also attempting to overwrite the existing VM.

Case #00859869 will have the detailed logs.

Thanks again.

[Threonine]

After another test in the same environment, it looks like this issue must've been a VMware OVF environment issue instead of a Veeam proxy issue. Another test with a different VM resulted in a successful restore via "nbd". Thanks, and sorry for the confusion.

[sfbi]

Hello!

I've the same issue with Veeam 9.0 and UPD2 (build 9.0.0.1715)

Code: Select all

19/01/2017 16:09:08 Error    Restore job failed Error: VDDK error: 16000 (One of the parameters supplied is invalid). Value: 0x0000000000003e80
                             Failed to write to a virtual disk
                             Failed to download disk.
                             Agent failed to process method {DataTransfer.SyncDisk}.
                             Exception from server: Shared memory connection was closed.
                             Failed to upload disk.

My backup proxy is without the option: "Failover to network mode" and Direct SAN transport work correctly for backup.

Enabling it the restore work correctly even if the network is not used! (tested with network monitor)

Another workaround to this issue/bug is select another datatore for restore.

I hope I was helpful to others

[flun]

Hi everyone

I have a questions about the DSA Feature of Veeam.
Is it possible to have the permissions to the SAN Storage via ISCI only on read and if we would peform a restore could it fallback onto the ESXi connection, where veeam has read/write permissions?

Basically:
Backup = via Storage Access
Restore = via ESXi Host

Thanks for your answers

[Vitaliy S.]

Yes, this scenario is possible, please take a look at Anton's reply above. Thanks!