SQL Server Transaction Log Backups to EMC Data Domain

[adweber328]

Hello,
We've been backing up our SQL servers with Transaction Log truncation/backup being done hourly by Veeam.
We want our transaction logs to back up to our EMC Data domain, which replicates off-site. However, we seem to have some issues with Veeam when we try to target our Data Domain. Normal backups run fine to the DD, and are very stable. Transaction log backups fail with this error:

Failed to save SQL Server transaction log backup file for database databasename to repository AgentManager: Failed to create agent session ticket for agent 'f106e9a4-5aa8-4961-9074-5aaf34b60fc1', host: 'veeamproxy.domain.name'.

I opened a ticket with Veeam support, which ran for multiple months of troubleshooting. The eventual answer from support was to remove the backups from the data domain, and target a different kind of storage. This is not what we want to do, because we need transaction logs in the event of a disaster. We could potentially lose a day's worth of data if we restore from the nightly backups. This could be quite damaging to our business.

Does anyone else back up to an EMC Data Domain with transaction logs? Is it working for you?

[Vitaliy S.]

Hi Alan, can you please post your support case ID for a reference?

[adweber328]

Hello, thanks for the response.
The case number was 01732623

[Vitaliy S.]

I see that you've tried to increase the backup interval for this job and it didn't' help. If I were you I would reopen a support case and ask escalation to a higher tier (dev guys) to check if anything else can be done on our side.

[adweber328]

Thank you. I opened up another case (this one Sev 2)
Some production backups started failing recently, so this is a priority now. The new case is 01771258.

I appreciate your responses!

[lowlander]

Hi,

we had similar problem, and also running a support case now. We discovered the same problem when using a different gateway server and log shipping server. Set the gateway server and log shipping server to the same server and log shipping would work as expected. However this is a workaround, actually you don't have any redundancy in gateway server and log shipping server in case of component failure. In our opinion logshipping should be working also in automatic mode instead of "fixed" mode.

[lowlander]

Seems to be a glitch on the gateway / proxy server thinking he was 32-bits instead of 64-bit. problem solved.