The case of failed Exchange & SQL backups when using VMware vSphere 6 and Veeam Backup & Replication 8

By: Netanel Ben-Shushan & Yaniv Totiashvili

 In my case, I tried to take Exchange Server 2013 backups using Veeam Backup & Replication v8, or Veeam B&R 8 for short.

The Exchange Server 2013 is a virtual machine (VM) that’s running on top of VMware vSphere 6 environment.

When I tried to take backup of Exchange Server using Veeam B&R 8 I notice that the backup takes long time and at the end I get the following error at the Veeam-level:

CBT (stands for Change Block Tracking) data is invalid, failing over to legacy incremental backup. No action is required, next job run should start using CBT again. If CBT data remains invalid, follow KB1113 to perform CBT reset. Usual cause is power loss.”

At the Exchange Server-level I notice that some of the databases was at dismount state with no ability to mount them back correctly (some of the databases’ transaction logs were corrupted after the backup completes), and I have to restore the databases manually from old valid backup or to use “Eseutil /P” to perform hard-recovery.

I contacted VMware support and they told me that I should contact Veeam support (The ping-pong method, LOL). I do so, and Veeam ask to me to wait for update 2 to be released. When Veeam B&R 8 Update 2 was released (download here), I installed it and try again to back up the Exchange with no luck.

During this time I’ve investigated in-depth this issue and found some issues regarding CBT at the VMware ESXi-level. I contacted VMware again and they released the following update that after installing it and performing Veeam back up again to the Exchange server make me sleep quietly.

More information about this issue can be found at VMware KB#2114076.

Leave a Reply

Your email address will not be published.

Reload Image

This site uses Akismet to reduce spam. Learn how your comment data is processed.

WhatsApp Logo IT World