I've found Veeam can cause problems in a mirrored environment for Cache.
We invested a great deal of energy trying to get to the bottom of this (with ISC) but never truly fathomed it out.
As you can imagine we tried all sorts of freeze thaw scripts and QOS settings. Whilst we did reduce the frequency we could not completely eliminate the problem.
In the end we came up with this strategy.
1. Run Veeam on the backup member nightly
2. Run Veeam on the primary member only once a week during the day
3. Only use Veeam as a way to restore the OS
4. This would exclude all DAT, Journal and WIJ files
5. Take a nightly dat backup of both members and stash several days of Journal files
6. Ensure the mirror pair are on completely different sets of hardware and ideally locations
7. Configure the DAT and Journal files to be on different LUNs on the VM setup
8. Understand how to physically recover these files should the server crash and not recover
Conclusion for me was that Veeam is a nice to have tool in a VM set-up, but it should not be a replacement for the battle tested cache backup solution.