i'm experiencing a difficult situation, where a super critical VM is totally unprotected. To start protecting it we need to perform configuration changes,
but before any change a backup is recommended

Scenario: Windows 2003 virtual machine (on ESXi 5.5), Exchange 2007 running on it. Legacy backup solution (Backup Exec 2012)
no longer working since months: the mail server is totally unprotected at the moment.
The VM was using RDMs in Physical Compatibility mode, but we have converted the RMDs in Virtual Compatibility mode.
We want to start protecting with Veeam, but we have concerns regarding the configuration of the machine and backup job,
and we want to take the less risk possibile.
Veeam is at 9.5update1, the integration with Netapp storage system works fine.
VM size is reported to be 2.7 TB, the job contains only this VM.
I've enabled application aware processing (test successful).
The job must use only storage integration (faillover to standard backup unchecked).
The NetApp volumes should have enough free space: the one holding log files is at 67% of used space.
I estimate 7-12 hours for the first full backup.
Is this configuration good and we can try to run the backup safely, or not?
Would it be better to convert disks from Virtual RDMs to VMDK before to proceed?
I've also opened a case with support (severity 4, id # 02072446) to upload configuration files and details.
Easy to guess, i have lot of concern with this issue, and i have fear that the snapshot operations could stuck the server driving i.e. to database unmount,
even if we are using Storage Integration, and the vmware snapshot will live just a few minutes. Without Storage Integration i would not even try it.
An comments and suggestions are appreciated. I'm feeling alone on a wrecked ship

thank you so much