Unfortunately, I am not watching backups happen as these are occurring during overnight hours. However, I have disabled backups for now and when I re-enabled will back up during the day so I can determine if this backup process is causing the issues or not. I suspect Nessus scanning by our security teams is disrupting traffic and might even be happening as snapshots are occurring which would cause backup issues. We've had Nessus scans kill our ILO interfaces in our chassis plus break the Brocade management interfaces at times too and even cause server VMs to be "sort of" network connected and we would eventually have to reset them to get them fully network connected.
That said, the odd thing is that when I SSH to the appliance or even open the browser to 5480 and connect to the appliance all services appear started and the appliance healthy. If I didn't know better, I would think some sort of websense or FW rule is preventing communication.
As for the backup configuration itself, I have it configured per Veeam's recommendations--using vmtool quiescence, no application processing obviously since it's not windows, and configured as its own job with no other VMs in the backup task. It's vsphere 6u2 with latest build updates on this appliance and my Veeam is now 9.5.
I opened a ticket with vmware and uploaded support bundles and the only thing they found was this: 497 instances of -- execution took too long but not sure what that means--postgres DB took too long, network connectivity took long, etc.? Waiting on a reply from them but not much to go on. O