Our network team is planning to reconfigure spanning tree between our Datacentres and have suggested there will be "up to 90s" of converge time which will affect all VLANs on the network.
A portion of our vSphere Datastores are NFS based on Data ONTAP 8.1.x FAS3270 appliances. We have dedicated heartbeat connectivity and this will not be affected.
We are planning to migrate critical system to FC Datastores but trying to understand what could happen to the VM's residing on NFS.
As the management interfaces will also be affected by this change, what do we need to do to prevent a panic and downtime of the Filers. If possible, we want to keep the Filers online to server the FC based VM's and minimise the need to shut down VM's.
Are there any timeouts we can, or should, change? Should we disable automatic takeover, etc?
If you have been through an exercise like this before, what recommendations would you make?