We have a scheduled downtime for maintenance of storage controller. We have faced issues with ESXi host whenever storage controller is rebooted in the past.
Some hosts faces PSOD, some goes to no responding mode. We want to prevent such issues for this downtime.
Please suggest proper preventive action that needs to be taken before shutting down the storage contoller.
What version of vsphere are you using? And how are you connecting to the storage? FC/iSCSI???
ESXi version 5.0 update 3 and storage connected via FC
If your storage array have redundant storage controller (processor) and your SAN is configured properly, including redundant paths to LUNs, just take one storage controller offline per time and your environment will not be affected.
But if there is no redundancy on SAN environment, you will need shutdown all virtual machines and ESXi hosts before take the storage array offline.
Just curious - Do you support both Storage and VMware components?
Redundancy is key point for such scenarios. If you have redundant paths during the absence of storage controller then you are good with current Maintenance activity. Else shutting down VM's then ESXi servers gracefully is good method.
Redundant HBA Card/Port
FC switch
.
.
Core Switch
Storage controller