No there's no way to avoid the update restart when upgrading firmware.
You also asked about getting to firmware 10.0.2, so I would suggest you upgrade to that level first then do the disk firmware upgrade afterwards.
Re: Downtime If you have configured your servers to the best practices and during a low IO period there is no downtime. Especially with firmware updates since they are planned restarts then are shorter than unplanned ones caused by hardware failures.
The firmware downloads for each revision has a copy of the OS considerations guide to help you set timeouts in the OS to wait longer before generating timeout errors while a failover is in process. It's important to do this not only for planned firmware upgrades, but also for unexpected failovers.
If you are running VMware, this document has how to set those timeouts.
All VMs would also need the disk timeouts set as well.
Thanks for replying. We have 2 SAN’s, 1 for live and 1 for DR (Hyper-V replication is between them). Both are running on PS series V8.1.0 and in both drives firmware got outdated. Our plan is to update DR SAN first and in later days we will update live SAN. I have few queries in this process
I have these doubts because basically I work on databases but now trying to learn and work on SAN.
re: 1. That is fine, it is not uncommon for customers to upgrade their DR first. Make sure that there are no active replications in progress then pause replication when you start the upgrade process. Replication in 8.x is compatible all the way up 10.x. This information is in the firmware release notes.
re: 2. Since you are jumping major revisions you won't be able to downgrade firmware w/o losing all the data on the array.
Drive FW != Array FW
For updating the Drive FW there is no restart needed.
For sure you should updating Array FW because 8.1 is very old. Take look to SANHQ aswell as for ASM for your Hyper-V Hosts.
Correct, but that's not the issue here. They must first update firmware on the array before they can do the drive firmware.
I heard that while doing PS firmware update, Iscsi initiator time outs (On the Hosts) should be set to 60 seconds to allow the controller to reconnect the volumes when restarting (Failing over).
Can anyone let me know where at how it should be changed.
The changing of the disk timeout is not solely for when doing a firmware upgrade. It's also to allow servers to properly wait during a controller failover due to a hardware issue.
With the firmware downloads, is a document "iSCSI Initiator and OS Considerations for HA"
That the instructions on how to set those timeouts. On servers with the HIT/ME software configured the timeout already.
If you make the change, the server would have to be restarted.
That is very true. DSUM makes it much easier.
However, it does change the need to change the DiskTimeOut value. Just to be clear.