vSphere - Handling Storage Failure?

[Paul Hutchings]

Those of you with vSphere environments, when despite your best efforts a host is running but has lost access to P4000 volumes, how do you deal with forcing HA to evacuate the VM's to another host that can see the storage?

[Alzhy]

Well I would think if you rhave HA licensed and DRS enabled -- that it should do so.

[Paul Hutchings]

We don't have DRS only HA (Standard Edition).

Plus the fact the storage has simply "disappeared" from both the host and guest doesn't seem to help.

Closest I seem able to get is an alarm that reboots the vSphere host - like I said, be interested if and how the rest of you address this possibility.

[Bryan McMullan]

I've never actually run into the problem and I've been running SAN/iQ w/ VMWare ESX (3.0-3.5-4.0) for a few years. The only time I've ever seen a problem was when I didn't set the server to use load balancing on the connection. Only other thing I've seen is that if the storage goes down, the VM's basically pause until the storage is available again (like when the VIP goes down and transfers to another node).

[Paul Hutchings]

I rebooted the server and now I see 3 connections in the CMC.

One "normal" and 2 additional which state "DSM" after the host/IP details.

I did reboot after I initially installed the DSM pack, the only thing I've just done is to reinstall it so I'll keep an eye on its behaviour.