Chronic ISCSI / storage failover / failback issue

[amf]

I am assigned to work on this chronic ISCSI/storage failover/failback  issue.

So here is the issue:

Everytime there is storage failover, the host loses the lun connectivity and there is timeout. Servers needs to be either rebooted or the luns needs to be rescanned in order to connect back. Any thoughts/suggesstions/soulutions would be appreciated. 

Thanks,

amf

[SeanHatfield]

The SAN host(s) may require remediation.

First validate the config against the IMT, then perform any required updates.

Likely causes:

-Lack of MPIO, or misconfigured MPIO (if cDOT)

-misconfigured CFO (if 7mode)

-misconfigured network switch (spanning-tree/portfast)

-Missing HUK.

-Unsupported SAN host

[paulstringfellow]

Out of interest what OS are the hosts running, if they're esx hosts I would also pursue vmware, this reconnection issue is a general esx problem.

[SeanHatfield]

Thats a good point.  For VMware hosts the recommended settings should be applied with the VSC, some of which require a host reboot to take effect.  Some guest instances may need timeouts set as well.

[amf]

Physical Servers running Windows 2008 R2