Aloha,
We have a troubling situation we don't understand. On each of our LUNs and all of our Windows guests we have MP set to round robin mode.
We are upgrading our RAM on the hosts. On each of the hosts that we have upgraded thus far, the MP mode has been reset to fixed on both
the hosts and the guests after the reboot. Has anyone else seen this??? Major humbug....
Mahalo,
Bill
This can be related. http://communities.vmware.com/thread/397244?start=0&tstart=0
There's likely a claim rule in place forcing the path selection policy. Are all your data stores backed by the same storage?
Log into one of the hosts using ssh and run the following
esxcli storage nmp satp list
Post what it says, if you see your storage provider you can run the following
esxcli storage nmp satp set -s VMW_SATP_PROVIDER -P VMW_PSP_RR
Then it should work from there on out.