I would appreciate it if anyone could confirm, or even suggest a fix for,
a problem we have with S10 x86 servers running MPXIO to connect to multiple
6140 units. In particular, if there is more than one path from a server to
each 6140 controller then a reboot of the server will cause some of the
volumes on the 6140 to fail-over to their non-preferred controller. However,
if there is just a single path per host/controller pair then there is no
problem. By more than one path I mean either multiple initiators on the same
server being able to see the same target port (out of the 4 available) on a
single 6140 controller, or a single initiator able to see more than 1 of the
4 physical target ports on a 6140 controller. We are controlling which HBA
initiators can see which controller port targets through the use of soft
zoning on a set of Qlogic FC switches.

While the system heals itself with all of the 6140 volumes eventually
reverting back to their preferred controller this is a particularly
annoying problem since there is another bug in Solaris which causes
some of our servers to kernel panic under certain I/O loads when this
unnecessary "flapping" happens.

Has anyone else seen this and/or found a solution?

P.S. At first glance this seems to rather defeat the M in MPXIO, so hopefully
we have just missed a magic setting somewhere.

Thanks.


-- 
Stuart Anderson  [EMAIL PROTECTED]
http://www.ligo.caltech.edu/~anderson
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Reply via email to