I decided to post this question to the mailing list because it needs ZFS
knowledge to be solved.
The situation is like this:
I have a blade server that boots from a LUN, which has no additional storage
or internal disk but that LUN used to boot.
MPxIO works perfectly; but the management wants to use EMC PowerPath,
because the company already has an investment on licensing.
After disabling MPxIO, rebooting and installing PowerPath 5.3 for Solaris 10
x86; the root pool is using only one of the paths to the array instead of
the PowerPath pseudo device name, as it should:
pool: rpool
state: ONLINE
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
c3t5***d0s0 ONLINE 0 0 0
errors: No known data errors
Before you say that that doesn't matter because PowerPath provides
redundancy no matter which path is used, let me say that this is not true
while booting and additionally, I tried off-lining that path and I almost
had a cardiac arrest when the rpool became corrupted. Fortunately, it just
affected a couple of log files that no one will miss. :)
The question then is, how can I make the rpool to use the emcpower0a device
instead of the c3t5***d0s0 Solaris native device?
I filed a case with EMC and Oracle, but the first one lacks ZFS knowledge
and the second one says that it's an EMC issue.
I can boot the server from a DVD-ROM to do some changes without the pool's
datasets being mounted, but I can't assign another LUN to mirror using the
correct device name.
What do you suggest? Changing the device path will work?
Thanks in advance,
Leandro.
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
