I have a two node cluster running Veritas HA 5.1 with a 1TB shared LUN attached.
The only way I could initialize the LUN device under VxVM was to enable MPXIO
and label the device as EFI format.
Initially Solaris was seeing four separate paths to the LUN until enabling
MPXIO.
Although Solaris and Veritas VM was seeing the SAN device it could not be
initialized under VxVM.
Solaris format command showed 4 separate paths on two separate controllers for
the same LUN besides the IBM SDD driver showing the one vpath.
2. c2t500507680110527Cd0 <IBM-2145-0000-1.00TB>
/p...@0/p...@0/p...@8/p...@0/p...@8/SUNW,q...@0/f...@0,0/s...@w500507680110527c,0
3. c2t5005076801105277d0 <IBM-2145-0000-1.00TB>
/p...@0/p...@0/p...@8/p...@0/p...@8/SUNW,q...@0/f...@0,0/s...@w5005076801105277,0
4. c4t500507680140527Cd0 <IBM-2145-0000-1.00TB>
/p...@0/p...@0/p...@8/p...@0/p...@a/SUNW,q...@0/f...@0,0/s...@w500507680140527c,0
5. c4t5005076801405277d0 <IBM-2145-0000-1.00TB>
/p...@0/p...@0/p...@8/p...@0/p...@a/SUNW,q...@0/f...@0,0/s...@w5005076801405277,0
6. vpath17a <IBM-2145-0000-1.00TB>
/pseudo/vpat...@17:17
Although the LUN could be formatted, etc. within Solaris, Veritas could not
initialize it due to device path issues and other anomalies.
After enableing MPXIO the single LUN shows as:
2. c6t60050768018E82965000000000000654d0 <IBM-2145-0000-1.00TB>
/scsi_vhci/s...@g60050768018e82965000000000000654
3. vpath17a <IBM-2145-0000-1.00TB>
/pseudo/vpat...@17:17
At that point Veritas can initialize the LUN.
san_vc0_0 auto:sliced dg101 dg1 online
Are there problems with this configuration?
Will it affect HA failover?
Should I disable DMP?
Thanks
--
This message posted from opensolaris.org
_______________________________________________
storage-discuss mailing list
storage-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/storage-discuss
