Our Netware 6.5 cluster on IBM (Shark/ESS) SAN is about 3 years old. When
we first set it up IBM recommended we pick a "standard" LUN size and
always use that. We came up with 25GB. Turns out nobody expected users
to gobble up disk as fast as they did (management here pretty much lets
users have what they want) so we were allocating 25GB LUNs left and
right. We have since started using larger LUNs but already have 71 LUNs
allocated to our cluster. All seems to be working fine but I'm wondering
if there is a problem with so many LUNs. We are using QLOGIC Fibre HBA's
and I know the current setting on those limits them to 128 LUNs but I know
that can be increased to 256. My concern is how well NetWare and NCS
handles large numbers of LUNs and is there a "breaking" point. I can't
seem to find any numbers on this documented. Thanks,