I know it's weird -- a summary without a post.
We've been working with a bunch of fully-loaded Enterprise 450 servers
with a terabyte of disk attached in separate D1000s. Because of this,
I've been working to understand the probe-scsi-all command and the way the
system lays out its SCSI controllers. I've put together an HTML table
describing my findings at:
Using this chart, you can tell what will happen to your beloved disk
volumes if you add a controller and some external disk, so you'll be able
to reset your md.tab or your Veritas volumes *before* everything blows
up...at least, that's the idea. Also, it'll help with planning where to
put the disk to spread out the load across PCI buses and PCI-UPA bridges.
Fun fun fun.
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:14:04 CDT