Thanks to:
Mariel Feder <unix.support@central.meralco.com.ph>
Manish Doshi <mdoshi@guanine.bchs.uh.edu>
Glenn Satchell<Glenn.Satchell@uniq.com.au>
Glenn gave the answer, not the one I was hoping for, but its still the
answer. He summed it up quite well and his response follows.
Disksuite (and Veritas Volume Manager for that matter) only report an
error on a disk if the underlying hardware returns an error for a
previous disk access command such as a read or write. When a disk in
the RAID5 stripe returns an error, Disksuite catches that error and
switches in the hot spare. Normally the hot spare disk is idle, ie it's
not being accessed by anything. Thus you'll need to read and/or write
to the hot disk to ensure that it is ok. A failure on one of your
active disks only shows up when you access that particular block on the
disk.
You may want to write a script which periodically tries to read or
write to the hot spare disk to ensure that it is ok.
ORIGINAL question:
>I am running on a Sparc 20, Solaris 2.5.1 and DiskSuite 4.1. I have
>configured a RAID array of 7 disks and 1 hot spare. Everything is> >working fine right now.
>The question I have is how will I know if I have a problem with the hot
>spare disk? Is there a utility to query the disk without invoking
>metatool and rescanning the configuration from the file pull down menu?
>I have simmulated a disk failure and neither metatool nor metahs
>reports the failure until I rescan the configuration. I do not want to
>have the potential of having the hot spare down when I need it. Any
>help will be appreciated. I will summarize.
-- Jon Levin Information Handling Services Unix Systems Administrator Inverness Business Park Phone: (303) 858-6058 15 Inverness Way East email:jon.levin@ihs.com Englewood, CO 80112
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:04 CDT