Hello,
The original questions appears at the end of the message.
Thanks to
alan_at_nabeth.cxo.dec.com
Mostafa Mahdy <mostafa_at_ncsbah.com>
Jim Bostwick <Jim_Bostwick_at_cargill.com>
"Ian 'Ivo' Veach" <ivo_at_scs.unr.edu>
"John J. Francini" <francini_at_progress.com>
for their responses. The consensus seems to be that AdvFS does not have
anything quite comparable to the superblock table. It is recommended to
keep file and paper copies the following around:
/etc/fstab
/etc/fdmns/*
disklabel -r [for all disks]
I gather that AdvFS maintains its own metadata about the disks, and using
verify, backups, and salvage (only as a last resort) is the appropriate
path to take in the event of disk problems with an AdvFS file domain.
It was also suggested that copies of /etc/rc.config and /sbin/swapdefaults
be kept for swap disk information, in the event of catastrophic failure
requiring rebuilding to the original configuration.
I hope I haven't garbled anything too badly in summarizing.
-- Bennet
>Date: Fri, 03 Dec 1999 15:10:00 -0800
>To: tru64-unix-managers_at_ornl.gov
>From: Bennet Fauber <bfauber_at_ucdavis.edu>
>Subject: AdvFS Disk Information
>
>Hello,
>
>Someone once told me that it was a Good Idea to print out the list of
superblocks that is generated by newfs so that if the superblock(s) go(es)
bad, the location of the alternate copies is available.
>
>Is there something comparable that is needed for AdvFS filesystems?
>
>I keep file and paper copies of the output of disklabel -r for all disks on
>the system and of /etc/fdmns at the moment. Does AdvFS keep track of
>everything else that it needs, or is there something else that would be
>critical to have around in the event of a partial disk failure?
<----- oOo ----->
Bennet Fauber
Social Science Data Service
University of California, Davis
Received on Mon Dec 06 1999 - 17:57:33 NZDT