If you assume that volume mount points change infrequently, except for the
occasional addition or removal of mount points when volumes are created or
destroyed, instead of scanning for mount points in "real time" each time you
want to do a backup, scan once for volume mount points and keep this
information in a file that the backup script refers to for each backup run.
Then, periodically rescan the AFS cell to recreate this file, with the period
being based on your experience of how often this information changes (once a
week, once a month, etc.). To augment this approach, you could make it part of
your administration policy to manually (or through a script) add or remove
entries to this file as you create and delete volumes.
Other things to think about: Since a volume can have multiple mount points,
you really only have to check one of the mount points, becasue all of its mount
points would lead you to the same set of files. So users who create their own
mount points for existing volumes wouldn't be an issue since the existing mount
point for that volume would presumably be in the file created above.
--
Norman Joseph, UNIX Systems Engineer [EMAIL PROTECTED] IC|XC
Concurrent Technologies Corporation 814/269.2633 --+--
NI|KA
ftp://ftp.ctc.com/pub/PGP-keys/joseph.asc
