On Wed, 2004-07-14 at 13:32, James Melin wrote:
> Well in a perfect world, I'd be able to do file level backup of a z/linux
> guest from z/os and write that backup to a tape managed by RMM in the 3494.
> This is an imperfect world, however.
>
> It is important that whatever backup method I employ be able to be tracked
> by RMM, because in a disaster recovery, we would be using z/os to restore
> z/VM and z/linux. Currently I have been doing Compatible Disk Layout
> backups of the volumes used by z/linux, but some of those volumes are
> logically carved into minidisks. (Such as 1200 cyl /tmp directories).
>
> So backing up a mod-9 with 5 1200 cyl minidisks and one 4017 cyl minidisk
> on it, is the granularity level of the mod-9.
>
> I was looking at the possibility of something from the VM side of things
> that can simply backup a minidisk so that the restore can be put anywhere,
> at minidisk granularity but still catalog stuff in the z/os RMM database.
> Probably a kludge at best, impossible at worst.

You could use CMSDDR to back up minidisks as files onto a volume that
then was backed up.  That'd be pretty easy to script.

...and I was going to suggest:

> Ultimately, I would love to be able to ship file level backup images via
> amanda to z/o to be migrated via HSM, but that is deep weeds rocket science
> to me. My understanding of such a theorettical scenario is that amanda
> would manage the archive contents and HSM/RMM manage the archive dataset
> and the tape that it is on.

Your understanding is correct.  From Amanda's viewpoint, its backups
would always be on a tape-faked-by-a-file (and presented via NFS).  That
file, in turn, might or might not be on DASD, depending on HSM migration
policy.

Adam

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to