On 2/4/2016 9:58 PM, Skip Robinson wrote:
I'd like to suggest an alternative recovery strategy...
Instead of standalone restore, plan to recover a system rather than a volume or
two. You don't need to restore sysres. You should have at least two sysres
volumes anyway that you swap between for maintenance. Would you really lose
both of them? IPL from the alternate. What you really need is a single volume
for critical components: master catalog, RACF, spool, PROC/PARMLIBs, whatever
it takes to get a sysprog logged on to run actual recovery jobs. It does not
have to be pretty, just minimally functional for a SME who knows how to put
Humpty Dumpty back together again. That volume is always available and can be
tested periodically to make sure it works. Just a thought.
There is no alternate sysres or existing system of any kind. (If there
was, we would use it.)
What we have is a brand new zBC12 CEC, with an unformatted DS8xxx DASD
box, in a brand new location twelve hundred miles away, with no starter
or bootstrap system of any kind. As Paul Robichaux would say, "We're
starting from dirt!"
Our plan is to send a sysprog carrying a tape to the new location with
SA DSS RESTORE on file #1, a DASD volume with SA DSS RESTORE IPL text as
file #2, and enough other volumes on files #3 through #n to bring up a
simple MVS system that can then restore everything else -- including
z/VM, zLinux, z/VSE, etc.
We need to test the process here to be sure it all works before anyone
flies halfway across the country to do it for real. The problem is, we
can't figure out how to mount the tape. Talk about "dirt!" :-0
--
Edward E Jaffe
Phoenix Software International, Inc
831 Parkview Drive North
El Segundo, CA 90245
http://www.phoenixsoftware.com/
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
