Re: Automating Disaster Recovery with FCOPY

New Message Reply Date view Thread view Subject view Author view Other groups

Subject: Re: Automating Disaster Recovery with FCOPY
From: David Stuart (David.Stuart@ventura.org)
Date: Mon Jul 21 2008 - 18:47:39 EDT


Tom, 

A couple ideas.  

See the System Control Statements manual, and especially the System Utilities guide.  You can build a customization table for Fastcopy that will get backed up to your backup automatically when specified on your Libr Backup Restore=SA (custtable=l.s.m.t).  My understanding (it's been a while) is that the customization table will be 'invoked' automatically during the SA IPL.  You could use the customization table to perform/automate all your Fcopy restores.  That would cut out some of the keying time and keying errors.  


What I did here is create a set of IPL and JCL proc's for a 2-pack mini-system.  All that is required is an Fcopy Restore of DOSRES and SYSWK1.  At that point, I can IPL, and I have CICS/TS and ICCF available.  My disk initialization job and the Define/Restore jobs for all my critical 'System level' utilities live in a dedicated ICCF library.  Plus, it allows me to run several things in parallel, once the disk init has finished.  


Dave 



Dave Stuart
Prin. Info. Systems Support Analyst
County of Ventura, CA
805-662-6731
David.Stuart@ventura.org

>>> "Tom Duerbusch" <DuerbuschT@stlouiscity.com> 7/21/2008 2:56 PM >>>
I know I've asked similar questions like this before, and didn't get "good" answers.  Perhaps I can say it in different terms...

(I just came back from another disaster recovery test....)

All of our backups for disaster recovery use VSE Fast Copy.  Good enough.

Nowadays, when you IPL a VSE Standalone tape, we come up with a mini VSE system.  That is, ICKDSF, Fast Copy, and Librarian are all selectable, in a very basic full screen display.

But watching the "monkey boys" try to key in all the statements necessary (restoring 16 volumes, plus 6 disk Inits), was a very error prone process.  (16 volumes get the minimum system up so the rest of the restores can be done online)

Under a disaster recovery (test or even live), we will always be running under VM (normally this is a native VSE system).  I want to reduce the manual keying as much as possible.  

In the old days, I could stack the commands under CMS and then let the standalone program go.  
But now, it is this full screen process.

Has anyone been able to either stack the commands to the standalone program, get the standalone program to receive its input from a CMS file (or some other type), run a different standalone program that is a command line version?  or something else?

I'm almost ready (well not really) to consider a HALLAPI interface to feed the commands to the console.  (Just shoot me for ever considering that <G>)

I can shave an hour or so off the most boring part of the disaster recovery test, if I get these people out of the way (and let me do it).  Monkey boys, poking at the keyboard and then looking at the screen to see if the letter was displayed.  TAKES FOREVER!

Thanks

Tom Duerbusch
THD Consulting


Law of Cat Acceleration

  A cat will accelerate at a constant rate, until he gets good and
  ready to stop.


New Message Reply Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b25 : Tue Aug 19 2008 - 18:05:08 EDT