If your directories are identical except for the volsers (and release and 
maintenance levels), go for it. Changing the IPL address should be 
sufficient.You have no synchronicity problems.What you will lose, if any exist, 
are the ordinary spool files


Regards,
Richard Schuh





________________________________
From: The IBM z/VM Operating System [mailto:[email protected]] On Behalf 
Of George Henke/NYLIC
Sent: Monday, October 18, 2010 3:11 PM
To: [email protected]
Subject: Re: *Rolling IPL* Style Migration


As long as the CMS users are not on the System Vols:  540RES, 540W01, 540W02, I 
should be ok because the Level 2 directory is identical to the Level 1 
directory except for these VOLSERs.

The same should be true for DB2, DIRMAINT, VMSECURE, etc.

As long as none of these things are on these 3 VOLSERs, it should be 
transparent to the migration.

If they are on these VOLSERs, they should not be for this very reason, and 
should be migrated off.




Michael Coffin <[email protected]>
Sent by: The IBM z/VM Operating System <[email protected]>

10/18/2010 05:16 PM
Please respond to
The IBM z/VM Operating System <[email protected]>


To
[email protected]
cc
Subject
Re: *Rolling IPL* Style Migration





Hi George,

If you are only using z/VM as a hypervisor for Linux guests (or otherwise don't 
have an active CMS user community) this will work.  If you have servers (e.g. 
DB2, SFS, etc.) and/or end-users that are updating CMS files - you have to keep 
in mind that those will get out of sync on your 2nd level system right after 
the spindles are copied.  If you have directory maintenance products (e.g. 
VM:Secure, DIRMAINT,etc.) and/or an ESM (e.g. VM:Secure, RACF, etc.) these 
products will also get out of sync, so find a way to ensure they stay 
synchronized between first and 2nd level systems.

-Mike

From: The IBM z/VM Operating System [mailto:[email protected]] On Behalf 
Of George Henke/NYLIC
Sent: Monday, October 18, 2010 4:53 PM
To: [email protected]
Subject: *Rolling IPL* Style Migration


I have no extra disk, no tape, but still need a fallback, I am thinking of 
implementing the following strategy to put maintenance into production and 
would appreciate it, if someone could find something wrong.

My 2d Level is a mirror of 1st Level except for VOLSERs which are unique.

I apply maintenance to 2d Level.

Maintenance hits 5 SDFs, among them, CMS.

At cutover:  I simply point to the 2d Level RES and IPL it from the HMC.

I bring up what was the 1st Level at 2d Level, after making the appropriate 
VOLSER changes for the 2d Level machine in the 1st Level Directory.

I  then have my prevous 1st Level machine at 2d Level as a fallback, if need be.

When I am ready to install a new version or put on more maintenance,, I DDR the 
1st Level disks to the 2d Level, and bgin the process all over again.

Aside from losing the 1st Level Spool files whenever I do this, I see no other 
problem.

I suppose I could always connect the 2d Level and 1st Level Spools with RSCS so 
I do not have that problem.

I could also use Dave Jones, *handy dandy* SFB utility to copy the Spool files 
to CMS and then  restore them to the new 1st Level system.

But aside from this weakness, I see no other reason why this should not work.

Please, someone, let me know if you do.

Reply via email to