As I've indicated in previous posts, we began doing DR with DASD mirroring
years before GRS star even existed. We would simply IPL one system
minimally, run a job to create a CFRM policy in the freshly formatted
couple data set, activate the policy, and bring up the rest of the system.
With GRS star, that procedure would no longer work because no system will
IPL at all unless the ISGLOCK structure can be allocated in a defined and
reachable CF.
IBM knew this would put a serious cramp in customers' DR procedures, so
they invented the CFRMPOL parameter in COUPLExx. The sole purpose of this
parameter is to name a policy to activate at IPL if the CFRM couple data
set indicates that no policy is already active. In other words, when IPLing
with a freshly formatted CFRM couple data set. You have the ability to
create a policy and store it in 'another' couple data set by volser, one
belonging to another sysplex or to no active sysplex at all, but you cannot
activate it 'remotely'. When a system is IPLed with that CFRM couple data
set, CFRMPOL tells XCF to activate the named policy (previously stored
there) at IPL time.
BTW anyone with a parallel sysplex should be running GRS star. Our
experience shows that even a single system performs better in star mode
than in ring mode.
.
.
JO.Skip Robinson
Southern California Edison Company
Electric Dragon Team Paddler
SHARE MVS Program Co-Manager
626-302-7535 Office
323-715-0595 Mobile
[EMAIL PROTECTED]
Mark Zelden
<[EMAIL PROTECTED]
CHNA.COM> To
Sent by: IBM [email protected]
Mainframe cc
Discussion List
<[EMAIL PROTECTED] Subject
.EDU> Re: Cleanup a unconnected Coupling
Facility
05/15/2008 08:42
AM
Please respond to
IBM Mainframe
Discussion List
<[EMAIL PROTECTED]
.EDU>
On Thu, 15 May 2008 10:17:09 -0500, Arthur Gutowski <[EMAIL PROTECTED]>
wrote:
>
>During sandbox testing, and implementing our first production system last
>night, I discovered that GRS Star does not IPL when you have a brand-
>spanking new CFRM dataset, even if it has a policy loaded into it, because
>that policy has not yet been STARTED. Because you cannot START the policy
>from the new CFRM dataset prior to IPL, and COMMNDxx is too late to start
>the policy, I found I had to override the IPL
SYSP=(xx,yy,...),GRS=TRYJOIN,
>get the Ring initialized, start the CFRM policy, then SETGRS MODE=STAR. I
>IPLed a second time, just to make sure all was well.
>
>Just some food for thought... If anyone else is in STAR, using the "clean"
>CFRM technique and has an easier way, I'd love to hear from you. Going
>forward, this might not be a concern for us, but the first one was a
little
>rough.
>
That is how we do it for the sysplexes running GRS STAR.
1) IPL with an IEASYSxx that has GRS=TRYJOIN,
2) Load a CFRM policy
3) START CFRM policy
4) SETGRS MODE=STAR.
Mark
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
