Anuradha

Well, I'm happy to see that you now can now "connect" to ISTMNPS.

What you originally described appeared as if it could be an invalid
environment as far as VTAM was concerned and so this "End Node" requirement,
very firmly stated in the "VTAM V4R4 for MVS/ESA Implementation Guide" (June
1997) redbook, seemed most likely. This redbook was the only one I
remembered which addressed this topic. However, the "SNA in a Parallel
Sysplex Environment" (December 1998) which you have been using is eighteen
months later and has the most recent changes at that time. Among the changes
in the meantime had been the relaxation to allow the use of a VTAM Network
Node. Interestingly enough - I've been refreshing my memory of how MNPS
works - the change exposes the customer to the possibility that sessions are
launched into an invalid configuration which was prevented when the End Node
restriction existed.

Turning to this post, it seems that the problem was explained when VTAM
started but I expect you just hadn't seen it. It would appear that when you
try the
V NET,CFS,ACTION=CONNECT,STRNAME=ISTMNPS command, VTAM doesn't respond
because it somehow has recorded that it has already tried and failed and so
isn't going to try again. VTAM perhaps should issue a message to remind you
that it has already tried and failed with an unrecoverable error.

I'm sure you looked up the explanation of the IXL013I and IXL015I messages
and then found your way to the information required to set the correct
structure size. I'm surprised that your system wasn't built with an
appropriate size as default. I'm glad you managed to work that out for
yourself because I have no experience actually fiddling with these coupling
facility structures.

Incidentally I notice you have - perhaps in desperation - at some time
issued MODIFY NET,MSGMOD=YES so that you get the 5-character module
identification in your VTAM (IST) messages.

Finally to answer your question, the reason you had to re-IPL in order to
have the changed sizes come into effect is almost certainly because this is
a major change to the coupling facility structures.

I glanced through the MNPS chapter in the "SNA in a Parallel Sysplex
Environment" redbook but I cannot find any text which suggests that this
sort of a change would not need a re-IPL. The command you mention is only to
cause the "connect" to take place. I don't know all that much about coupling
facility structures but it seems logical that a change involving the size of
the structure would need, firstly, that the change is made in all systems in
the parallel sysplex which are going to connect to the structure, secondly,
all "disconnect" and, thirdly, all "connect" (again). This is because the
structure is built when the first "connect" is made. Perhaps the "Parallel
Sysplex Operational Scenarios" redbook has more to say on this subject
somewhere.

Chris Mason

----- Original Message ----- 
From: "Anuradha V" <[EMAIL PROTECTED]>
Newsgroups: bit.listserv.ibm-main
To: <IBM-MAIN@BAMA.UA.EDU>
Sent: Friday, 01 September, 2006 3:04 AM
Subject: Re: VTAM not connecting to ISTMNPS - VTAM recycle required ?


> Hi All,
>
> Is VTAM recycle required when the ISTMNPS structure is changed?
>
> I understand from the Redbooks that its is not.
> VTAM is supposed to automatically detect the changes.
> If needed we could do 'V NET,CFS,ACTION=CONNECT,STRNAME=ISTMNPS'.
>
> While VTAM started for the first time it had the following error:
> IXL013I IXLCONN REQUEST FOR STRUCTURE ISTMNPS FAILED.
> IST1365I FSNCN CONNECTION ATTEMPT TO STRUCTURE ISTMNPS FAILED
> IST1366I FSNCN MVS MACRO IXLCONN FAILED - RTN CODE= 0C - REAS
> IXLCONN RETURN CODE: 0000000C,  REASON CODE: 02010C08
>         CONADIAG0:   00000002
>         CONADIAG1:   00000008
>         CONADIAG2:   00000C08
> IXL015I STRUCTURE ALLOCATION INFORMATION FOR
> STRUCTURE ISTMNPS, CONNECTOR NAME NETB_CDRMDOS
> CFNAME     ALLOCATION STATUS/FAILURE REASON
> --------   ---------------------------------
> ICF11      INVALID STRUCTURE SIZE
> ICF21      INVALID STRUCTURE SIZE
>
> I then increased the INITSIZE and SIZE.
>
> In my environment a VTAM recycle(IPL) alone made VTAM aware of the ISTMNPS
> structure change.
>
> Regards,
> Anu

----------------------------------------------------------------------
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

Reply via email to