Roger,

I have to apologise for my last post. I became so mesmerised by your
attempts to use SNI definitions to deal with your requirement for NJE that I
completely overlooked the obvious solution. Since the problem is you have
two NetIDs why not just make it all one NetID. It is very likely that all
you need do is change the VTAM start option NETID from NETID=ADCD and
NETID=P390 to NETID=ROGER (or whatever) in the two cases. The 087C sense
code may imply you have a working link but, in case you still have problems,
you should report whether or not your adjacent link stations over the
presumably simulated channel connection become active - that's the names on
the relevant PU statements.

So if you have a working link you just need the following to get the NJE
connection working in SNA terms:

The CDRMs, the same in both nodes:

        VBUILD TYPE=CDRM
ADCDSSCP CDRM  CDRDYN=YES,        ** AUTHORIZE DYNAMIC CDRSC DEF.   ** X
               CDRSC=OPT,         ** AUTHORIZE DYNAMIC CDRSC DEF.   ** X
               ELEMENT=1,         ** HOST ELEMENT ADDRESS           ** X
               ISTATUS=ACTIVE,    ** CDRM INITIAL ACTIVATION STATUS ** X
               RECOVERY=YES,      ** CDRM AUTOMATIC RECOVERY        ** X
               SUBAREA=8,         ** NETWORK UNIQUE SUBAREA ADDRESS ** X
               VPACING=63         ** CDRM REQS BEFORE PACING RESP   **
*
P390SSCP CDRM  CDRDYN=YES,        ** AUTHORIZE DYNAMIC CDRSC DEF.   ** X
               CDRSC=OPT,         ** AUTHORIZE DYNAMIC CDRSC DEF.   ** X
               ELEMENT=1,         ** HOST ELEMENT ADDRESS           ** X
               ISTATUS=ACTIVE,    ** CDRM INITIAL ACTIVATION STATUS ** X
               RECOVERY=YES,      ** CDRM AUTOMATIC RECOVERY        ** X
               SUBAREA=6,         ** NETWORK UNIQUE SUBAREA ADDRESS ** X
               VPACING=63         ** CDRM REQS BEFORE PACING RESP   **


Although the CDRM definitions can be the same in both VTAM nodes, one is the
"home" CDRM and one is the "away" CDRM. In fact CDRDYN applies only to the
"home" CDRM and may (an additional option introduced because of the need to
specify the associated function independently of the need for CDRM
statements when VTAM is a pure APPN node), be specified as a start option.
Conversely CDRSC applies only to the "away" CDRM. In order to allow a common
CDRM major node (VTAMLST member) definition VTAM accepts the presence of the
"opposition player" without demur.

Incidentally, taking note of your comments, I'm happy to see that the
systems happen to have different subarea numbers and, I hope, don't have the
same SNA resource names anywhere.

You will not need any CDRSC (or adjacent SSCP table) definitions if you
ensure that DYNASSCP=YES is specified in the start options.

Isn't that so much easier, simple and straightforward compared to what you
were trying before?

Just a small observation about the APPL definitions. You do not need
AUTH=PASS so the PASS suboperand can be thrown away. Also you do not need
MODETAB=ISTINCLM since what this says is you want a specific mode table
ISTINCLM as well as an always present mode table ISTINCLM - clearly making
MODETAB=ISTINCLM not even a default but redundant.

Now I look a bit harder at your APPL definitions - and your JES2
definitions - I see some unnecessary complications - and I wonder where you
got the examples from that caused you to do it this way. Surely you need
only AS06JES2 and A08JES2 as names. You do not need the ACBNAME operands and
only the names A06JES2 and A08JES2 should appear in the appropriate places
in the JES2 definitions.

I used to have NJE definitions in my education test systems but I've quite
forgotten what the definitions looked like having set them up once, got them
working and then forgotten about them (except, as I now seem to remember,
when JES2 development introduced some changes in a new release which
required definition changes - and I thought until then that was a big IBM
no-no). I'm pretty sure I do remember having only the "network" names ( the
names on the APPL statements) and no ACBNAMEs for the JES2 APPL statements.

Chris Mason

----- Original Message ----- 
From: "Chris Mason" <[EMAIL PROTECTED]>
Newsgroups: bit.listserv.ibm-main
To: <IBM-MAIN@BAMA.UA.EDU>
Sent: Wednesday, 11 January, 2006 9:09 PM
Subject: Re: z/OS 1.6 JES2 NJE link


> Roger,
>
> It would appear that one of your VTAM systems has NetID P390 and the other
> has NetID ADCD. In order to connect two VTAMs with different NetIDs over a
> channel-to-channel connection, you need to use APPN.
>
> Note: I was going to say "Low Entry Networking" (LEN) first - as well as
> APPN - but then I remembered that LEN is not possible over a channel -
> although you could use it over, say, Ethernet if your systems both emulate
> 3172s, for example. I guess you have a "P390" and maybe a FLEX-ES running
> the ADCD package and I'm afraid I don't know these systems well enough to
> know what facilities they are clever enough to emulate.
>
> It's rather amazing that you have managed to set up so much of a set of
SNI
> definitions. I'm afraid the part you are missing is vital - and somewhat
> complicated if you are not steeped in it - namely, one or two NCPs. I
expect
> you found a redbook for your samples - but didn't read all of it - very
> dangerous :-)
>
> Incidentally, with these subarea SNA definitions, you should first have
set
> up a CDRM/SSCP to CDRM/SSCP session, in other words, P390.P390SSCP needs
to
> be in session with ADCD.ADCDSSCP first. Without this, there is not a
> snowball in - am I allowed such a word in the newsgroup?* - anyhow you
know
> what it is - chance of P390.A06JES2 getting in session with ADCD.A08JES2.
>
> * Once, in an earlier manifestation of the newsgroup function - I used a
> word which refers to drawing in air through pursed lips together with
> "grandmother", "teach" and "eggs". The moderator felt that the expression
> needed to be rephrased in order to avoid offence!
>
> Since you have just the two VTAM systems and all you want to run is NJE,
it
> should be fairly easy. If you need help, let me know.
>
> You should know there is a specifically VTAM newsgroup to which you can
> subscribe via this URL:
>
>
http://www-1.ibm.com/support/docview.wss?rs=2192&context=SSRRLB&dc=D700&uid=swg27005100&loc=en_US&cs=utf-8&lang=en
>
> It will normally get at least very willing VTAM development help.
>
> Chris Mason

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