Patrick, I've been keeping my eye on this thread as the responses flew by hoping a definite answer might emerge. The definitive answer doesn't seem to have appeared so far.
Many years ago I used to teach TCP/IP for MVS. My last version of the presentation in dated November 1995 and I think it corresponds to V3R1. When I was first confronted with installing TCP/IP for MVS I was appalled at the mess of libraries, partitioned data sets, that were thrown at you with no explanation as to what purpose each served - what a can of worms! But I was eventually asked to insert a day teaching the MVS flavour of TCP/IP into an otherwise basics-cum-OS/2 class. Now I was being forced to try to make sense of all these libraries. So with that background I can report that what I said all that time ago about SEZATCP is that it is a library for load modules, it had at that time 40 members and was used by "started tasks". What I very probably meant by "started tasks" is that it seemed to contain the names which followed the PGM= operand of the EXEC statement in the many sample task procedure members which were described in the then "Customization and Administration Guide". Two rather old pennies. Chris Mason ----- Original Message ----- From: "Chase, John" <[EMAIL PROTECTED]> Newsgroups: bit.listserv.ibm-main To: <[email protected]> Sent: Monday, 05 June, 2006 2:03 PM Subject: Re: SEZATCP - Steplib or LNKLST? > > -----Original Message----- > > From: IBM Mainframe Discussion List On Behalf Of McKown, John > > > > > -----Original Message----- > > > From: IBM Mainframe Discussion List On Behalf Of Patrick Lyon > > > > > > [ snip ] > > > > > > Does the TCPIP stack need SEZATCP at all? If so, should SEZATCP be > in > > > the LNKLST? Does everyone else have only SEZALOAD in the LNKLST? > > > > From reading the z/OS 1.6 manual (the TCPIP one, not the CICS > > one), it says that SEZATCP should be on the DFHRPL, not the > > STEPLIB. It also says that this library contains the CICS > > interface code, not "general purpose" code. So I would not > > put it in the LNKLST, in the TCPIP procedure, or in the > > STEPLIB for CICS. It appears to be only needed in the DFHRPL > > concatenation. > > And it is also true that, once up and running, the CICS loader loads > *exclusively* from DFHRPL; it does not "fall thru" to either STEPLIB or > LNKLST. If CICS Program Management doesn't find what it's looking for > in DFHRPL, "it ain't there". > > -jc- ---------------------------------------------------------------------- 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
