Michael Schmutzok wrote:
Scenario: We recently (March 2005) moved from OS/390 V2R10 to z/OS V1R4. We
have a StorageTek V2X1 we are using as our primary storage and a newer V2X2 as
a mirror. We want to flip-flop the two boxes so that the newer box is the
primary and the older box is the mirror. Currently, the primary is defined as
being online at IPL and the mirror is defined as being offline at IPL.
Problem: I went into HCD for the first time since we moved to the z/OS system
to change the online status for the two boxes to reflect what it will be when
we break the mirror and set the new box as the primary (i.e. change the current
DASD to offline from online and change the new DASD from offline to online). No
problems doing this * until I tried to create the new production IODF dataset.
I then received this warning:
W CBDA333I EDT 00 of OS configuration S1 does not use tokens for its
esoterics.
I do not recall receiving this message before when we were on the OS/390
system. I checked some old IODF's and they did not have any tokens specified. I
checked the previous message manuals but could not find this message until the
V1R4 version of z/OS. All of this leads me to believe that some hole got
plugged or some restriction is now being checked/enforced.
Question: I understand from the manual that the token is supposed to be a number from 1 to 8999 and that if I specify it for one esoteric, I have to specify it for all other esoterics. The warning message goes on to warn about possible problems in allocating and deleting datasets that are cataloged using esoterics.
Question 1: Am I in big doo-doo if I continue to leave the token field blank as I have before?
Question 2: If so, what is my rationale/logic for determining what to put in as values for each of the esoteric names?
It was discussed on this forum recently (maybe 3 months ago).
The message comes from new version of HCD, it is not realted to the
hardware or OS level.
My advice: Just assign some unique numbers and forget about it.
A1: IMHO no, but you'll still be attacked by these warnings.
A2: As far as I vaguely remember it is related to very specific catalog
issue: you can *manually* (DEF NONVSAM) catalog a dataset with esoteric
unitname instead of "typical" generic unitname.
(Hint: esoteric: SYSDA, generic: 3390).
After that you could change you IODF definition and old name "MYDISK"
would niw mean tape drive...
HTH
--
Radoslaw Skorupka
Lodz, Poland
----------------------------------------------------------------------
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
