Back in olden days there was a RDINUM JES2 parameter specifying max
number of Internal Readers, and I believe it was a fairly low value
(less than 5 as I recall on our initial MVS/XA system in the 1980's),
either the default or recommended value at the time. I remember having
to increase that value over the years, and I think the default was
eventually increased as well. That RDINUM parameter seems to have
since disappeared with no current equivalent.
That suggests that the original reasons for putting a fixed limit on the
number of concurrent internal readers are no longer valid. and that
whatever internal JES2 resources an internal reader now requires may be
dynamically allocated rather than coming from some fixed pool.
The timing issues with single vs multiple C/Is are un-related to
internal readers, just with concurrent paths to the JES2 input
queues. Would potentially be an issue under VM where z/OS thinks it
has multiple physical card readers as well.
Joel C. Ewing
On 3/4/22 10:06, Paul Gilmartin wrote:
On Fri, 4 Mar 2022 08:29:44 -0600, Joel C. Ewing wrote:
Note also that it has been mentioned before in this forum in other
contexts: submitting two jobs with the same job name in sequence and
expecting "DELAY" to force them to run in the sequence submitted is only
valid if you have a single JES2 Converter//I/nterpreter running, which I
suspect is no longer true at most installations.
Does each SYSOUT(,INTRDR) get a separate C/I? I was told at one
time to use INTRDR sparingly; it used a scarce resource. But, later,
that such a constraint had been relaxed.
--
Joel C. Ewing
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
