If nothing else is in the dataset, a half track block can hold 349 records. Would want space for an EOF record too, so 680 should result in a 1 track dataset. 4080 block size would be 12 blocks so 612. On Tue, Oct 16, 2018 at 11:07 AM Sam Golob <[email protected]> wrote: > > Hi Folks, > > I hope this post isn't considered a waste of everyone's time. I > came across this "design element" (so to speak) in my work concerning > the Broadcast Dataset (aka SYS1.BRODCAST, in its more restrictive sense). > > Question: When you are reformatting the active Broadcast Dataset > (SYS1.BRODCAST) with the SYNC command, you create space for 100 global > Notices (the messages that everybody sees when they LOGON). How do you > change this number? > > Answer: You probably would never want to change this number, > because 100 notices is adequate for most purposes. But what if you > really DO want to change this number? Then IBM tells you to zap a > fullword field at the beginning of TSO initiation module IKJEFXSR, and > re-IPL (maybe with CLPA, just to be safe). Again, you'd probably never > want to do this, because 100 notices is adequate for almost everyone. > > Where is that number (100, or F'64') kept? It is actually a > fullword in the CVT itself, at displacement X'5A8'. It is not "chained > off the CVT". It is actually a fullword IN the CVT...!!!! Wow! > > So how would a fullword in the CVT be changed? Presumably only at > IPL time. That's why IBM's method of changing it is so cumbersome, > since (I think) IKJEFXSR, which starts up TSO, has to put that number > into the CVT. > > Anyway, if someone REALLY wants to change this number (temporarily, > for the duration of the IPL), I wrote an authorized TSO command called > BDMNNOTC, which will take a number as a parameter, convert it to > fullword binary, and zap it into CVT + X'5A8'. This command is on CBT > File 731 (www.cbttape.org). It works instantly. So you say: BDMNNOTC > 50, or BDMNNOTC 150, and if you do a SYNC afterwards, that number of > NOTICES records will be formatted into the (active) Broadcast Dataset. > If you are really doing this, please see the notes in the program > source, on CBT File 731. > > I submitted a request to IBM quite a few years ago, that maybe an > entry should be put into PARMLIB member IKJTSOxx, with the format > NUMNOTICES(nnn), which would change this number whenever you do a > PARMLIB UPDATE(xx) TSO command or a SET IKJTSO=xx console command. > Since this requirement probably has the lowest priority that could ever > be assigned to an enhancement request, I'd guess that it probably would > never get done. But meanwhile, the sysprog community has my BDMNNOTC > solution, if they want it. > > I guess I've taken enough of your time. Hope it's food for some > thinking--and maybe it'll eventually lead to some productive progress > somewhere. > > Thanks for listening. All the best of everything to all of you. > > Sincerely, Sam > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN
-- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
