On Sat, 2 Jun 2007 14:25:10 +1000, Shane <[EMAIL PROTECTED]> wrote: >... >The reason is because they use a single byte id - this is what the >tracker apparently looks for. >Using console-id zero is fine - certainly works up to 1.7. I'll (also) >find out when I get hold of 1.8. >...
Well, that's part of my confusion. I have trouble telling whether console 0 is 1 byte of zeros or 4 bytes of zeros. I don't think you could give a console number to MGCR - it uses "0". (1 byte or 4 bytes? I don't know.) Maybe MGCR (SVC 34) isn't supported under 1.8, but I suspect that would be more clearly documented. We've got 1.8 running on several (non-production) LPARs and have had not trouble so I'm probably worrying about nothing, but I know we've got some home-grown programs that issue MVS commands and I'm afraid we're going to run into some problem with this. BTW, MGCR is documented as supporting only START and REPLY commands (and has been so documented for at least 30 years - probably longer). And I know that 30 years ago I used it to issue a MODIFY, in spite of the documentation. But, of course, MGCRE didn't exist then. Does the documentation really mean it now? Can I assume that any home-grown programs issuing MODIFYs or VARYs is using MGCRE? Pat O'Keefe ---------------------------------------------------------------------- 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
