Recently some old warhorses have been retired from our site, and a bunch of 
smart young contractors were ferried in 
to replace our old reliable mainframe based ISPF green screen Change Management 
System,  which the O W had written and knew from inside out, with a superduper 
webbased package thingy that has far more facilities.   Surprisingly , it 
works.  

But what they also did was to rip up the insides of the old system and write in 
interfaces to their snazzy new stuff.  The interfaces work too, but the changes 
they have made to accomodate the interfaces do **not** work.  We are getting 
lots of SQL abends. Some of the new stuff uses RRSAF because of the need to 
coordinate rollback between SQL and MQ requires us to call MQCOMMIT (?? or some 
such program), and that seems to cause extra trouble.

I need some explanation of the basics here - 


I understand you can have only one DBthread per logical screen but is that 
strictly true?

Is it one thread per TCB??

Is a new TCB created by SPLIT/START, SELECT CMD(xxxxx), or what?

What is the safest way to coordinate multiple bits inside an ISPF logical 
screen so that they don't interfere??

Sorry, it's 3 am here and I am a bit ill on top of everything else.  Any help, 
pointers, etc gratefully received.

Andy R
 




YCHrtuue

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to