Hi,

I wrote a PC/mainframe hybrid based facility that has, as a small part of
it's capabilities, a console command monitor/processor (it monitors all
console commands) and constantly checks the time of the messages (amoung
other things) against the PC's internal clock, (which is kept up to date via
many "occasional" updates of the PC's clock with an external source, when it
notices that the clocks are different, it will/can, depending on a lot of
rules, issue the SET console command to alter it.  This is only a minor
function of the interface, but I have successfully gone through two seasonal
time changes on a z/890 and a z/990 with no problems, (at least none that I
have noticed:).  While this isn't the primary purpose of this facility, it's
been very useful, and I don't see why it can't be accomplished very simply
on it's own with some simpler or a subset of the code. 

There are some issues I ran into with the messages themselves not always
being issued to the interface (or to anywhere)in a timely manner, but I
think I have worked out a good compromise on the "average" times generated
by the messages over the course of an hour, plus I also use another
interface I have to the mainframe's TCP component that gets the mainframe
system time for me quite efficiently so that I can doble check what is being
seen.  I have also worked out a fudge factor of up to 2 seconds, so that it
doesn't end up trying to adjust the clock all day long just to keep things
within 2 seconds.  

Brian

----------------------------------------------------------------------
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

Reply via email to