Like many companies, we use a z/OS guest at IBM Dallas for our testing. We also 
cross-compile our source on Windows, using the Dignus compiler. Obviously, this 
results in lots of FTPing bits from our systems to Dallas. Moreover, we do tons 
of automated testing submitting jobs and gathering output directly to and from 
the JES queue with FILETYPE=JES.

Usually this all works fine. Some days, however, are a disaster, with FTP 
hanging constantly, both during "traditional" file transfers and when 
communicating with the SPOOL via FILETYPE=JES.

We've been unable to identify any commonality to the bad days. At various times 
we've suspected shortage of SPOOL space; something latent in not having IPLed 
recently (we do wind up IPLing fairly frequently, due to running out of 
reusable address spaces after stopping and restarting our cross-memory STC 
repeatedly); shortage of initiators; and other things along those lines, but 
have always been able to disprove those theories.

For example, today, with only two of us around (and thus even more control than 
usual), is a bad day: test jobs keep hanging in FTP. We reIPLed, and SPOOL and 
initiators are fine, but the hangs continue.

We've talked with IBM and our own IT, and nobody seems to have any clue about 
how to diagnose this.

Any ideas? It's not only frustrating, but when we're on a release deathmarch, 
is particularly disruptive to schedules!

Thanks in advance,

...phsiii

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

Reply via email to