>The dumps take too long to capture and process and we're getting to the >point where they can't be captured at all. >We're going to need a more intelligent approach to first failure data >capture or we're going to be in a world of hurt.
I am more concerned with a) the brave new worlds (aka all OMVS stuff): We were given the instruction from IBM to 'dump all jobs with name xyz*' and we soon found out that that is impossible, because there are about 65 of those jobs and the dump command will 'only' dump 15 address spaces. Of course, in the dump we finally sent, the once crucial thing with a number at the end that causes the hang isn't dumped. (And there was no way from the outside to determine what that address space was.) b) the way all OMVS level1/2/3 is handling problem determination: If it's not reproducible at will, too bad for the customer! And even if you can reproduce the problem, let's go back and ask for more and more traces. c) the disappearing skill to read a dump *at all*! It is not a good thing for IBM if the customer can do it better than the 'experts' in level2! Not being a fan of UNIX, can someone enlighten me where the advantage is in running IPCS in a shell (presumably doing command in batch) as opposed to a batch job running those same commands with print noterm? Regards, Barbara Nitz -- Psssst! Schon vom neuen GMX MultiMessenger gehört? Der kanns mit allen: http://www.gmx.net/de/go/multimessenger ---------------------------------------------------------------------- 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
