Sysudump is empty Sent from my iPhone
On May 30, 2013, at 4:11 PM, DASDBILL2 <[email protected]> wrote: > First, something "bad" happened somewhere and very likely left some forensic > evidence. Then EXCP's recovery routines were invoked and decided to classify > the error as an S800-4 ABEND. Then maybe a dump was produced. I suggest you > look at the System Trace that is in your SYSUDUMP (you do have one, right?), > find the SVC 13 (base 10) that was issued that resulted in the S800-4 error > message and dump, then work backwards in the system trace looking for > previous entries that are (1) from the same address space as yours and (2) > have an "*" in a column that is almost always blank. That "*" indicates that > something most unusual happened as part of that entry. Then try to analyze > that error rather than the S800-4 situation . > > > > Another possible approach is to find the SDWA and/or RTM2WA control blocks in > your SYSUDUMP and look for fields in those control blocks whose DSECT > descriptions say something like "earliest error", "original error", or some > words like that. Those kinds of fields will probably correspond to that > system trace entry with the "*" in it. If you can't find fields like that > quickly in the SDWA or RTM2WA, then look at each field in both control > blocks, read the DSECT description for that field, and think about what the > description says - is that field associated with the original error or with > something that happened much later in the recovery/termination process? > There are hundreds of fields. It takes time and determination. Usually > there is forensic evidence somewhere, just like on television. > > > > Another possible source of early forensic evidence is software records in > SYS1.LOGREC. > > > > Also look for an EXCP Debugging Area, pointed to by the TCB that was ABENDed. > If you can find that, you may have a PSW and registers at the instant that > the original error occurred. You may have to ask for help in finding a DSECT > of the EXCP Debugging Area, however. I used to know where that one is > documented, but I haven't needed to use it in lots of decades, so I don't > remember now. > > Bill Fairchild > Franklin, TN > > > ----- Original Message ----- > From: "Micheal Butz" <[email protected]> > To: [email protected] > Sent: Thursday, May 30, 2013 2:56:39 PM > Subject: Re: System abend 800 reason code 4 > > For reason code 4 the explanation says > A program issued a SVC 114 the EXCPVR macro but a error occurred durning > page-fix or page-unfix processing A page-fix error can occur > If the EXCP processor tries to fix pages > That are not assigned to the callers ASID > > The blksize in the program is 27930 > For this program maybe for dfsms > To get that kind of buffer is problematic > > Ill look at the dcb parms to see what I can do > > Sent from my iPhone > > On May 30, 2013, at 4:34 AM, Binyamin Dissen <[email protected]> > wrote: > >> On Wed, 29 May 2013 22:22:33 -0400 Micheal Butz <[email protected]> >> wrote: >> >> :> Running a program under TSO TEST >> :>I got the above abend after a BSAM READ. Would anyone know what this is >> about >> >> What did you understand from looking up the abend in SYSTEM CODES? >> >> -- >> Binyamin Dissen <[email protected]> >> http://www.dissensoftware.com >> >> Director, Dissen Software, Bar & Grill - Israel >> >> >> Should you use the mailblocks package and expect a response from me, >> you should preauthorize the dissensoftware.com domain. >> >> I very rarely bother responding to challenge/response systems, >> especially those from irresponsible companies. >> >> ---------------------------------------------------------------------- >> For IBM-MAIN subscribe / signoff / archive access instructions, >> send email to [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
