A way around the problem of creating mostly unnecessary PH logs which require cleanup is to create a wrapper program that:
- takes [bpfile and] progname as argument(s) - sets logname = progname:"_":date():"_":time() - does a SETPTR with "BANNER ":logname to create a unique log file - runs progname - checks @SYSTEM.RETURN.CODE after the program - if successful, EXECUTE "DELETE &PH& ":logname Now &PH& is left with logs for only those runs which failed. Of course, you'll also have to change the paragraph to run the wrapper with appropriate args (you won't need bpfile if you catalog the prog). -Keith ----Original Message---- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Wednesday, October 19, 2005 7:55 AM To: [email protected] Subject: RE: [U2] UV Abort > <quote who="Keith W. Roberts"> >> What did the phantom log say? You might DISPLAY something after the >> RUN cmd to make sure the fault occurs after the program completes. > > I haven't been able to replicate this behavior. Also, the BRIEF > keyword causes the output to &PH& to be suppressed. I may remove that > to see, but without being able to cause the command to fail or see a > pattern to the failure will make catching the error more difficult. > At least there will be a different &PH& entry for each iteration of > the process... > > Thanks for the suggestion. > > KLP > >> >> -Keith >> >> ----Original Message---- >> From: [EMAIL PROTECTED] >> [mailto:[EMAIL PROTECTED] On Behalf Of >> [EMAIL PROTECTED] Sent: Tuesday, October 18, 2005 3:03 PM >> To: [email protected] >> Subject: [U2] UV Abort >> >>> UV 10.1.7 on AIX 5.2.0 (5200-05) >>> >>> We have a cron job with the following line in a script: >>> /u1/uv/bin/uv "phantom BRIEF BO060.CRON" >>> >>> a VOC entry which looks like: >>> PA Daily cron process scheduled at 2pm >>> RUN BP BO060 >>> >>> This process picks up back-orders, thus the BOnnn name. >>> >>> Periodically, when this runs, I get the following lines in my email: >>> Abnormal termination of UniVerse. >>> Fault type is 11. layer type is BASIC run machine. >>> Fault occurred in BASIC program BO060 at address 0. >>> /u1/atsphx/bo060[3]: 634918 Segmentaion fault (coredump) >>> >>> A cursory _strings_ look at the core file shows nothing other than >>> the normal shell environments stuff. >>> >>> Address 0 in a VLIST is: >>> crtcrlf "Running BO060" >>> >>> I commented that line out and recompiled to see if running a >>> phantom BRIEF balks (sometimes) at a CRT line. I know it shouldn't, >>> but >>> hey, I'm gonna >>> try and see. Since it's rare that it happens, I'm not sure if I'll >>> even know if that fixed it. >>> >>> Any ideas? Suggestions? Oh, the program DID NOT ACTUALLY ABORT. It >>> completed and printed the reports correctly. HUH? Yes, I ran >>> it again and >>> confirmed that the reports I generated were identical to the >>> ones printed >>> at 2pm via cron job. >>> >>> -- >>> karl ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
