The definition of "AVAILABLE" file for VSAM is a little odd (obscure? strange?)
Check out: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/igy3pg30/1.10.6 A VSAM file *must* be "available" to OPEN I-O. See also the information at: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/igy3pg30/1.10.3.2 .1 HOWEVER, (after finding all of that), I found what I think is MOST relevant in the latest Enterprise COBOL V3R4 manual (with revision bars indicating it is new). From: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/igy3pg30/1.10.3.2 .1 " When you are loading an extended-format VSAM data set, file status 30 will occur on the OPEN if DFSMS system-managed buffering sets the buffering to local shared resources (LSR). To successfully load the VSAM data set in this case, specify ACCBIAS=USER in the DD AMP parameter for the VSAM data set to bypass system-managed buffering." "Greg Shirey" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... > Hello all, > > We modified our ACS routines so that our production VSAM files will be > defined as extended format VSAM, after months of testing with "test" VSAM > files and having no issues. > > Early this morning, a production program that is in a job that runs every > day issued an OPEN I-O on a data set and got a status code of 30, which led > to a the program abend. A couple of steps earlier in the job, the VSAM file > is deleted and defined, but nothing is REPRO'd into it. The programmers are > saying that the job and the program have not changed, implying that they > have been able to open for I-O an empty VSAM file, but are now unable to > because of the change to extended format VSAM. (They got the job to run by > "seeding" the empty VSAM file before running the program that opens it.) > > We've been searching for some definitive information on whether this is > expected behavior for extended format VSAM, but are not having much luck > searching the bookshelves. I thought I might turn to the collected wisdom > of the list for feedback. > > We were also surprised by the return code of 30 - Permanent I/O Error? ---------------------------------------------------------------------- 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
