IDMS

We have a COBOL batch-autostatus program running under the CV that appears to be looping on the call to IDMSIN1 within the IDMS-STATUS-LOOP paragraph. The following output is observed:

PROGRAM NAME ------             
ERROR STATUS ------ 0326        
ERROR RECORD ------ DLNTAX      
ERROR SET --------- CALC        
ERROR AREA -------- DLNTAX-AREA
LAST GOOD RECORD --             
LAST GOOD AREA ---- MULT-AREA   
                --- ............

with the last line - which appears to be the 'DISPLAY SSC-STATUS-LABEL '--- ' SSC-STATUS-VALUE' statement near the end of the IDMS-STATUS-LOOP routine - being displayed until the job gets a S722 abend.

The 'LAST DBKEY -------- ', 'LAST DBKEY HEX ---- ', etc. lines that normally appear after the 'LAST GOOD AREA ---- ' line don't appear anywhere in the output, which makes us think that the condition(s) in IDMSIN1 that 'allows' the code to proceed past the 'DISPLAY SSC-STATUS-LABEL '--- ' SSC-STATUS-VALUE' statement is never 'satisfied' & that the value for SSC-IN01-REQ-RETURN never exceeds 0 which concludes the 'perform IDMS-STATUS-LOOP until SSC-IN01-REQ-RETURN > 0.' statement. The developer insists that this worked prior to release 19.0 (we've been running 19.0 since 8/2016 & haven't had this problem before or currently with any other programs), but I have no way to verify that.

We have modified the IDMS-STATUS routine to use paragraphs vs. a section due to the way a vast majority of our customers have coded their programs, but the changes appear to properly emulate the 'canned' version of IDMS-STATUS. I've attached our version of IDMS-STATUS along with the user routine that is called when an IDMS error has occurred & it in turn calls IDMS-STATUS.

My question - I apologize for the length of the post - has anyone else had this problem or have any suggestions as to what the cause may be? Thanks for your time & any assistance, I appreciate it.

There are no reports of loops in the call to IDMSIN1.
IDMSIN1 just formats the DBKEY, page group, Radix and DML sequence and returns it to the COBOL program as a character string for display in the IDMS-STATUS section. On the last line it sets the return code to 4.
Firstly I would try recompiling it with the default IDMS-STATUS versions 11, 13 or 14 depending on the mode and see if that works.

With the little snippet you posted it looks like you could potentially be issuing a finish after you have lost your run-unit.  I have seen user-inflicted error status loops before.  We have one around here I found that I believe just pure luck has prevented us from having bad situations with. 

Thanks to those who responded - the cause has been determined to be due to a 3rd party product. I won't mention the product here, but if you would like more information, please send me an email - marshall.fuqua@iowa.gov.