Is there any way to SCAN ADABAS file for column names (not the value!).
E.g. If there are 500 ADABAS files (DDMs) and 100 out of them have a column ACCT_NO, how we should find those 100 files? Does NATURAL or ADABAS support this type of SCAN/SEARCH method or some logic?
What I know is DDMs are stored in FDIC file, so how to we SCAN FDIC for a column name?
It is very likely your shop has PREDICT (which is the product Douglas Kelly referenced in his link), and the type of thing you are asking about is available in that product. Your DBA should be able to provide that information for you.
In addition, you may utilize the Natural SCAN utility in each library to look for anywhere that such a field name is referenced, which has the advantage of being self-service for you, but isn’t as clean as the PREDICT option.
I got my field (account_no) in 132 files, is it possible to list these files into a mainframe dataset, i.e. can we issue these steps via a JCL and list the output to a dataset?
How do I use this parameter MAINPR=1? Shall I pass it within my CMSYNIN data?
I tried executing following code:
//CMSYNIN DD *
SYSDIC
LIST ELEMENT ACCOUNT-NO
FIN
But got the below error in my CMPRINT, how do I correct this?
1NEXT SYSDIC
116:30:56 ***** P R E D I C T 4.5.2 *****
- Startup Procedure -
- CATASTROPHIC ERROR
--------------------
Error ........ 6504
Line number .. 0790
Program ...... N-INST
(RESET)
Level ........ 2
NAT6504 SAG EDITOR buffer pool work file open error 4.
Start with a review of the JCL expansion in your JES output. Look in the Natural PROC for DDname CMEDIT (although your DBA has the option to use a different DDname). This is the dataset causing the NAT6504 subcode 4.
********************************* TOP OF DATA **********************************
===> <=== UNKNOWN DRIVER NAME- USING THE DEFAULTS
******************************** BOTTOM OF DATA ********************************