VSAM Return Code 44 Received on Write to Fixed Length File

John Weber

unread,

Nov 19, 2012, 6:11:55 PM11/19/12

to

We are using a batch process to write account records to a VSAM file
while also writing records to an audit file which is also a VSAM file.

Both files include an index and several alternate indexes.
Each alternate index allows duplicates (NONUNIQUEKEY).

The audit file receives a VSAM return code of 44 and discontinues writing
after writing 3,258 of 10,000 records.

Neither file is a variable file so the explanation of return code 44 is not
relevant to the problem or solution. (44 - REWRITE of different size record)

The sizing in our allocations of our primary data and secondary index elements
are appropriate for both files given the data in this exercise.

Please advise on problem solving this issue and/or what additional information is needed.

Thank you...

John

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to list...@listserv.ua.edu with the message: INFO IBM-MAIN

Lizette Koehler

unread,

Nov 19, 2012, 6:28:03 PM11/19/12

to

Could you post the IDC messages for the 44?

Lizette

John Weber

unread,

Nov 19, 2012, 7:20:56 PM11/19/12

to

The only sysprint from the job is the status code display from the program.

John McKown

unread,

Nov 19, 2012, 7:43:55 PM11/19/12

to

My habit with VSAM is to use the second COBOL FILE STATUS variable to get
the actual VSAM return and reason codes. Often that is the only way to
determine the actual reason.

What comes to mind in your case is that the RECORDSIZE on your AIX is too
small to record all the base keys associated with a single AIX key. This
can cause a 44. Normally these keys are either ALL SPACES or ALL LOW-VALUES
or maybe ALL HIGH-VALUES.

John Weber

unread,

Nov 19, 2012, 7:52:31 PM11/19/12

to

Interesting about a secondary status variable. We'll investigate that.

Regarding the AIX definitions. We do not include a RECORDSIZE parameter.

John McKown

unread,

Nov 19, 2012, 8:09:46 PM11/19/12

to

Then u get some default. Do a LISTC on the AIX to see what it is. You might
want to PRINT the records in the AIX itself just to see. Or on an EXAMINE
on it.

John Weber

unread,

Nov 19, 2012, 8:36:48 PM11/19/12

to

What are we looking for in the LISTCAT output? RECORDSIZE is not found. Is this because our VSAM file has a fixed length? Thanks.

> > -- For IBM-MAIN subscribe / signoff / archive access instructions,

> > send email to list...@listserv.ua.edu with the message: INFO
> > IBM-MAIN
> >
> >
> > --------------------------------------------------------------------

> > -- For IBM-MAIN subscribe / signoff / archive access instructions,

John Weber

unread,

Nov 21, 2012, 1:23:17 PM11/21/12

to

Thank you for the assistance with our problem.

It turns out our secondary index which allows non-unique records was running into the default limit.

Happy Thanksgiving everyone.

John

> > -- For IBM-MAIN subscribe / signoff / archive access instructions,

> > send email to list...@listserv.ua.edu with the message: INFO
> > IBM-MAIN
> >
> >
> > --------------------------------------------------------------------

> > -- For IBM-MAIN subscribe / signoff / archive access instructions,

mailk...@gmail.com

unread,

Sep 6, 2015, 4:01:06 PM9/6/15

to

Hi,

I know this is old discussion, sorry to ask now. now I am getting 44 now when I try to rewrite in cobol program, please let me know how did you resolve this last time.