Might these be not "broken", but continuation records? For example.
when TYPE30 records have more DD segments than will fit in a record,
continuation records are written with some sections of the TYPE30
omitted in the continuation records. However, the identification
section is present in the continuation records to allow proper
identification of the task related to the additional DD segments.
Michael
At 11:28 AM 9/14/2021, Pierre Fichaud wrote:
To All,
For SMF types 70-79 (RMF), RMF may create broken records
because the volume of data can't fit in 32K.
So there may be 2 or more "broken" records created.
I have an IBM sample file that has the original unbroken record
split into 3 broken records.
The records are consecutive in the sample file.
But I am getting the records in my SMF exit and I must save the
broken records to put them together at a later time.
There doesn't seem to be a unique identifier in the product
section connecting the 3 records.
I will use the SMF type/subtype/date/time in the SMF header for now.
But the time in 100ths of seconds doesn't seem granular enough to me.
Can someone help me here ?
Thanks in advance, Pierre.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
