User tries to recall an output from tape that was 168 million lines.
User job eventually ended with an RC=16. It had multiple SARDBI23 messages warning that SYS0.SAR.MCS03 Data file(s) was at x% starting at 80% and increasing until it got to 100% and the job failed with the below message.
SARDBI07 No more data file space in database SYS0.SAR.MCS03
However there are no SARDBI23 messages in the SAR task log and it looks like processing continued on as normal.
What dataset or where does SAR temporarily hold these outputs as they're being recalled?
Is it in the actual database files or is it separate?
And if it is in the database itself and the size is as large as it was in this instance, does it pose a risk to normal processing?
Release : 14.0
Component : View
The SARBCH job that is performing the Load is in the background , allocating database blocks from the Free block chain in the Ca View database.
The output is not committed to disk until it is determined that enough space is availalble in the Free block chain , if there is not then the output is effectively deleted from disk and the allocated blocks are returned to the free block chain and processing will continue as normal .
These messages will be seen in the batch job performing the load and not the Ca View STC