High volumes of SQL errors can cause the job log for the QSQSRVR prestart jobs to become very large, and may cause problems with storage.
Any supported version of IBM i.
Please check if your prestart jobs use the default prestart job description or some other job description, and verify that the job description uses the default values for the job message queue maximum size and full action parameters.
From a 5250 command line with *SECOFR class user profile:
Use option "10. Prestart job entries". Find QSQSRVR on the list, and use option 5 to display it. In the "Display Prestart Job Entry Detail" display, the job description should be listed. The default value is QDFTSVR in library QGPL. If the job description is different, make a note of the actual value for use in the next step. Use F3 to get back to the command line. Then use this command (remember to use the different job description if you are using something different):
On the "Display Job Description" display, page down to find these:
Job message queue maximum size . . . . . . . . . : 4
Job message queue full action . . . . . . . . . : *WRAP
The values '4' and '*WRAP' are recommended to prevent the job log from becoming too large. If you have different values for these parameters, please consider using these recommended values (remember to change the job description in his command to whatever you are actually using):
CHGJOBD JOBD(QGPL/QDFTSVR) JOBMSGQMX(04) JOBMSGQFL(*WRAP)
Reference: Changing job attributes for prestart jobs