Kenneth, I would agree that when the jobs hit JES2 the damage is done. To handle things in the future you could look at increasing the size of your checkpoint datasets. Maybe double or triple the current size so that if needed you can dynamically adjust things. Naturally if you are using a coupling facility structure as one of your checkpoint that would also have to increase. To refresh my memory I took a quick look in the JES2 Initialization and Tuning Guide for z/OS 2.4 and the max size a checkpoint dataset can be is 17,000 tracks.
>From the manual: If you don't want to calculate the size, you can use a trial-and-error approach by over-allocating the space (two or three times your current allocation) and let JES2 tell you with the $HASP296 or $HASP542 message if you need more. The largest checkpoint that JES2 will currently use is approximately 17,000 tracks on a 3390 device (Approximately 200,000 4K records). Allocating a new data set of this size will ensure that it can accommodate increasing any section to the maximum size. Paul Feller GTS Mainframe Technical Support -----Original Message----- From: IBM Mainframe Discussion List <[email protected]> On Behalf Of Kenneth J. Kripke Sent: Monday, May 9, 2022 8:00 AM To: [email protected] Subject: $HASP052 JES2 RESOURCE SHORTAGE OF BERT XX% UTILIZATION REACHED [EXTERNAL] Hello; We had an incident where about 8,000 jobs were submitted via a distributed process. All the jobs were of the same name. It is my understanding that the BERT contains the CAT and JQE as well as the WSCQ as indicated in a bit indicator. If the jobs were placed on hold via the $H GRP,JM=, I am guessing The control blocks will still be built, hence, no benefit. Is my assumption correct? I would like to pose the question on how others have resolved the problem. Sincerely; Kenneth J. Kripke [email protected] <mailto:[email protected]> ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- Please note: This message originated outside your organization. Please use caution when opening links or attachments. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
