Re: Almost 26 years on this platform, and got my first CPF4235 message

Well the good news is that 2.6 Million is no longer a hard limit. If you create an iASP and put the output queues in there instead the limit goes up, Waaaaay up. (I forget the number) Then if that's not enough create another iASP! Additional bonus is this way the job control blocks are cleaned up and IPL times are better as all that stuff need not be maintained/cleaned up at IPL.

Sure Sure this means you'll use up a lot of disk but anyone who has been in one of my User Group presentations knows, "I sell disk!"

- Larry "DrFranken" Bolhuis

www.frankeni.com
www.iDevCloud.com
www.iInTheCloud.com

On 11/15/2013 9:29 AM, Paul Nelson wrote:

There was little information provided by the help text, but a Goggle search
showed that one of my client's sandbox machines hit the 2.6 million spool
file limitation.

The developers were trying to run a simulation of production for pressure
testing, and had a couple of programs submitting report generation programs
to a bunch of different job queues.

I ended up having to hold all their job queues and kill the jobs because
every one of them was getting the error message. I first noticed it when I
got an alert from the data center that BRMS was puking.

The sad part is that when they complained that their jobs weren't running
anymore, I told them that the job queues were held so I could do some
housekeeping. The lead guy actually asked me what a job queue was. He also
said they didn't need to keep any of the reports. The CLROUTQ command took
over 6 hours to flush them all out. I now have a scheduled job that holds
all their job queues, runs the CLROUTQ command, and then releases their job
queues. Then it waits 2 hours and repeats the process.

The moral of the story? Outsourcing to inexperienced people across the pond
is like giving whiskey and car keys to a pack of teenage boys.

OK, I'm done venting.

Paul Nelson
Cell 708-670-6978
Office 512-392-2577
nelsonp@xxxxxxxxxxxxx