Re: Aborted 5250 job sucking cpu

[pnelson@xxxxxxxxxx]

PICNIC  (Problem in chair, not in computer)

Do the users have an initial program that's trying to kill the system? 

I usually set the system value to *endjob, so I'll at least get a job log. 
I also deal with an application where the risk of record locks is almost 
zero, so I'll set them to *endjobnolist, because I don't care if they 
click the X in the top right corner instead of signing off.


Paul Nelson
Arbor Solutions, Inc.
708-670-6978  Cell
pnelson@xxxxxxxxxx




rob@xxxxxxxxx
Sent by: midrange-l-bounces@xxxxxxxxxxxx
07/25/2003 03:04 PM
Please respond to Midrange Systems Technical Discussion

 
        To:     MIDRANGE-L@xxxxxxxxxxxx
        cc: 
        Subject:        Aborted 5250 job sucking cpu


Luckily we have a multiprocessor iSeries thus we can still get work done. 
However, when someone's 5250 session ends abnormally all you see are the 
standard messages in QSYSOPR, (which are the same if the session is shut 
down normally):
Session to device (deleted)S1 ended normally. 
Session stopped by a request from device (deleted)S1.
But they jump up to eating an entire one of our processors and a big chunk 

of our interactive capability.  Job still shows on WRKACTJOB dang near 
three hours later.  We ended the job *immed and it's gone now.  Strange, 
but I just noticed in the joblog that it happened within a minute of 
signin.  Like "Oops, I got this message that my message queue is allocated 

to another job.  I'll enter and let it get to the program menu and 
shutdown the session; still signed on.  I only need one session anyway."

Would changing this system value be the fix we need, and if so, what value 

do you recommend?  What drawbacks are there?
System value . . . . . :   QDEVRCYACN 
Description  . . . . . :   Device I/O error action 
Action . . . . . . . . :   *MSG                       *MSG 
                                                      *DSCMSG 
                                                      *DSCENDRQS 
                                                      *ENDJOB 
                                                      *ENDJOBNOLIST


>From the joblog:
CPF5140    Diagnostic              70   07/25/03  12:05:33.402008
  Message . . . . :   Session stopped by a request from device 
(deleted)S1. 
  Cause . . . . . :   The request shutdown was caused by either the user 
turning
    the power off, by a device error, or the ASCII controller inactivity 
timer 
    expired. Recovery  . . . :   Close the files and vary the device off 
(VRYCFG
    command). If the problem occurs again, enter the ANZPRB command to run 

 
    problem analysis. 
CPF5503    Diagnostic              30   07/25/03  12:05:33.402328
  Message . . . . :   Input or Output request failed.  See message 
CPF5140. 
  Recovery  . . . :   See the message CPF5140. Correct the errors and then 

try 
    the request again. 
CPC1125    Completion              50   07/25/03  14:49:32.168440
  Message . . . . :   Job 610712/(deleted)/(deleted)S1 was ended by user 
ROB.

The idea of changing all of our app's to handle *MSG just doesn't tickle 
my toes.

Rob Berendt
-- 
"They that can give up essential liberty to obtain a little temporary 
safety deserve neither liberty nor safety." 
Benjamin Franklin 
_______________________________________________
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing 
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.