Re: Tight loop - stopping the error

[CRPence <CRPbottle@xxxxxxxxx>]

  HLDJOB or ENDJOB should work sufficiently well, to stop a coded loop. 
 However two things about the proposed loop.

  First, I am not entirely sure that the default handler for a CLP [in 
sending the CPA0701] will effect the described hard loop on the failed 
attempt to send that inquiry to the QSYSOPR message queue.  Regardless 
of whether that would be an origin for a loop, a better idea IMO is to 
code the loop in the program itself.  That means the request should have 
a MONMSG coded to ignore the failure, and a GOTO to try again; or coded 
in a while or another supported loop syntax.

  Second, I think a better choice is to minimize the amount of work to 
cause the T-AF entry.  There should be no reason to send a message, and 
especially not to a system message queue, and even more importantly not 
to the *SYSOPR message queue since every message if it actually made it 
there will then be copied to QHST thus flooding the history log files 
with the useless impromptu message... making interrogating the history 
log seriously painful; understood that the setup should prevent that, 
but why take the chance?  Using just CHKOBJ against an object, even 
QSYSOPR *MSGQ, should cause the authority failure entry to be logged; 
presumably with less processing by the OS code, to effect that request.

  For example:

    pgm /* replace ## to reflect the 'not aut' msg */
    again:
     chkobj qsys/qsysopr *msgq
     monmsg cpf98## exec(rcvmsg ... rmv(*yes))
     goto again
    endpgm

Regards, Chuck

John Earl wrote:
> <<SNIP>>
>
> Here is one idea that I had, but I don't know how to stop it once 
> we begin.  Any ideas on this particular approach would be greatly 
> appreciated.  Here is the plan:
>
> 1. Create a User called "Fred" 
> 2. Give "Fred" *EXCLUDE authority to > the QSYSOPR message queue.
> 3. Have "Fred" submit a batch job that does a
>    SNDMSG 'Hello World' TOMSGQ(QSYSOPR).
>
> Because "Fred" has *EXCLUDE authority to QSYSOPR, this command will 
> generate an 'T' 'AF' audit journal entry in QAUDJRN.  Because it is
> done in a batch job, the batch job attempt to throw and exception
> error message into QSYSOPR, but because "Fred" is still *EXCLUDED
> from message queue QSYSOPR, the process will generate another 'T'
> 'AF', and yet another error message headed for QSYSOPR, etc. etc.
> in a very tight little loop.
>
> This seems like a good way to generate a ton of audit log entries,
> but the question is this:  How do you stop it before it runs away
> with your system?  Can anyone think of a way to interrupt this
> process?