Re: Divide by zero and monitor group...

Yes, I did not mention that "intent" - there are many processes that can put messages into the job log, and I found that the method I cited is the only way to clear some of these things. These would be messages that are not receivable any other way.

If one has monitored for a message in CL, for example, then the RCVMSG with DELETE(*YES) - hope my dim memory got that right! - will take care of things. Something similar would be done in RPG.

Here is the link to the article - http://www-01.ibm.com/support/docview.wss?uid=nas8N1015739&context=SGYQGH

The title is "Receiving Messages from the Joblog after the Program is Off the Callstack" - what you do with the messages is whatever you need - not always just removing them.

Enjoy!
Vern

On 7/2/2015 12:31 AM, CRPence wrote:

On 01-Jul-2015 20:25 -0600, Vernon Hamberg wrote:

On 7/1/2015 6:54 PM, Rich Marion wrote:

This is the main reason why I try to avoid MONITOR. It confused
other programmers where I work when they read through job logs.

There's a nice method for removing distracting messages - it
involves sending a message before whatever process you want to clean
up in, get a msg key from that one, then do another at the end - then
loop through the possible message keys (4-byte integer, basically) -
IBM has knowledge base article on it, I don't have the link right
now.

Vendors should be using this kind of technique to clean up their
logs - far too easy just to leave junk lying around!!

That methodology is for intended for messages sent to programs that are no longer on the stack.

The monitored messages will have been sent to the active program; naturally, that program is still on the stack :-) so either call the /remove message/ API for expected\handled message(s) to remove them or call the /move message/ API to bubble\percolate the messages up the stack.