Re: Signal API's

Bill Greenfield wrote:

I have done some more playing around with my perceived 'Signal' problem. I have posted portions of 'Stored Procedure' code below
so that any glaring errors can be caught by the community. Also,
I am including portions of the joblog that I forced. If you look
at the subset of the joblog, you will see that a CPE3407
diagnostic message is been sent. Note on the job log the that my
code executed a 'Return' in some CL program at 11:41:06.xxxx.
Next you will see the CPE3407 diagnostic message being issued at
11:41:16 which happens to be the approximate number of seconds that the 'Alarm' was set to (10 seconds). I do not know what all of this means but searching the internet for CPE3407 and QZBSCOMM
does not provide my little brain much useful information. What I
read from this is that "You can't do that dude" (interruption a
function call).

FWiW: The errno is 3407 with name EINTR [recorded as EINTER in places]. The message identifier is CPE3407. The /symptom keyword/ for the message ID is msgCPE3407; i.e. the search token for a symptom is different than the message identifier, in order that a symptom can be located without also finding all ordinary documentation [or casual mention] of the message identifier. For example, in the APAR at the link included below, properly utilizes that /symptom string keyword/ albeit in its different and presumably unrelated /symptom string/:
http://www-01.ibm.com/support/docview.wss?uid=nas22e24beadf9801a2b86256ed4003c940f

FWiW: The message CPE3407 was shown & noted to be sent as a diagnostic. There is no indication from the given joblog, that the logging of that message had caused any negative effects; e.g. an escape message.

<<SNIP>>

Regards, Chuck