The Question is:
Recently the queue_manager process failed because the enqlm had been exceeded.
The limit was set to 5000. I checked the pql parameters and adjusted them as
they were at 2000 for pql_menqlm and 4000 for pql_denqlm, and now set to 3000
and 5000 respective
ly. The process has not failed again. However, in monitoring the process and
its enqlm I have noticed a vasalation ranging from 500 to over 3000 with no
particular connection to the number of entries in the queues.
Can you tell me how the queue_manager uses locking?
I am running OpenVMS 7.3 with all current patches as of 10/19/2002.
The Answer is :
For assistance with a failed queue manager -- whether due to ENQLM
or otherwise -- please contact the support center directly. This
(mis)behaviour is a bug (obviously), and needs to be resolved.
That the queue manager is reportedly sensitive to the system parameter
PQL default settings would imply an additional problem, the queue
manager and other similar OpenVMS processes should not be dependent
on these PQL settings or other similar system parameters.
For details on the queue manager internals, please see the OpenVMS
source listings kit.