Re: SQL7917

On 26 Apr 2012 05:46, Mike Cunningham wrote:

Rob on Thursday, April 26, 2012 8:41 AM wrote:

http://www.ibm.com/developerworks/forums/thread.jspa?messageID=14494840&tstart=0

I did find that but the thread was 18 months old and talks about a
way to stop the error but does not explain how to determine what is
being locked and who has the lock. And says going to dynamic will
have a negative impact on performance

FWiW Dynamic SQL is already being used in the scenario given in this thread; albeit the effect could have occurred on a CALL to a compiled stored procedure(s) on the server. An extension to the dynamic allows for storing some statements and their plans in an SQL Package on the server; i.e. Extended Dynamic. Note that the severity of the message is zero.

The section of spooled joblog of the "QZDASOINIT job for the ODBC connection" showing that message and preceding messages was not included in the OP. So, is the message seen only within a debug [via options or actual start-debug] joblog? If seen without debug, then turn on the debug messages to see if there are more details; I infer SQL7917 still has no pointer nor other replacement text available at whatever release to identify the object that was not updated.? What release; most relevant for finding possible defects? The DB monitor [STRDBMON; there may be an option to start the monitor, as I believe there is an option for debug messaging, for the connection, else they can be started against the ODBC job from elsewhere] I seem to recall tracks access plan rebuilds; perhaps also to where they are [attempted to be] stored.

Regards, Chuck