Re: Job hangs after processing 220 records.

On 02-Jul-2014 08:33 -0500, Marie O'Rourke wrote:

<<SNIP>> Calling IBM support will be the last option, once I have
investigated it to my full potential, I will pass to the ISeries team
to investigate and then it might go to IBM... <<SNIP>>

With regard to the MCH3603 alone, any delay in followup pursuant to resolution, IMNSHO could be a huge mistake. If there is an indication that the OS asked to perform for example a *destroy object* request, but the LIC has notified with the x2403 exception that the object type of the object to be destroyed was incompatible with the selected method against that object, then that is a _clear indication_ that the OS might be in other instances, successfully but unintentionally destroying objects for which the selected method had _by chance_ been correctly matching the object type.

Minimally, I would immediately apply all HIPer maintenance and anything that is known to address a condition in which the MCH3603 was identified as a possible symptom.

Note: Not knowing the invoker of #cfochkr, which may appear in an existing vlog, or in a VLog forced per a process dump initiated for the MCH3603, if the error originates with a /destroy/ operation is only speculation.

P.S. IIRC one of the APARs I had mentioned in my first reply dealt with object references in the SQL Plan Cache. So just the act of IPL might give temporary relief; without any preventive fix applied during that IPL, only until the unresolved issue reappears.