Problems with server result caching on 2 node 11.2.0.3 RAC

I'm working on getting result caching set up on a 2-node RAC cluster running 11.2.0.3.0 on Linux x86-64.

I've got stuff configured, and it seemed to be working. But, then after some time, sometime between yesterday afternoon and late this morning, it seemed to break. Stuff that clearly should be cached, wasn't getting cached. After some time spent spinning my wheels, I discovered that on one node, it was working, and on the other it was not.

Further, I discovered that running 'select dbms_result_cache.status from dual', returns ENABLED on the node that works and CORRUPTED on the node that doesn't work.

I've been trying to find out about this 'CORRUPTED' state, what causes it, how to resolve it (perferably without a bounce) and how to get things working again. I've been googling and searching MOS, but information seems to be pretty scarce.

Anyone else out there w/ experience using server result cache? Any thoughts or suggestions?

Thanks,

-Mark

--

http://www.freelists.org/webpage/oracle-l Received on Wed Jun 11 2014 - 22:12:37 CEST