Peter, Actually, that's good news from my viewpoint since it removes the QuerySQLCache variable from the equation.
How are you performing the enhancement processing [1] for your Entities? Are you doing it during build time, or dynamically via an agent or loader? Any differences between the two executions in this regard? It's looking like a potential race condition for the meta data repository. Looking for more ideas... Kevin [1] http://webspherepersistence.blogspot.com/2009/02/openjpa-enhancement.html On Wed, Jul 15, 2009 at 8:48 AM, Peter Henderson < [email protected]> wrote: > Kevin, > > I have added that property and ran several tests. The problem remains. Most > of the time one of the worker threads throws, occasionally start up is > successful. > > Thanks > > Peter. > > > > > Kevin Sutter wrote: > >> Hi Peter, >> Another item to try is to turn off the sql generation cache. I'm taking a >> long shot here, but the call stack shows that we're attempting to get the >> query from the cache. If there's a problem with synchronizing the cache >> access, then turning it off would help isolate the problem. Documentation >> for this sql generation cache can be found here [1], but to turn it off >> just >> set this property: >> >> <property name="openjpa.jdbc.QuerySQLCache" value="false"/> >> >> Kevin >> > > -- > Peter Henderson > Director Starjar Limited. > > Mobile: +44 (0) 778 233 8645 > Email: [email protected] > Web: www.starjar.com >
