Answer
This error comes from IFS Middleware's own periodic connection-pool health check — it pings the database to confirm pooled sessions are still genuinely alive before handing them out to data sources, and 'pingDatabase returns not OK' means that ping is failing, which strongly indicates a TNS-level connectivity interruption between the Middleware Server and the database server rather than an application-level bug. A brief interruption can sometimes self-recover, but an extended one can cause middleware deployments to fail and, as seen here, generate runaway log volume from the repeated failed ping attempts. Resolving it typically requires restarting the IFS services (and sometimes the database as well) once the underlying network/connectivity interruption between middleware and database has been addressed — the log flooding itself is a symptom of the connectivity problem, not something to fix by adjusting logging settings.