How does the high availability algorithm decide from which connection pool in the MultiPool list a connection will be used?
The MultiPool will always try to provide a connection from the JDBC connection pools in the order they are defined for the MultiPool. This means that for every connection request, the MultiPool tries to provide a connection from the first pool, then should this fail it will try the second pool, and if this also does not work, the third pool and so on. As a pool is never marked as bad or not available by the MultiPool itself, the next request for a connection from the MultiPool will again try to provide a connection in the same manner as described before. For WebLogic Server versions 8.1 SP3 / 7.0 SP5 and upwards, this is changed, i.e., if a connection request to one of the pools in the list fails, it is marked as bad. Please check Enhancements for details.
Related Questions
- How does the high availability algorithm decide from which connection pool in the MultiPool list a connection will be used?
- How can I disable the first (or any) JDBC connection pool so that the Multipool goes directly to the next pool in the list?
- What algorithm is used in the Jdbc pool for re-using the pooled connections ?
