SunJSSE consumers often parse handshake messages and extensions via consumers into spec objects. These spec objects are sometimes attached to the handshake context and held at least through the handshake and possibly are passed onto child session contexts. This includes vectors that contain duplicate identifier values currently not filtered out. This can result in increased memory consumption. It can also force increased processing time if those vectors need to be walked or have their values converted to enumeration members by their ID value.

Consumers parsing vectors should filter out duplicate IDs before storing them in spec objects or otherwise attaching them to the handshake context.

Producer filtration of duplicate identifiers is out of the scope for this bug.