I am seeing the following warning in my Kafka connect logs :
[2024-02-12 10:35:02,489] WARN [sink|task-4] [com.couchbase.core][TooManyInstancesDetectedEvent] The number of simultaneously connected Cluster instances (7) exceeds the configurable warning threshold of 1. This is a diagnostic message to help detect potential resource leaks and inefficient usage patterns. If you actually intended to create this many instances, please ignore this warning, or increase the warning threshold by calling Cluster.maxAllowedInstances(int) on startup. However, if you did not intend to have 7 Cluster instances connected at the same time, this warning may indicate a resource leak. In that case, please make sure to call cluster.disconnect() after a Cluster and its associated Buckets, Scopes, Collections, etc. are no longer required by your application. Also note that Cluster, Bucket, Scope, and Collection instances are thread-safe and reusable until the Cluster is disconnected; for best performance, reuse the same instances throughout your application's lifetime. (com.couchbase.core:136)
|
|
| |
As the long indicates, this might be OK, but it is unclear to me if the Kafka Connector Sink was meant to run this way. If so, perhaps it should call `maxAllowedInstances`? Or disable this log?
Improvement
Minor