azagrebin commented on a change in pull request #6782: [FLINK-9083][Cassandra
Connector] Add async backpressure support to Cassandra Connector
URL: https://github.com/apache/flink/pull/6782#discussion_r224475206
##########
File path:
flink-connectors/flink-connector-cassandra/src/main/java/org/apache/flink/streaming/connectors/cassandra/CassandraSink.java
##########
@@ -354,6 +359,23 @@ protected Cluster buildCluster(Cluster.Builder builder) {
return this;
}
+ /**
+ * Sets the maximum allowed number of concurrent requests for
this sink.
+ *
+ * <p>This call has no effect if {@link
CassandraSinkBuilder#enableWriteAheadLog()} is called.
+ *
+ * @param maxConcurrentRequests maximum number of concurrent
requests allowed
+ * @param timeout timeout duration when acquiring a permit to
execute
+ * @param unit timeout unit when acquiring a permit to execute
+ * @return this builder
+ */
+ public CassandraSinkBuilder<IN> setMaxConcurrentRequests(int
maxConcurrentRequests, long timeout, TimeUnit unit) {
Review comment:
I would also add a short hand with default max timeout basically blocking
the task thread:
`setMaxConcurrentRequests(int maxConcurrentRequests);`
In general, I think users would be more interested in blocking indefinitely
case which just propagates the back pressure upstream and it can be also seen
in UI. As I understand, now throwing TimeoutException will fail and job, it
will be restarted, probably fail with timeout again and so on. Then it is
better just to block and avoid restarts.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
