zhijiangW commented on a change in pull request #11351: [FLINK-16404][runtime]
Avoid caching buffers for blocked input channels before barrier alignment
URL: https://github.com/apache/flink/pull/11351#discussion_r409271481
##########
File path:
flink-runtime/src/test/java/org/apache/flink/runtime/io/network/netty/NettyPartitionRequestClientTest.java
##########
@@ -142,6 +147,45 @@ public void testDoublePartitionRequest() throws Exception
{
}
}
+ @Test
+ public void testResumeConsumption() throws Exception {
+ final CreditBasedPartitionRequestClientHandler handler = new
CreditBasedPartitionRequestClientHandler();
+ final EmbeddedChannel channel = new EmbeddedChannel(handler);
+ final PartitionRequestClient client =
createPartitionRequestClient(channel, handler);
+
+ final NetworkBufferPool networkBufferPool = new
NetworkBufferPool(10, 32, 2);
+ final SingleInputGate inputGate = createSingleInputGate(1);
+ final RemoteInputChannel inputChannel =
createRemoteInputChannel(inputGate, client, networkBufferPool);
+
+ try {
+ client.resumeConsumption(inputChannel);
Review comment:
is it possible to call `inputChannel#resumeConsumption` instead of
`client.resumeConsumption`, then we can further verify the interaction between
input channel and client.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to 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
