GJL commented on a change in pull request #10666: [FLINK-15214] Introduce
multiple submission e2e test
URL: https://github.com/apache/flink/pull/10666#discussion_r363827085
##########
File path: tools/travis/splits/split_container.sh
##########
@@ -51,6 +51,7 @@ run_test "Run kubernetes test"
"$END_TO_END_DIR/test-scripts/test_kubernetes_emb
run_test "Run kubernetes session test"
"$END_TO_END_DIR/test-scripts/test_kubernetes_session.sh"
if [[ "${HADOOP_INTEGRATION}" = "with-hadoop" ]]; then
run_test "Run Mesos WordCount test"
"$END_TO_END_DIR/test-scripts/test_mesos_wordcount.sh"
Review comment:
I agree that a separate test makes it easier to locate the root cause.
However, due to being expensive, E2E tests should not be used for exhaustively
testing all cases. Currently, running all E2E tests take more than 24 CPU hours
on Travis. `e2e - container - hadoop 2.8` takes ~43 minutes on Travis, which is
not great but acceptable. Imo, if we decide to add more Mesos tests, we should
start to consolidate soon.
cc: @tillrohrmann
----------------------------------------------------------------
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
