Will Berkeley has uploaded this change for review. ( http://gerrit.cloudera.org:8080/13453
Change subject: [java] Deflake TestAsyncKuduSession with temporary workaround ...................................................................... [java] Deflake TestAsyncKuduSession with temporary workaround TestAsyncKuduSession.test is kind of a mess. In particular, it has one section where it tries to insert 21 ops into a session with 20 ops of buffer space. It expects this to trigger a PleaseThrottleException. However, it is inherently racy because the session may manage to flush one or more buffers of ops to the server so no throttling is needed. It's always been an icky test but it wasn't flaky until 2bf80b5147f09c980c2f2268ee3467a5b3305a57 cleaned up the session code a bit. Now it's about 30% flaky in RELEASE mode according to the flaky test dashboard. Until I can spend some time to redo the test and understand exactly what about the previous change triggered the flakiness, I'm turning off the check for an exception. I also moved a debug log statement so there's less noise when calling flush in non-AUTO_FLUSH_BACKGROUND modes. Change-Id: I751f1ed3a151eee611a8b88c1527e12271024bf6 --- M java/kudu-client/src/main/java/org/apache/kudu/client/AsyncKuduSession.java M java/kudu-client/src/test/java/org/apache/kudu/client/TestAsyncKuduSession.java 2 files changed, 6 insertions(+), 10 deletions(-) git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/53/13453/1 -- To view, visit http://gerrit.cloudera.org:8080/13453 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: newchange Gerrit-Change-Id: I751f1ed3a151eee611a8b88c1527e12271024bf6 Gerrit-Change-Number: 13453 Gerrit-PatchSet: 1 Gerrit-Owner: Will Berkeley <[email protected]>
