MINOR: reduce commit interval and cache size for integration test

[mjsax]

No description provided.

[mjsax]

Call for review @enothereska @dguy

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk8-scala2.11/2619/
Test FAILed (JDK 8 and Scala 2.11).

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk7-scala2.10/2615/
Test PASSed (JDK 7 and Scala 2.10).

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk8-scala2.12/2615/
Test PASSed (JDK 8 and Scala 2.12).

[dguy]

LGTM

[enothereska]

@mjsax is there an error that we're trying to fix by changing the commit interval. Note that with a non-0 cache size, even with a commit interval of 0, there is still a chance that the cache dedups. So if there is a dedup error, changing the commit interval won't be sufficient.

[mjsax]

Yes -- I am aware of that. This is only about timeouts -- by default, we commit every 30 seconds, and ofter tests wait for 30 seconds to receive output -- reducing commit interval should stabilize (and maybe speed up) the tests by avoiding race conditions.

[mjsax]

Rebased to get latest test stabilizations from trunk and to trigger new build.

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk8-scala2.11/2650/
Test PASSed (JDK 8 and Scala 2.11).

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk7-scala2.10/2646/
Test PASSed (JDK 7 and Scala 2.10).

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk8-scala2.12/2646/
Test FAILed (JDK 8 and Scala 2.12).

[mjsax]

@enothereska Doesn't seem to solve it for all cases... one build fails with

java.lang.AssertionError: Condition not met within timeout 30000. Expecting 1 records from topic output- while only received 0: []
	at org.apache.kafka.test.TestUtils.waitForCondition(TestUtils.java:265)
	at org.apache.kafka.streams.integration.utils.IntegrationTestUtils.waitUntilMinKeyValueRecordsReceived(IntegrationTestUtils.java:206)
	at org.apache.kafka.streams.integration.utils.IntegrationTestUtils.waitUntilMinKeyValueRecordsReceived(IntegrationTestUtils.java:175)
	at org.apache.kafka.streams.integration.KTableKTableJoinIntegrationTest.verifyKTableKTableJoin(KTableKTableJoinIntegrationTest.java:221)
	at org.apache.kafka.streams.integration.KTableKTableJoinIntegrationTest.shouldInnerLeftJoin(KTableKTableJoinIntegrationTest.java:142)

[enothereska]

Yeah, let me have a look.

[mjsax]

@enothereska I think it makes still sense to merge this PR -- can you have a look?

[enothereska]

Ok LGTM. This can probably wait for @guozhangwang to be checked in since other committers might be busy with 0.10.2 bug fix release and this is not urgent.

[mjsax]

Rebases this.

Call for final review and merging @guozhangwang

[guozhangwang]

@mjsax should this go to 0.10.2 as well?

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk7-scala2.10/3027/
Test PASSed (JDK 7 and Scala 2.10).

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk8-scala2.11/3032/
Test PASSed (JDK 8 and Scala 2.11).

[asfbot]

Refer to this link for build results (access rights to CI server needed):
https://builds.apache.org/job/kafka-pr-jdk8-scala2.12/3028/
Test PASSed (JDK 8 and Scala 2.12).

[mjsax]

@guozhangwang Did not think about it -- might make sense to stabilize 0.10.2 builds, too.

[guozhangwang]

LGTM and merged to trunk and 0.10.2.