Yang Chen created HBASE-26217:
---------------------------------
Summary: Clean shared state pollution to avoid flaky tests.
Key: HBASE-26217
URL: https://issues.apache.org/jira/browse/HBASE-26217
Project: HBase
Issue Type: Bug
Components: test
Reporter: Yang Chen
h4. PR Link:[https://github.com/apache/hbase/pull/3615]
h2. What is the purpose of this change
* This PR cleans the state polluted by
{{org.apache.hadoop.hbase.monitoring.TestTaskMonitor.testDoNotPurgeRPCTask}}.
* It may be better to clean state pollutions so that some other tests won't
fail in the future due to the shared state pollution.
h2. Reproduce the test failure
* Run the test twice in the same JVM.
h2. Expected result:
* The tests should run successfully when multiple tests that use this state
are run in the same JVM.
h2. Actual result:
* We get the failure:
{{[ERROR] Failures: [ERROR] testDoNotPurgeRPCTask:144 RPC Tasks have been
purged! expected:<10> but was:<20>}}
h2. Why the test fails
* Each time this test runs,10 RPCtasks are created and added to the ArrayList
{{rpcTasks}}, but {{rpcTasks}} is not completely purged when the test ends. So
next time the test starts, the remaining RPC tasks lead to an assertion failure.
h2. Fix
* Clean all tasks when the test ends.
* Another solution is to create a new {{TaskMonitor}} each time to avoid
pollution similar to the other tests (e.g., {{testTaskMonitorBasics}},
{{testTasksGetAbortedOnLeak}}) in {{TestTaskMonitor}}. [Create a new
TaskMonitor each time to avoid flaky tests.
LALAYANG/hbase#1|https://github.com/LALAYANG/hbase/pull/1]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
