[
https://issues.apache.org/jira/browse/HDFS-13975?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16642710#comment-16642710
]
Ayush Saxena commented on HDFS-13975:
-------------------------------------
Thanks [~jlowe] for putting this up.
But in this test I guess it was previously realized that it can fail with this
exception and was considered fine as of then.
If you see the code it is explicitly mentioned there.
{code:java}
// accept runtime if it is under 3.5 seconds, as we need to wait for
// IN_PROGRESS report from DN, and some spare to be able to finish.
// NOTE: This can be a source of flaky tests, if the box is busy,
// assertion here is based on the following: Balancer is already set
// up, iteration gets the blocks from the NN, and makes the decision
// to move 2 blocks. After that the PendingMoves are scheduled, and
// DataNode heartbeats in for the Balancer every second, iteration is
// two seconds long. This means that it will fail if the setup and the
// heartbeat from the DataNode takes more than 500ms, as the iteration
// should end at the 3rd second from start. As the number of
// operations seems to be pretty low, and all comm happens locally, I
// think the possibility of a failure due to node busyness is low.
assertTrue("Unexpected iteration runtime: " + runtime + "ms > 3.5s",
runtime < 3500);
{code}
The timeout is for the best case scenario.For Average cases it doesn't provide
any margin.IIUC Increasing that 3.5s limit is the only way to get this test a
little ahead.Since all other factors which lead to this timeout seems beyond
control.But not sure that going that way can take away the logic for this
assertion.
> TestBalancer#testMaxIterationTime fails sporadically
> ----------------------------------------------------
>
> Key: HDFS-13975
> URL: https://issues.apache.org/jira/browse/HDFS-13975
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 3.2.0
> Reporter: Jason Lowe
> Priority: Major
>
> A number of precommit builds have seen this test fail like this:
> {noformat}
> java.lang.AssertionError: Unexpected iteration runtime: 4021ms > 3.5s
> at org.junit.Assert.fail(Assert.java:88)
> at org.junit.Assert.assertTrue(Assert.java:41)
> at
> org.apache.hadoop.hdfs.server.balancer.TestBalancer.testMaxIterationTime(TestBalancer.java:1649)
> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> at
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
> at
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> at java.lang.reflect.Method.invoke(Method.java:498)
> at
> org.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:47)
> at
> org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:12)
> at
> org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:44)
> at
> org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:17)
> at
> org.junit.internal.runners.statements.FailOnTimeout$StatementThread.run(FailOnTimeout.java:74)
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
