[
https://issues.apache.org/jira/browse/HDFS-4055?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Eli Collins updated HDFS-4055:
------------------------------
Summary: TestAuditLogs is flaky (was: TestAuditLogs has some chances to
fail.)
> TestAuditLogs is flaky
> ----------------------
>
> Key: HDFS-4055
> URL: https://issues.apache.org/jira/browse/HDFS-4055
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Binglin Chang
> Assignee: Binglin Chang
> Attachments: HDFS-4055-TestAuditLogs-bug.patch
>
>
> {code}
> InputStream istream = webfs.open(file);
> int val = istream.read();
> istream.close();
> verifyAuditLogsRepeat(true, 3);
> assertTrue("failed to read from file", val >= 0);
> {code}
> {code}
> byte[] toWrite = new byte[bufferLen];
> Random rb = new Random(seed);
> long bytesToWrite = fileLen;
> while (bytesToWrite>0) {
> rb.nextBytes(toWrite);
> {code}
> InputStream.read() return the first byte of the file, the bytes in the file
> is generated in using Random.nextBytes(), so you get 1/256 chance the first
> byte is 0, so some times it may fail.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
