[jira] [Created] (HDFS-9727) Improve the exception's log output in DataXceiver
Lin Yiqun created HDFS-9727:
---
Summary: Improve the exception's log output in DataXceiver
Key: HDFS-9727
URL: https://issues.apache.org/jira/browse/HDFS-9727
Project: Hadoop HDFS
Issue Type: Improvement
Affects Versions: 2.7.1
Reporter: Lin Yiqun
Assignee: Lin Yiqun
Priority: Minor
There are some not appropriate logLevel output in many palaces of
{{DataXceiver}}. When some exceptions happened, the output logLevel often the
info level not warn. Like these:
{code}
try {
blockSender = new BlockSender(block, blockOffset, length,
true, false, sendChecksum, datanode, clientTraceFmt,
cachingStrategy);
} catch(IOException e) {
String msg = "opReadBlock " + block + " received exception " + e;
LOG.info(msg);
sendResponse(ERROR, msg);
throw e;
}
{code}
And this case is not only one, it looks frequent in DataXceiver methods.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
[jira] [Created] (HDFS-9728) TestDFSShell should not create MiniDFSCluster in each test case
John Zhuge created HDFS-9728: Summary: TestDFSShell should not create MiniDFSCluster in each test case Key: HDFS-9728 URL: https://issues.apache.org/jira/browse/HDFS-9728 Project: Hadoop HDFS Issue Type: Improvement Reporter: John Zhuge Assignee: John Zhuge Priority: Critical Create the MiniDFSCluster in @BeforeClass method, and shut it down in @AfterClass method. This fix will speed up unit testing of HDFS component. When a test case needs 2 clusters, use the default one, and create a second one in the test case. If there are too many test cases that need 2 clusters, move them to a different test file. Only when a custom cluster specific to that test case is needed, create the cluster in the test case. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
