You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-dev@hadoop.apache.org by "Shilun Fan (Jira)" <ji...@apache.org> on 2025/10/09 01:35:00 UTC
[jira] [Created] (MAPREDUCE-7521) Fix TestUberAM failures after JUnit 5 migration
Shilun Fan created MAPREDUCE-7521:
-------------------------------------
Summary: Fix TestUberAM failures after JUnit 5 migration
Key: MAPREDUCE-7521
URL: https://issues.apache.org/jira/browse/MAPREDUCE-7521
Project: Hadoop Map/Reduce
Issue Type: Improvement
Components: mapreduce-client
Affects Versions: 3.5.0
Reporter: Shilun Fan
Assignee: Shilun Fan
*Background:*
After migrating the project to {*}JUnit 5{*}, the {{TestUberAM}} unit test started failing intermittently. The error message observed during test execution is as follows:
{code:java}
Remote File /apps_staging_dir/jenkins/.staging/job_1759472296837_0013/job.jar could only be written to 0 of the 1 minReplication nodes.
There are 2 datanode(s) running and 2 node(s) are excluded in this operation. {code}
This error originates from
{code:java}
org.apache.hadoop.hdfs.server.blockmanagement.BlockManager.chooseTarget4NewBlock {code}
indicating that the MiniDFSCluster used in the test failed to find available data nodes for block writes during execution.
*Change Summary:*
* Removed redundant invocation of the parent class’s {{BeforeAll}} method in subclasses to prevent double initialization.
* Increased MiniDFSCluster data node count from 2 to 5 to improve fault tolerance.
* Verified that all affected tests pass successfully.
*Impact Scope:*
* Affects only AM tests based on MiniDFSCluster.
* No impact on business logic or production code.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: mapreduce-dev-unsubscribe@hadoop.apache.org
For additional commands, e-mail: mapreduce-dev-help@hadoop.apache.org