[HADOOP-13776] remove redundant classpath entries in RunJar

• Type: Bug
• Status: Patch Available
• Priority: Major
• Resolution: Unresolved
• Component/s: util
• Assignee: Sangjin Lee
• Reporter: Sangjin Lee
• Created: 31/Oct/16 22:28
• Updated: 01/Nov/16 03:50

Description

Today when you run a "hadoop jar" command, the content of the jar gets added to the classpath twice, once in the jar form, and again in an unpacked form.

We should include the content of the jar to the classpath only once. We should keep the jar in the classpath (to support setJarByClass and other useful use cases) but remove the root of the unpacked directory.

Issue Links

• is related to: HADOOP-13620 Mapreduce job failure on submission
• is related to: HADOOP-13410 RunJar adds the content of the jar twice to the classpath