Like HADOOP-7491 which added these environment variables to the case where the hadoop script is run with a classname, they should also be used when run with jar as the command. All the other commands use these env vars, I can't see why jar doesn't.

Since every case in the big if block have would now had:

HADOOP_OPTS="$HADOOP_OPTS $HADOOP_CLIENT_OPTS"

I just removed each individual line and added a single line below the block. It has the nice side affect of making that big block more readable

-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12537914/HADOOP-8623.patch
against trunk revision .

+1 @author. The patch does not contain any @author tags.

-1 tests included. The patch doesn't appear to include any new or modified tests.
Please justify why no new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.

+1 javac. The applied patch does not increase the total number of javac compiler warnings.

+1 javadoc. The javadoc tool did not generate any warning messages.

+1 eclipse:eclipse. The patch built with eclipse:eclipse.

+1 findbugs. The patch does not introduce any new Findbugs (version 1.3.9) warnings.

+1 release audit. The applied patch does not increase the total number of release audit warnings.

-1 core tests. The patch failed these unit tests in hadoop-common-project/hadoop-common:

org.apache.hadoop.ha.TestZKFailoverController

+1 contrib tests. The patch passed contrib unit tests.

Test results: https://builds.apache.org/job/PreCommit-HADOOP-Build/1220//testReport/
Console output: https://builds.apache.org/job/PreCommit-HADOOP-Build/1220//console

This message is automatically generated.

the same is being repeated in hadoop-hdfs-project/hadoop-hdfs/src/main/bin/hdfs and hadoop-mapreduce-project/bin/mapred scripts. I think we can move assignment of HADOOP_OPTS to before the case statement and remove the line from various if statements in bin/mapred and bin/hdfs.

Actually thinking about it more, hdfs and mapred could have separate configs and can be invoked independently from the bin/hadoop script so the current changes look good to go.

+1

+1 ditto

Thanks for contributing the patch Steven. I have added you as a contributor to Hadoop common. You can now assign the jiras to yourself.

I committed the patch to trunk. Will merge it into 2.x next.

I merged this to branch-0.23 as well.

Integrated in Hadoop-Hdfs-0.23-Build #388 (See https://builds.apache.org/job/Hadoop-Hdfs-0.23-Build/388/)
svn merge -c 1366126 FIXES: HADOOP-8623. hadoop jar command should respect HADOOP_OPTS. Contributed by Steven Willis. (Revision 1391276)

Result = UNSTABLE
jlowe : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1391276
Files :

• /hadoop/common/branches/branch-0.23/hadoop-common-project/hadoop-common/CHANGES.txt
• /hadoop/common/branches/branch-0.23/hadoop-common-project/hadoop-common/src/main/bin/hadoop

I am out of the office until Tuesday October 16th. Please contact Seth Madison for any urgent issues.