[
https://issues.apache.org/jira/browse/HADOOP-10787?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14976118#comment-14976118
]
Varun Vasudev commented on HADOOP-10787:
----------------------------------------
Thank for the patches [~aw]. The latest patch doesn't apply to trunk. Can you
please take a look?
Also, mapred still has a reference to TOOL_PATH
{code}
archive-logs)
CLASS=org.apache.hadoop.tools.HadoopArchiveLogs
hadoop_debug "Injecting TOOL_PATH into CLASSPATH"
hadoop_add_classpath "${TOOL_PATH}"
hadoop_debug "Appending HADOOP_CLIENT_OPTS onto HADOOP_OPTS"
HADOOP_OPTS="${HADOOP_OPTS} ${HADOOP_CLIENT_OPTS}"
{code}
> Rename/remove non-HADOOP_*, etc from the shell scripts
> ------------------------------------------------------
>
> Key: HADOOP-10787
> URL: https://issues.apache.org/jira/browse/HADOOP-10787
> Project: Hadoop Common
> Issue Type: Improvement
> Components: scripts
> Affects Versions: 3.0.0
> Reporter: Allen Wittenauer
> Assignee: Allen Wittenauer
> Priority: Blocker
> Labels: scripts
> Attachments: HADOOP-10787.00.patch, HADOOP-10787.01.patch,
> HADOOP-10787.02.patch
>
>
> We should make an effort to clean up the shell env var name space by removing
> unsafe variables. See comments for list.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
