[
https://issues.apache.org/jira/browse/HADOOP-7089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12981098#action_12981098
]
Eli Collins commented on HADOOP-7089:
-------------------------------------
Some extra explanation of the rationale: the current code only resolves the
final symlink in the path, ie only starts _if the file that sources
hadoop-config.sh is a symlink_, where the patch here unconditionally resolves
all the symlinks. So the current code sometimes does, and sometimes does not,
fully resolve links in the path to HADOOP_HOME. Therefore the current code
should be agnostic as to whether you fully resolve HADOOP_HOME (ie it should be
OK to always resolve the symlinks).
I think it makes sense to resolve links (and this probably was the intent of
the original symlink resolution code) as that way the logic is uniform, ie the
way HADOOP_HOME is determined is the same regardless of whether
hadoop-config.sh is sourced from a symlink.
> Fix link resolution logic in hadoop-config.sh
> ---------------------------------------------
>
> Key: HADOOP-7089
> URL: https://issues.apache.org/jira/browse/HADOOP-7089
> Project: Hadoop Common
> Issue Type: Bug
> Components: scripts
> Reporter: Eli Collins
> Assignee: Eli Collins
> Priority: Minor
> Attachments: hadoop-7089-1.patch, hadoop-7089-1.patch,
> hadoop-7089-2.patch
>
>
> The link resolution logic in bin/hadoop-config.sh fails when when executed
> via a symlink, from the root directory. We can replace this logic with cd -P
> and pwd -P, which should be portable across Linux, Solaris, BSD, and OSX.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
