Jason Lowe commented on YARN-7677:

I realize now that the theoretical example cannot work in practice.  In order 
for there to be a "hook" variable for the user to leverage, the variable would 
need to have escaped variable expansion by the shell when it was originally 
set.  The variable would need to be set in the NM's environment like, 
JAVA_HOME="/some/node/path/\$JDKVER".  While that could be a valid path for the 
user when it is expanded in the container launch script, it is not a valid 
setting for JAVA_HOME in the nodemanager itself.  NM whitelist variables are 
going to be variables coming from a shell environment and not from XML property 
settings, so it's highly unlikely they will retain unexpanded variable 

In short, I'm cool with simply placing the NM whitelist variables first and 
simplifying YARN-5714 to list the variables in the launch script in the order 
they appear in their corresponding configuration properties.  My apologies for 
the detour.

> Docker image cannot set HADOOP_CONF_DIR
> ---------------------------------------
>                 Key: YARN-7677
>                 URL: https://issues.apache.org/jira/browse/YARN-7677
>             Project: Hadoop YARN
>          Issue Type: Bug
>    Affects Versions: 3.0.0
>            Reporter: Eric Badger
>            Assignee: Jim Brennan
>            Priority: Major
>         Attachments: YARN-7677.001.patch, YARN-7677.002.patch
> Currently, {{HADOOP_CONF_DIR}} is being put into the task environment whether 
> it's set by the user or not. It completely bypasses the whitelist and so 
> there is no way for a task to not have {{HADOOP_CONF_DIR}} set. This causes 
> problems in the Docker use case where Docker containers will set up their own 
> environment and have their own {{HADOOP_CONF_DIR}} preset in the image 
> itself. 

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org

Reply via email to