[
https://issues.apache.org/jira/browse/WHIRR-695?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13569620#comment-13569620
]
Steve Loughran commented on WHIRR-695:
--------------------------------------
I think this'll have to be a WONTFIX, as it's jps's fault.
# {{jps}} only does local user, not {{ps -ef}} -read the small print on the man
page. As a result, you don't find java processes started by other users, which
is exactly what whirr does.
# it also used to depend on saving pid information to /tmp, which means when
that dir got purged of old files, the pid list would go away, and jps would
stop finding them.
{{ps | grep}} are more reliable, I'm afraid
> Hadoop ecosystem tools should show up when 'jps' is run at the console
> ----------------------------------------------------------------------
>
> Key: WHIRR-695
> URL: https://issues.apache.org/jira/browse/WHIRR-695
> Project: Whirr
> Issue Type: Bug
> Reporter: Eli Reisman
> Priority: Minor
>
> When running the hadoop-oriented service recipes on Whirr, my Hadoop tools do
> not show up on 'jps' but do show up on 'ps -ef | grep java' as running when I
> SSH into my cluster nodes.
> For people who are new to EC2/cloud cluster configuration, it is reassuring
> to run jps and see you framework processes running and good to go, just like
> in the Hadoop cluster config instructions.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
