[
https://issues.apache.org/jira/browse/AMBARI-21475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16091253#comment-16091253
]
Rob Phipps commented on AMBARI-21475:
-------------------------------------
I have done a bit more digging, and have traced the error back to the
assumption in {{/usr/lib/ambari-agent/lib/resource_management/core/shell.py}}
that a non-zero exit code means a task failed, and that {{_call}} in this file
is invoked with {{throw_on_failure}} set to true. I am assuming if it was
possible to pass back false to that function from earlier on, but looking at
the chain of files that {{service elasticsearch status}} has to pass through to
get to the {{shell.py}} script, that may be easier said than done. Any ideas?
> Elasticsearch systemd support
> -----------------------------
>
> Key: AMBARI-21475
> URL: https://issues.apache.org/jira/browse/AMBARI-21475
> Project: Ambari
> Issue Type: Bug
> Affects Versions: 2.5.3
> Environment: CentOS 7 x64
> Reporter: Rob Phipps
> Attachments: es-stop-fail.png
>
>
> Hi,
> It appears the Elasticsearch plugin can't recognise the responses from the
> new systemd service control messages, as despite the logs showing the service
> has been stopped successfully, the task is failing in the Ambari UI.
> Elasticsearch plugin version 2.3.3
> Log:
> !es-stop-fail.png|large!
> Thanks,
> Rob
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
