Nagios Logs are too noisy

[Swann Croiset]

Nagios is one of the most verbose program within the cluster. This uselessly fills Elasticsearch because the majority of log are the result of the stacklight collector which sends service health status (EXTERNAL COMMAND: PROCESS_SERVICE_CHECK_RESULT).

Because Nagios is configured to use both Syslog and /var/nagios/nagios.log and the /var/log/messages is collected by default by Heka.

There are 2 choices to fix this:

1/ disable the Nagios option "use_syslog=1"
2/ disable useless log (log_external_commands=0 and log_external_commands=0).

I preconise to use #2 to keep other logs into Elasticsearch (less verbose and potentially useful).

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/274047

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/274047
Committed: https://git.openstack.org/cgit/openstack/fuel-plugin-lma-infrastructure-alerting/commit/?id=6a1f72fb2d3106c95e4f7b890ac38cca05e4bfa9
Submitter: Jenkins
Branch: master

commit 6a1f72fb2d3106c95e4f7b890ac38cca05e4bfa9
Author: Swann Croiset <email address hidden>
Date: Fri Jan 29 13:40:30 2016 +0100

    Avoid useless Nagios logs

    Change-Id: I9dd49f1df5fa060416b33ec1119c7dd3752889f8
    Fixes-bug: #1539567