The Elasticsearch cluster isn't monitored anymore after a scale down of the environment

[Simon Pasquier]

Tested with MOS 8 (ISO #452).

Steps to reproduce

1. Create an environment with 3 Elasticsearch nodes and at least one infrastructure alerting node.
2. Check that the Elasticsearch cluster is monitored in Nagios
3. Remove one ES node from the environment.
4. The Elasticsearch cluster isn't present in Nagios anymore.

Expected result

The Elasticsearch cluster is monitored by Nagios.

[Simon Pasquier (simon-pasquier)]

We need to use hiera_hash() instead of hiera() at [1].

[1] https://github.com/openstack/fuel-plugin-lma-collector/blob/58d7984813be5e29b0004b597331911116b476cf/deployment_scripts/puppet/manifests/hiera_override.pp#L15

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/277393

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/277393
Committed: https://git.openstack.org/cgit/openstack/fuel-plugin-lma-collector/commit/?id=be85d7ebacb1977b09fb292eb5e2dae353ba5cd7
Submitter: Jenkins
Branch: master

commit be85d7ebacb1977b09fb292eb5e2dae353ba5cd7
Author: Simon Pasquier <email address hidden>
Date: Mon Feb 8 14:43:20 2016 +0100

    Always use hiera_hash() for the lma_collector key

    Since 'lma_collector' may be spread across several Hiera files, we
    should always use the hiera_hash() function to access it.

    Change-Id: Iadeda5f777253e1a5f8c408b527b381ccdf0df6b
    Closes-Bug: #1538070