Producing dashboards with GrimoireELK

GrimoireELK is an interim system oriented to produce Kibana-based dashboards with Perceval and friends. It provides a Python module (grimoire_elk) with facilities for driving Perceval, enriching data, and uploading / downloading it to / from ElasticSearch. It includes as well some scripts (mainly p2o.py) to retrieve data from repositories related to software development, and produce everything needed to have Elasticsearch indexes that provide the data for a nice Kibiter/Kibana dashboard.

In summary, p2o.py:

• drives Perceval to retrieve data from repositories,
• uploads the resulting data as raw indexes (collections of JSON documents) to ElasticSearch
• enriches those raw indexes (produce new data with fields suitable to be used by Kibana dashboards)
• uploads that resulting data as enriched indexes to ElasticSearch

In addition, kidash, available in the Python package of the same name, can upload dashboard definitions (including visualizations, searches and everything needed by them) to produce a Kibiter/Kibana dashboards.

In this chapter we will explore how to use these tools to produce complete Kibiter/Kibana-based dashboards. Before following to the rest of the chapter, ensure that both grimoire-elk and kidash Python packages are installed (see Installing GrimoireLab).