Analyse ARender performances in ELK stack.
ARender returns statistics on its usage, like the loading time of a document and the opened document type. Information is stored in log files. It is possible to analyse these logs with the ELK stack.
You need to use following applications from the ELK stack : Elastisearch and Kibana :
- Elasticsearch : stores and indexes data. It is a NoSQL base allowing to manage big data.
- Kibana : is a Web interface allowing to search and visualize graphical data.
In Kibana data are shown in a graphical user friendly way. Graphics creation is simple and there are lots of customization possible.
See below examples :
Average document loading time :
MIME Type distribution :
To import visualizations and dashboards with .ndjson files
- Kibana version 7.2.0 and above
- ElasticSearch version 7.2.0 and above
To import visualizations and dashboards with .json files
- Kibana version 5.x and above
- ElasticSearch version 5.x and above
Follow official documentation to install ElasticSearch : https://www.elastic.co/guide/en/elasticsearch/reference/current/
Follow official documentation to start ElasticSearch : https://www.elastic.co/guide/en/elasticsearch/reference/current/starting-elasticsearch.html
Follow official documentation to install Kibana : https://www.elastic.co/guide/en/kibana/current/install.html
Follow official documentation to start Kibana : https://www.elastic.co/guide/en/kibana/current/start-stop.html
Kibana configuration prerequisites
Before configuring Kibana, you need to open at least one document in ARender in order to have indexes created in Elasticsearch.
Creation of index pattern
Open Kibana in your browser. Local URL : http://localhost:5601/app/kibana
Go to the Management section :
Go to the Index Patterns :
Add the 2 indexes :
Import dashboard example
Go to Saved Objects :
For version allowing .ndjson file import (Since Kibana 7.2.0):
For version only allowing .json file import :
Open dashboard :