juju deploy kibana
Your Window into the Elastic Stack Read more
Discuss this charm
Share your thoughts on this charm with the community on discourse.
This charm provides Kibana v5.x from https://www.elastic.co/products/kibana. Kibana lets you visualize your Elasticsearch data and navigate the Elastic Stack.
Deploy and expose the Kibana web interface with:
juju deploy kibana juju expose kibana
Kibana is not much use on its own; you'll probably want the full stack:
juju deploy ~elasticsearch-charmers/bundle/elk-stack juju expose kibana
Browse to http://ip-address to begin searching.
The Kibana source tarball will be downloaded and verified according to the configured values in 'kibana_source' and 'kibana_source_checksum', which defaults to:
It will, obviously, require internet access on deployed nodes.
This behavior can be bypassed by providing
payload prior deployment. The given payload will be used, without any
checksum verification, and unpacked to
Listens on localhost:5601. Uses nginx to redirect ip-addr:80 to localhost:5601. This should allow for haproxy etc to be used in front. You could also configure apache reverse proxy in front if you want SSL / .htaccess
supports multiple ES servers in cluster. This should help with balancing the load on ES and dealing with failure.
see logstash-indexer charm's README.md file for usage examples.
Deploy / Add Dashboards
The Kibana charm ships with an action to deploy dashboards into the Kibana instance. You will however, have to have them loaded into the charm before the user deploys it, which means a pull request to the upstream charm to include you're awesome dashboard for everyone!
For example, to deploy the included 'beats' dashboard with < Juju 2.0:
juju action do kibana/0 load-dashboard dashboard=beats
For 2.0 +
juju run-action kibana/0 load-dashboard dashboard=beats
This invokes the action, which in turn looks in the 'dashboards' directory for
Which is the repository layout adopted by Elasticsearch for their own
dashboards. To include your own, simply make a directory, mirror the
script to curl to your elasticsearch instance (by default: localhost:9200 will
work as its proxying the connection to your ES cluster) and it will declare any
indexes as well as load any dashboard JSON definitions you include. For more
information, consult the
Custom dashboards can be added by uploading them via
juju scp my-dashboard/ kibana/0:dashboards/