Centralized Logging

Semantic Logger directly supports a centralized logging system consisting of:


For simplicity these installation notes use homebrew on Mac OSX. Following the product links for information on installing on other platforms.


Install Elasticsearch:

brew install elasticsearch

Follow the on-screen instructions to auto-start elasticsearch.


Install Kibana:

brew install kibana

Follow the on-screen instructions to auto-start kibana.

Semantic Logger

Configure the application to forward all log messages to Elasticsearch.

Add the following code to a Rails initializer if running rails with the rails_semantic_logger gem, or add the code directly to your application:

  appender: :elasticsearch,
  url:      "http://localhost:9200"

Restart the application and let it run to generate several log messages.

Check the text file for any errors. For example: log/development.log

Configure Kibana

Assuming Kibana was installed locally with its default port. Open a browser for Kibana http://localhost:5601/.

Add a new index pattern to view in Kibana:

Select Settings -> Indices -> Add New

Select checkbox Index contains time-based events.

Enter Index name: semantic_logger-*

Give Kibana a moment to scan the index, then select Time-field name and choose timestamp, then press the Create button.

View logs in browser

If not already open, open Kibana in the browser http://localhost:5601/.

Select Discover.

The log entries generated by the application should now be displayed. If not, change the time interval on the top right of the screen.

On the left under Available Fields add the following columns to the view:

Example Searches

Some example searches:

Only show error level messages:

level: error

Only show messages from host mymachine

host: mymachine

Search logging tags for the string 17262353

tags: 17262353

Other solutions

Semantic Logger also supports several other centralized logging solutions:

See appenders to configure appenders to forward log messages to the centralized logging solutions / log aggregators.

Next: Programmers Guide ==>