Ingest data from Logstash to Azure Data Explorer

Important

This connector can be used in Real-Time Intelligence in Microsoft Fabric. Use the instructions in this article with the following exceptions:

Logstash is an open source, server-side data processing pipeline that ingests data from many sources simultaneously, transforms the data, and then sends the data to your favorite "stash". In this article, you'll send that data to Azure Data Explorer, which is a fast and highly scalable data exploration service for log and telemetry data. You'll initially create a table and data mapping in a test cluster, and then direct Logstash to send data into the table and validate the results.

Note

This connector currently supports only json data format.

Prerequisites

Create a table

After you have a cluster and a database, it's time to create a table.

  1. Run the following command in your database query window to create a table:

    .create table logs (timestamp: datetime, message: string)
    
  2. Run the following command to confirm that the new table logs has been created and that it's empty:

    logs
    | count
    

Create a mapping

Mapping is used by Azure Data Explorer to transform the incoming data into the target table schema. The following command creates a new mapping named basicmsg that extracts properties from the incoming json as noted by the path and outputs them to the column.

Run the following command in the query window:

.create table logs ingestion json mapping 'basicmsg' '[{"column":"timestamp","path":"$.@timestamp"},{"column":"message","path":"$.message"}]'

Install the Logstash output plugin

The Logstash output plugin communicates with Azure Data Explorer and sends the data to the service. For more information, see Logstash plugin.

In a command shell, navigate to the Logstash root directory, and then run the following command to install the plugin:

bin/logstash-plugin install logstash-output-kusto

Configure Logstash to generate a sample dataset

Logstash can generate sample events that can be used to test an end-to-end pipeline. If you're already using Logstash and have access to your own event stream, skip to the next section.

Note

If you're using your own data, change the table and mapping objects defined in the previous steps.

  1. Edit a new text file that will contain the required pipeline settings (using vi):

    vi test.conf
    
  2. Paste the following settings that will tell Logstash to generate 1000 test events:

    input {
        stdin { }
        generator {
            message => "Test Message 123"
            count => 1000
        }
    }
    

This configuration also includes the stdin input plugin that will enable you to write more messages by yourself (be sure to use Enter to submit them into the pipeline).

Configure Logstash to send data to Azure Data Explorer

Paste the following settings into the same config file used in the previous step. Replace all the placeholders with the relevant values for your setup. For more information, see Creating a Microsoft Entra Application.

output {
    kusto {
            path => "/tmp/kusto/%{+YYYY-MM-dd-HH-mm-ss}.txt"
            ingest_url => "https://ingest-<cluster name>.kusto.windows.net/"
            app_id => "<application id>"
            app_key => "<application key/secret>"
            app_tenant => "<tenant id>"
            database => "<database name>"
            table => "<target table>" # logs as defined above
            json_mapping => "<mapping name>" # basicmsg as defined above
    }
}
Parameter Name Description
path The Logstash plugin writes events to temporary files before sending them to Azure Data Explorer. This parameter includes a path where files should be written and a time expression for file rotation to trigger an upload to the Azure Data Explorer service.
ingest_url The Kusto endpoint for ingestion-related communication.
app_id, app_key, and app_tenant Credentials required to connect to Azure Data Explorer. Be sure to use an application with ingest privileges.
database Database name to place events.
table Target table name to place events.
json_mapping Mapping is used to map an incoming event json string into the correct row format (defines which property goes into which column).

Run Logstash

We're now ready to run Logstash and test our settings.

  1. In a command shell, navigate to the Logstash root directory, and then run the following command:

    bin/logstash -f test.conf
    

    You should see information printed to the screen, and then the 1000 messages generated by our sample configuration. At this point, you can also enter more messages manually.

  2. After a few minutes, run the following Data Explorer query to see the messages in the table you defined:

    logs
    | order by timestamp desc
    
  3. Select Ctrl+C to exit Logstash

Clean up resources

Run the following command in your database to clean up the logs table:

.drop table logs