Kafka moduleedit

The kafka module collects and parses the logs created by Kafka.

The module has additional support for parsing thread ID from logs.

When you run the module, it performs a few tasks under the hood:

  • Sets the default paths to the log files (but don’t worry, you can override the defaults)
  • Makes sure each multiline log event gets sent as a single event
  • Uses an Elasticsearch ingest pipeline to parse and process the log lines, shaping the data into a structure suitable for visualizing in Kibana
  • Deploys dashboards for visualizing the log data

Read the quick start to learn how to configure and run modules.


The kafka module was tested with logs from versions 0.9, 1.1.0 and 2.0.0.

Configure the moduleedit

You can further refine the behavior of the kafka module by specifying variable settings in the modules.d/kafka.yml file, or overriding settings at the command line.

You must enable at least one fileset in the module. Filesets are disabled by default.

The following example shows how to set paths in the modules.d/kafka.yml file to override the default paths for logs:

- module: kafka
    enabled: true
      - "/path/to/logs/controller.log*"
      - "/path/to/logs/server.log*"
      - "/path/to/logs/state-change.log*"
      - "/path/to/logs/kafka-*.log*"

To specify the same settings at the command line, you use:

-M "kafka.log.var.paths=[/path/to/logs/controller.log*, /path/to/logs/server.log*, /path/to/logs/state-change.log*, /path/to/logs/kafka-*.log*]"

Variable settingsedit

Each fileset has separate variable settings for configuring the behavior of the module. If you don’t specify variable settings, the kafka module uses the defaults.

For advanced use cases, you can also override input settings. See Override input settings.

When you specify a setting at the command line, remember to prefix the setting with the module name, for example, kafka.log.var.paths instead of log.var.paths.

log fileset settingsedit


The path to your Kafka installation. The default is /opt. For example:

- module: kafka
    enabled: true
    var.kafka_home: /usr/share/kafka_2.12-2.4.0
An array of glob-based paths that specify where to look for the log files. All patterns supported by Go Glob are also supported here. For example, you can use wildcards to fetch all files from a predefined level of subdirectories: /path/to/log/*/*.log. This fetches all .log files from the subfolders of /path/to/log. It does not fetch log files from the /path/to/log folder itself. If this setting is left empty, Filebeat will choose log paths based on your operating system.

Time zone supportedit

This module parses logs that don’t contain time zone information. For these logs, Filebeat reads the local time zone and uses it when parsing to convert the timestamp to UTC. The time zone to be used for parsing is included in the event in the event.timezone field.

To disable this conversion, the event.timezone field can be removed with the drop_fields processor.

If logs are originated from systems or applications with a different time zone to the local one, the event.timezone field can be overwritten with the original time zone using the add_fields processor.

See Processors for information about specifying processors in your config.

Example dashboardedit

This module comes with a sample dashboard to see Kafka logs and stack traces.

filebeat kafka logs overview


For a description of each field in the module, see the exported fields section.