The OSS Elastic Beats are a great group of log shippers. They are cross-platform, lightweight, and can ship data to a number of tools including Humio as long as you stick to the OSS builds.
Starting from version 6.7.0 of the libbeat only the OSS versions can ship to Humio. The non-OSS Beats check that the server is a licensed elastic server due to this change to the beats client library: “Check license x-pack”.
There are currently five official Beats. The Elastic documentation site and Humio’s documentation offer resources that describe how to use each of them.
Filebeat—Ships regular log files.
Metricbeat—Ships metrics from your OS and common services.
Packetbeat—Analyzes network packets and common protocols like HTTP.
Winlogbeat—Ships Windows event logs
All Beats are built using the libbeat library and share output configuration. Humio supports parts of the ElasticSearch ingest API, so to send data from Beats to Humio, you use the ElasticSearch output (the documentation is identical for all Beats).
You can use the following
elasticsearch output configuration template:
output: elasticsearch: hosts: ["$YOUR_HUMIO_URL/api/v1/ingest/elastic-bulk"] username: any-organization password: $INGEST_TOKEN
$YOUR_HUMIO_URL variable is the base URL of your Humio server whether cloud or self-hosted. The
$INGEST_TOKEN variable is the ingest token for your repository (a string such as fS6Kdlb0clqe0UwPcc4slvNFP3Qn1COzG9DEVLw7v0Ii).
Starting from version 7.7.0, all beats clients default to using the Elasticsearch index template feature which does not work with Humio. This needs to be disabled in the configuration:
output: elasticsearch: setup.template.enabled: false
To optimize performance for the data volumes you want to send, and to keep shipping latency down, change the default settings for
flush_interval. Don’t raise
bulk_max_size much: 100–300 is the appropriate range. While doing so may increase throughput of ingest it has a negative impact on search performance of the resulting events in Humio.
All Beats also have a
fields section in their configuration. You can add fields to all events by specifying them in the
fields: service: user-service datacenter: dc-a
Fields can be turned into tags by including a
@tags field that lists the names of fields to turn into tags. This applies to fields both from the fields sections and from the events being shipped. Refer to datasources for information on tags.
For on-premise installations of Humio, it is possible to specify a global parser that can ingest data into multiple repositories.