Hints based autodiscover

Filebeat supports autodiscover based on hints from the provider. The hints system looks for hints in Kubernetes Pod annotations or Docker labels that have the prefix co.elastic.logs. As soon as the container starts, Filebeat will check if it contains any hints and launch the proper config for it. Hints tell Filebeat how to get logs for the given container. By default logs will be retrieved from the container using the filestream input. You can use hints to modify this behavior. This is the full list of supported hints:

Filebeat gets logs from all containers by default, you can set this hint to false to ignore the output of the container. Filebeat won’t read or send logs from it. If default config is disabled, you can use this annotation to enable log retrieval only for containers with this set to true. If you are aiming to use this with Kubernetes, have in mind that annotation values can only be of string type so you will need to explicitly define this as "true" or "false" accordingly.

Multiline settings. See Multiline messages for a full list of all supported options.

JSON settings. See ndjson for a full list of all supported options.

For example, the following hints with json options:

co.elastic.logs/json.message_key: "log"
co.elastic.logs/json.add_error_key: "true"

will lead to the following input configuration:

filestream

parsers:
  - ndjson:
      message_key: "log"
      add_error_key: "true"

A list of regular expressions to match the lines that you want Filebeat to include. See Inputs for more info.

A list of regular expressions to match the lines that you want Filebeat to exclude. See Inputs for more info.

Instead of using raw docker input, specifies the module to use to parse logs from the container. See Modules for the list of supported modules.

When module is configured, map container logs to module filesets. You can either configure a single fileset like this:

co.elastic.logs/fileset: access

Or configure a fileset per stream in the container (stdout and stderr):

co.elastic.logs/fileset.stdout: access
co.elastic.logs/fileset.stderr: error

When an entire input/module configuration needs to be completely set the raw hint can be used. You can provide a stringified JSON of the input configuration. raw overrides every other hint and can be used to create both a single or a list of configurations.

co.elastic.logs/raw: "[{\"containers\":{\"ids\":[\"${data.container.id}\"]},\"multiline\":{\"negate\":\"true\",\"pattern\":\"^test\"},\"type\":\"docker\"}]"

Define a processor to be added to the Filebeat input/module configuration. See Processors for the list of supported processors.

If processors configuration uses list data structure, object fields must be enumerated. For example, hints for the rename processor configuration below

processors:
  - rename:
      fields:
        - from: "a.g"
          to: "e.d"
      fail_on_error: true

will look like:

co.elastic.logs/processors.rename.fields.0.from: "a.g"
co.elastic.logs/processors.rename.fields.1.to: "e.d"
co.elastic.logs/processors.rename.fail_on_error: 'true'

If processors configuration uses map data structure, enumeration is not needed. For example, the equivalent to the add_fields configuration below

processors:
  - add_fields:
      target: project
      fields:
        name: myproject

is

co.elastic.logs/processors.1.add_fields.target: "project"
co.elastic.logs/processors.1.add_fields.fields.name: "myproject"

In order to provide ordering of the processor definition, numbers can be provided. If not, the hints builder will do arbitrary ordering:

co.elastic.logs/processors.1.dissect.tokenizer: "%{key1} %{key2}"
co.elastic.logs/processors.dissect.tokenizer: "%{key2} %{key1}"

In the above sample the processor definition tagged with 1 would be executed first.

Define an ingest pipeline ID to be added to the Filebeat input/module configuration.

co.elastic.logs/pipeline: custom-pipeline

When hints are used along with templates, then hints will be evaluated only in case there is no template’s condition that resolves to true. For example:

filebeat.autodiscover.providers:
  - type: docker
    hints.enabled: true
    hints.default_config:
      type: container
      paths:
        - /var/lib/docker/containers/${data.container.id}/*.log
    templates:
      - condition:
          equals:
            docker.container.labels.type: "pipeline"
        config:
          - type: container
            paths:
              - "/var/lib/docker/containers/${data.docker.container.id}/*.log"
            pipeline: my-pipeline

In this example first the condition docker.container.labels.type: "pipeline" is evaluated and if not matched the hints will be processed and if there is again no valid config the hints.default_config will be used.

Kubernetes autodiscover provider supports hints in Pod annotations. To enable it just set hints.enabled:

filebeat.autodiscover:
  providers:
    - type: kubernetes
      hints.enabled: true

You can configure the default config that will be launched when a new container is seen, like this:

filebeat.autodiscover:
  providers:
    - type: kubernetes
      hints.enabled: true
      hints.default_config:
        type: container
        paths:
          - /var/log/containers/*-${data.container.id}.log 1

1. CRI path

You can also disable default settings entirely, so only Pods annotated like co.elastic.logs/enabled: true will be retrieved:

filebeat.autodiscover:
  providers:
    - type: kubernetes
      hints.enabled: true
      hints.default_config.enabled: false

You can annotate Kubernetes Pods with useful info to spin up Filebeat inputs or modules:

annotations:
  co.elastic.logs/multiline.pattern: '^\['
  co.elastic.logs/multiline.negate: true
  co.elastic.logs/multiline.match: after

When a pod has multiple containers, the settings are shared unless you put the container name in the hint. For example, these hints configure multiline settings for all containers in the pod, but set a specific exclude_lines hint for the container called sidecar.

annotations:
  co.elastic.logs/multiline.pattern: '^\['
  co.elastic.logs/multiline.negate: true
  co.elastic.logs/multiline.match: after
  co.elastic.logs.sidecar/exclude_lines: '^DBG'

When a container needs multiple inputs to be defined on it, sets of annotations can be provided with numeric prefixes. If there are hints that don’t have a numeric prefix then they get grouped together into a single configuration.

annotations:
  co.elastic.logs/exclude_lines: '^DBG'
  co.elastic.logs/1.include_lines: '^DBG'
  co.elastic.logs/1.processors.dissect.tokenizer: "%{key2} %{key1}"

The above configuration would generate two input configurations. The first input handles only debug logs and passes it through a dissect tokenizer. The second input handles everything but debug logs.

Hints can be configured on the Namespace’s annotations as defaults to use when Pod level annotations are missing. The resultant hints are a combination of Pod annotations and Namespace annotations with the Pod’s taking precedence. To enable Namespace defaults configure the add_resource_metadata for Namespace objects as follows:

filebeat.autodiscover:
  providers:
    - type: kubernetes
      hints.enabled: true
      add_resource_metadata:
        namespace:
          include_annotations: ["nsannotation1"]

Docker autodiscover provider supports hints in labels. To enable it just set hints.enabled:

filebeat.autodiscover:
  providers:
    - type: docker
      hints.enabled: true

You can configure the default config that will be launched when a new container is seen, like this:

filebeat.autodiscover:
  providers:
    - type: docker
      hints.enabled: true
      hints.default_config:
        type: container
        paths:
          - /var/log/containers/*-${data.container.id}.log 1

1. CRI path

You can also disable default settings entirely, so only containers labeled with co.elastic.logs/enabled: true will be retrieved:

filebeat.autodiscover:
  providers:
    - type: docker
      hints.enabled: true
      hints.default_config.enabled: false

You can label Docker containers with useful info to spin up Filebeat inputs, for example:

co.elastic.logs/module: nginx
co.elastic.logs/fileset.stdout: access
co.elastic.logs/fileset.stderr: error

The above labels configure Filebeat to use the Nginx module to harvest logs for this container. Access logs will be retrieved from stdout stream, and error logs from stderr.

You can label Docker containers with useful info to decode logs structured as JSON messages, for example:

co.elastic.logs/json.keys_under_root: true
co.elastic.logs/json.add_error_key: true
co.elastic.logs/json.message_key: log

Nomad autodiscover provider supports hints using the meta stanza. To enable it just set hints.enabled:

filebeat.autodiscover:
  providers:
    - type: nomad
      hints.enabled: true

You can configure the default config that will be launched when a new job is seen, like this:

filebeat.autodiscover:
  providers:
    - type: nomad
      hints.enabled: true
      hints.default_config:
        type: filestream
        id: ${data.nomad.task.name}-${data.nomad.allocation.id}1
        paths:
          - /opt/nomad/alloc/${data.nomad.allocation.id}/alloc/logs/${data.nomad.task.name}.*

1. unique ID required

You can also disable the default config such that only logs from jobs explicitly annotated with "co.elastic.logs/enabled" = "true" will be collected:

filebeat.autodiscover:
  providers:
    - type: nomad
      hints.enabled: true
      hints.default_config:
        enabled: false
        type: filestream
        id: ${data.nomad.task.name}-${data.nomad.allocation.id}1
        paths:
          - /opt/nomad/alloc/${data.nomad.allocation.id}/alloc/logs/${data.nomad.task.name}.*

1. unique ID required

You can annotate Nomad Jobs using the meta stanza with useful info to spin up Filebeat inputs or modules:

meta {
  "co.elastic.logs/enabled"           = "true"
  "co.elastic.logs/multiline.pattern" = "^\\["
  "co.elastic.logs/multiline.negate"  = "true"
  "co.elastic.logs/multiline.match"   = "after"
}

If you are using autodiscover then in most cases you will want to use the add_nomad_metadata processor to enrich events with Nomad metadata. This example configures Filebeat to connect to the local Nomad agent over HTTPS and adds the Nomad allocation ID to all events from the input. Later in the pipeline the add_nomad_metadata processor will use that ID to enrich the event.

filebeat.autodiscover:
  providers:
    - type: nomad
      address: https://localhost:4646
      hints.enabled: true
      hints.default_config:
        enabled: false 1
        type: filestream
        id: ${data.nomad.task.name}-${data.nomad.allocation.id} 2
        paths:
          - /opt/nomad/alloc/${data.nomad.allocation.id}/alloc/logs/${data.nomad.task.name}.*
        processors:
          - add_fields: 3
              target: nomad
              fields:
                allocation.id: ${data.nomad.allocation.id}

processors:
  - add_nomad_metadata: 4
      when.has_fields.fields: [nomad.allocation.id]
      address: https://localhost:4646
      default_indexers.enabled: false
      default_matchers.enabled: false
      indexers:
        - allocation_uuid:
      matchers:
        - fields:
            lookup_fields:
              - 'nomad.allocation.id'

1. The default config is disabled meaning any task without the "co.elastic.logs/enabled" = "true" metadata will be ignored.
2. Unique ID is required.
3. The add_fields processor populates the nomad.allocation.id field with the Nomad allocation UUID.
4. The add_nomad_metadata processor is configured at the global level so that it is only instantiated one time which saves resources.