Skip to content

Latest commit

 

History

History
87 lines (65 loc) · 4.85 KB

discovery.md

File metadata and controls

87 lines (65 loc) · 4.85 KB
Raw

Flex Auto Container Discovery

⚠️ Note: this is deprecated functionality that is still provided for backwards compatibility. We encourage you to use the improved, fully-supported container auto-discovery mechanism for On-Host Integrations.

Flex has the capability to auto discovery containers in your surrounding environment, and dynamically monitor them regardless of changing IP addresses and ports.

This requires access to /var/run/docker.sock, same as the New Relic Infrastructure Agent, so it is convenient to include Flex in the newrelic/infrastructure image or your k8s daemon set.

Configuration

  -container_discovery
        Enable container auto discovery
  -container_discovery_dir string
        Set directory of auto discovery config files (default "flexContainerDiscovery/")
  -docker_api_version string
        Force Docker client API version

Versions

V2 Container Discovery

V2 simplifies discovery by allowing everything to be defined by your Flex configuration file: you can place config files within your standard flexConfigs folder. The container_discovery flag still needs to be enabled to true.

These configs are prefixed with cd- (for example, flexConfigs/cd-redis.yml). Use ${auto:ip}, ${auto:host} and ${auto:port} as your substitution flags.

In this example, container_discovery is enabled and a YAML file called cd-redis.yml is placed in your flexConfigs folder:

name: redis
apis: 
  - name: redis
    entity: ${auto:ip}
    commands: 
      - dial: ${auto:ip}:${auto:port}
        run: "info\r\n"
        split_by: ":"

This tells Flex to look for containers with the name that contains redis by default, as the name is taken from the text after cd-.

To override this behavior or further configure how you would like to target containers, consider the example below (all are optional parameters):

name: redis
container_discovery:
  mode: contains # default: contains, other options: prefix, suffix, contains, regex
  type: container # default: container, other options: image
  target: some-other-container-name # default: text after the "cd-" in the config file name
  ip_mode: private # default: automatically set, other options: public, private
apis: 
  - name: redis
    entity: ${auto:ip}
    commands: 
      - dial: ${auto:ip}:${auto:port}
        run: "info\r\n"
        split_by: ":"

V1 Container Discovery

In a Kubernetes environment, service environment variables can be passed to every container, so your Flex container can access them. You can use environment variable substitution in a normal config file without using the container discovery feature. Environment variables can be passed to config files by using a double dollar sign (for example, $$MY_ENVIRONMENT_VAR).

  1. Add a label or annotation that contains the keyword - flexDiscovery.
  2. For Kubernetes, add it as an environment variable.
  3. Add a flex discovery configuration to that same label or annotation, for example, flexDiscoveryRedis="t=redis,c=redis,tt=img,tm=contains

You could have varying configs on one container, like flexDiscoveryRedis1, flexDiscoveryZookeeper, etc.

Flex Container Discovery Configs are placed within "flexContainerDiscovery/" directory. For an example see flexContainerDiscovery/redis.yml.

Use ${auto:host} and ${auto:port} anywhere in your config: these are dynamically substituted per container discovered, and makes it possible to have multiple containers re-use the same config with different ip/port configurations.

You can also opt to apply the same environment variable(s) to the Flex container itself, which allows you to monitor other containers without modifying your existing environment. If you have configs that overlap between something set on a container, and something set on the Flex container itself, the Flex container will take precedence.

Flex uses the environment variables that are presented to the Docker Engine; if you have modified the discovery environment variable(s) after deployment for whatever reason, the Docker Engine will be unaware of these changes.

Flex Discovery Configuration Parameters

  • tt=targetType - Target type (default is img)
  • t=target - Keyword to target based on the targetType (for example, redis)
  • tm=targetMode - Prefix or regex to match the target (default is contains)
  • c=config - Config file used to create the dynamic configs (defaults to the target value)
  • p=port - Target port
  • ip=ipMode - IP mode. Default is private - it can be set to public

If a config is nil, use the target (t), as the YAML file to look up. For example, if target (t) = redis, look up the config (c) redis.yml if the config is not set.