Docker Community Forums

Share and learn in the Docker community.

Monitoring nodes


(Kim Neunert) #1

Hi,
i’m searching for a convenient way of monitoring my docker-cloud nodes. How are they utilized? How is memory-consumption? What about file-system?

Usually i do that with collectd.

In the monitoring-section of the jumpstarts, there is a datadog and a newRelic “bridge-image”, however i’m not using that stuff, but hosting a graphite/Grafana on docker-cloud. So regarding that basic setup some questions:

  • does it make sense to roll my own collectd inside a docker-container pushing data to my graphite?
  • I would need to run that instance in priviledged-mode, right? Anything else to take care?
  • Is there some other cool option i’m missing?
  • Wouldn’t it be cool if you would provide a standard-solution in the jumpstart-section :slight_smile:

Thanks

Kim


(Chris Jones) #2

I am using site24x7.com linux node monitors. Grabs all your CPU, RAM, Disk, Network stats and also gives good status on the Docker engine including all those same stats per-container.


(Sematext Group Inc.) #3

You could use sematext/sematext-agent-docker
https://hub.docker.com/r/sematext/sematext-agent-docker/

See the “deploy to cloud” button to get your stackfile. Just enter Applications tokens for SPM and Logsene, generated in Sematext UI.
It monitors Host metrics, Docker Events, Container Metrics and can collect all container logs (including log parsing for many log formats and GeoIP enrichment and Kibana…). Look at this to get started quickly: https://sematext.com/blog/2015/12/15/docker-logging-partner-sematext-logsene/


(Geoff Bowers) #4

I use the Datadog agent; free for up to five nodes.

You can add a utility stack to your Docker Cloud set up that automatically deploys the monitoring agent as a container on every node:

datadog:
  image: 'datadog/docker-dd-agent:latest'
  deployment_strategy: every_node
  environment:
    - API_KEY=***YOUR-API-KEY***
    - HOSTNAME=$DOCKERCLOUD_NODE_HOSTNAME
  privileged: true
  restart: on-failure
  volumes:
    - '/var/run/docker.sock:/var/run/docker.sock'
    - '/proc:/host/proc:ro'
    - '/sys/fs/cgroup:/host/sys/fs/cgroup:ro'

You’ll have to set up a trial account at https://www.datadoghq.com/ to get your API key. The service is free indefinitely for less than 5 nodes.

Hope that helps :unicorn:


(Ykyuen) #5

Maybe this tool could help.