Acadock Monitoring - Docker container monitoring v1.2.1

This webservice provides live data on Docker containers. It takes data from the Linux kernel control groups and from the namespace of the container and expose them through a HTTP API.

This software is currently used in production.

Configuration

From environment

• PORT: port to bind (4244 by default)
• DOCKER_URL: docker endpoint (http://127.0.0.1:4243 by default)
• REFRESH_TIME: number of second between CPU/net refresh (1 by default)
• PROC_DIR: mountpoint for procfs (default to /proc)
• CGROUP_DIR: mountpoint of cgroups (default to /sys/fs/cgroup)
• CGROUP_SOURCE: "docker" or "systemd" (docker by default) docker: /sys/fs/cgroup/:cgroup/memory/docker systemd: /sys/fs/cgroup/:cgroup/memory/system.slice/docker-#{id}.slice
• DEBUG: output of debugging information (default "false", switch to "true" to enable)

Docker

Run from docker:

docker run -v /sys/fs/cgroup:/host/cgroup:ro         -e CGROUP_DIR=/host/cgroup \
           -v /proc:/host/proc:ro                    -e PROC_DIR=/host/proc \
           -v /var/run/docker.sock:/host/docker.sock -e DOCKER_URL=unix:///host/docker.sock \
           -p 4244:4244 --privileged --pid=host --network=host \
           -d scalingo/acadock-monitoring

• --pid=host: The daemon has to find the real /proc/#{pid}/ns directory to enter a namespace
• --network=host: Acadock should in the host namespace to access other containers network namespaces (for network metrics)
• --privileged: Acadock has to enter the other containers namespaces

API

• Memory consumption

  GET /containers/:id/mem

  Return 200 OK Content-Type: application/json

{
  "mem_usage": 123,
  "mem_limit": 5000,
  "max_mem_usage": 500,
  "max_swap_mem_usage": 200,
  "swap_mem_usage": 145,
  "swap_mem_limit": 1000
}

• CPU usage (percentage)

  Return 200 OK Content-Type: text/plain GET /containers/:id/cpu

• Network usage (bytes and percentage)

  Return 200 OK Content-Type: application/json GET /containers/:id/net

• Mem+CPU+Network for a container

  Return 200 OK Content-Type: application/json GET /containers/:id/usage

• Mem+CPU+Network for all containers

  Return 200 OK Content-Type: application/json GET /containers/usage

Release a New Version

Bump new version number in:

• CHANGELOG.md
• README.md

Commit, tag and create a new release:

version="1.2.1"

git switch --create release/${version}
git add CHANGELOG.md README.md
git commit -m "Bump v${version}"
git push --set-upstream origin release/${version}
gh pr create --reviewer=EtienneM --fill-first

Once the pull request merged, you can tag the new release.

git tag v${version}
git push origin master v${version}
gh release create v${version} --generate-notes --prerelease
gh release view v${version} --web

Build locally the archives. You can use the following command, it will automatically use the version of the last tag created.

goreleaser release --skip=publish,announce,sign --clean

On the web interface, unset the pre-release checkbox, check the "Set as the latest release" checkbox, and upload the archives in the dist folder.

Last, update the default version installed in the cookbook.