axibase/collectd repository overview

⁠Collectd Docker Image

This image provides a Dockerized collectd configuration to gather operating system statistics from the underlying Docker host instead of a container where it's launched.

The collectd agent started within the container is automatically configured to send metrics into an Axibase Time Series Database instance using the write_atsd⁠ plugin.

The target ATSD instance is specified in the --atsd-url argument.

⁠Prepare Image

⁠Docker Hub

• Use an existing image published on Docker Hub⁠

docker pull axibase/collectd

⁠Build Image from Sources

• Download Dockerfile⁠ to a Docker host connected to hub.docker.com

• Build image

docker build -t axibase/collectd .

⁠Download Image

• Download a pre-built image file from axibase.com⁠

• Import the image into the Docker host

docker load < docker-axibase-collectd.tar.gz

⁠Start Container

⁠Basic Configuration

• Start Docker container

docker run -d -v /:/rootfs:ro --pid=host --net=host \
    --name=collectd axibase/collectd \
    --atsd-url=tcp://atsd_host:tcp_port \
    --fqdn=false

• The fqdn option is used to set Collectd FQDNLookup⁠ setting. It controls how a hostname is chosen. When enabled, the hostname of the node is set to the fully qualified domain name

⁠lvs Configuration

This configuration reads lvs command output and therefore must be launched with elevated privileges. The collected data is useful when docker is configured in direct-lvm⁠ mode.

docker run -d -v /:/rootfs:ro --privileged=true \
    --pid=host --net=host \
    --name=collectd axibase/collectd \
    --atsd-url=tcp://atsd_host:tcp_port \
    --fqdn=false \
    --lvs

The lvs output is processed with lvs.sh script, invoked by the collectd's native exec⁠ plugin. The lvs.sh script reads the output into a tabular structure and generates PUTVAL commands containing LVM statistics for each logical volume:

PUTVAL nurswgdkr001/exec-lvs-data%/gauge-volume_group=vg0;logical_volume=thinpool N:4.52

⁠Credits

• Carles Amigó⁠ for the idea to re-mount rootfs.

⁠Troubleshooting

If there are some problems grep collectd logs for errors to find out the reason:

docker exec -it collectd cat /var/log/collectd.log | grep error