Monitors docker events and restarts unhealthy containers.
The healthiness of the container is assessed through configurable healthcheck commands.
docker run -d -v /var/run/docker.sock:/var/run/docker.sock mcasimir/docker-autoheal
Starts your containers with some labels to enable the autoheal:
docker run \
-l com.github.mcasimir.autoheal.check.cmd="curl -f -I --connect-timeout 1 -X HEAD http://127.0.0.1:3000" \
my-container-image
label | type | default | description |
---|---|---|---|
com.github.mcasimir.autoheal.check.grace |
number |
15000 |
Grace period between last container start/restart and first healthcheck |
com.github.mcasimir.autoheal.check.frequency |
number |
5000 |
Healthcheck frequency |