A docker container to monitor other docker containers based on the dead man's switch principle.
Some containers are neither able to self-recover nor self-terminate when they run into trouble.
If you have a container that you expect regular logging output from, and long periods of no output are unexpected; then this container can restart containers that have been silent for longer than a predetermined time.
docker run -d --rm -v /var/run/docker.sock:/var/run/docker.sock ijmacd/dms <monitored_container> [<timeout>] [<interval>]
<monitored_container> is the name of the container you want monitored.
<timeout> is the period of silence before the container is restarted. This is
optional and uses the relative format from the linux date command. It defaults
to '-1hour'
<interval> is also optional and is the monitoring interval in seconds. This
defaults to '60' seconds.
This container produces regular output so it's possible to put a dead man's switch on the dead man's switch.
docker run -d --rm -v /var/run/docker.sock:/var/run/docker.sock --name one ijmacd/dms <monitored_container>
docker run -d --rm -v /var/run/docker.sock:/var/run/docker.sock --name two ijmacd/dms one
docker run -d --rm -v /var/run/docker.sock:/var/run/docker.sock --name three ijmacd/dms two