|IamTheFij bb6c7f903a Remove Gopkg.*||преди 6 месеца|
|.drone.yml||преди 6 месеца|
|.gitignore||преди 7 месеца|
|Dockerfile||преди 6 месеца|
|Dockerfile.multi-stage||преди 6 месеца|
|LICENSE||преди 1 година|
|Makefile||преди 6 месеца|
|README.md||преди 6 месеца|
|go.mod||преди 6 месеца|
|go.sum||преди 6 месеца|
|main.go||преди 7 месеца|
|manifest.tmpl||преди 8 месеца|
Simple scheduling for short-running Docker containers
Dockron requires access to the Docker, so it may need to be run as root, or, if in a Docker container, need the socket mapped as a volume.
As simple as:
It will then run in the foreground, periodically checking Docker for containers with labels containing a cron schedule.
By default, Dockron will periodically poll Docker for new containers or schedule changes every minute. You can specify an interval by using the
Dockron is also available as a Docker image. The multi-arch repo can be found at IamTheFij/dockron
From either an
arm64 machine, you can run Dockron using:
docker run -v /var/run/docker.sock:/var/run/docker.sock:ro iamthefij/dockron -watch
First, be sure your container is something that is not long running and will actually exit when complete. This is for batch runs and not keeping a service running. Docker should be able to do that on it's own with a restart policy.
Create your container and add a label in the form
dockron.schedule="* * * * *", where the value is a valid cron expression (See the section Cron Expression Formatting).
Dockron will now start that container peridically on the schedule.
For more information on the cron expression parsing, see the docs for robfig/cron.
Dockron is quite simple right now. It does not yet:
I intend to keep it simple as well. It will likely never:
Either use a separate tool in conjunction with Dockron, or use a more robust scheduler like Tron, or Chronos.
If you have go on your machine, you can simply use
make build or
make run to build and test Dockron. If you don't have go but you do have Docker, you can still build docker images using the provide multi-stage Dockerfile! You can kick that off with