An example project to show various Flink job cluster deployments on kubernetes, including an implementation of filesystem-based high availability.
A script that builds the Flink docker image with our streaming job embedded. This image is used for both job manager and task manager.
Same script as build-image.sh, but uses minikube's docker to build the image.
This allows to run the deployments on a local minikube cluster.
Deploys a flink job cluster in a high-avalability setup. To configure HA, edit the configmap with your desired HA. Link to relevant blog post.
Deploys a flink job cluster using the filesystem high availability implemented in this repo. This uses a shared persistent volume on the local host for HA as well as state storage. Link to relevant blog post.
Deploys a flink job cluster using the filesystem high availability implemented in this repo. This uses an EBS persistent volume for HA as well as state storage. Link to relevant blog post.
Questions and PRs are welcome :)