libesz / etcd_healthchecker Goto Github PK

Demo Etcd + Helm project

A working kubernetes cluster is assumed with:
- kubedns or other service discovery
- etcd operator with the custom resource registration

The Chart in the chart/ directory will deploy:
- an etcd cluster, using the operator
- an example application which
  - monitors the etcd cluster size
  - and watches a key in the db
- a post-install hook which writes the watched key as a bootstrap trigger

Chart parameters:
- cluster_name, used to name the etcd cluster and also the app pod
- cluster_size, for the etcd cluster

Usage:
- Install a Release with specific values:
  helm install chart/ --name meetup --set cluster_size=5 --set cluster_name=meetup
- Upgrade e.g. to change db cluster size
  helm upgrade meetup chart/ --reuse-values --set cluster_size=3
- Delete the Release
  helm delete meetup --purge