cnwan-operator-charts

Test reporitory for cnwan-operator's helm charts. NOTE: this documentation is a TODO and is done in a low-effort style, as it is not the definitive one.

Usage

Helm must be installed to use the charts. Please refer to Helm's documentation to get started.

Once Helm has been set up correctly, add the repo as follows:

helm repo add sunsince90 https://SunSince90.github.io/cnwan-operator-charts
helm repo update

If you had already added this repo earlier, run helm repo update to retrieve the latest versions of the packages. You can then run helm search repo cnwan-operator to see the charts.

To install the cnwan-operator chart:

helm install my-cnwan-operator sunsince90/cnwan-operator

To uninstall the chart:

helm delete my-cnwan-operator

At any point you can run

helm show values sunsince90/cnwan-operator

to see what you can override with --set <key>:<value>

Note that you put all values in just one --set, but in these examples those values are on separate --set for sack of clarity.

Install with Service Directory

if you install this on GKE then you can avoid setting google cloud data. Otherwise you need to provide operator.googleServiceDirectory.defaultRegion and operator.googleServiceDirectory.projectID with --set.

kubectl create ns cnwan-operator-system

helm install my-operator sunsince90/cnwan-operator \
--set operator.namespaceListPolicy=allowlist \
--set operator.serviceAnnotations="{traffic-profile}" \
--set operator.serviceRegistry=ServiceDirectory \
--set-file operator.googleServiceDirectory.serviceAccount=path/to/gcloud-credentials.json \
-n cnwan-operator-system

Install with etcd

With etcd dependency

This will install etcd on your cluster with the name service-registry.

kubectl create ns cnwan-operator-system

helm install my-operator sunsince90/cnwan-operator \
--set operator.serviceAnnotations="{traffic-profile}" \
--set operator.serviceRegistry=etcd \
--set operator.etcd.install=true \
-n cnwan-operator-system

# Wait a few minutes until service registry gets into ready state
watch kubectl get pods -n cnwan-operator-system

Without etcd

If you already have etcd running somewhere -- remember to specify the correct endpoints: