This program can be used in synchronisation with KubeEdge in a High Available Cluster, where the cloudcore is been executed only one time (Not HA Setup of the KubeEdge Cloudcore).
This program is only written for in cluster usage but a few more points are also needed. This will be configured through environment variables. The following table contains all keys a description and default values.
Environment Variable | Description | Default Vaule |
---|---|---|
IP_KUBEEDGE_NAMESPACE | is the kubeedge namespace (default should be kubeedge) | no default value is set |
IP_KUBEEDGE_CONFIGMAP | is the name of the configmap, which will be created from kubeedge | tunnelport |
IP_KUBBEDGTE_STREAMPORT | is the streamport, which will be used by the tunnelport | no default value is set |
An example deployment is given in deployment/k8s
. You can apply this with the following command:
kubectl apply -f deployment/k8s