google-container-engine-dse

Google Deployment Manager scripts for deploying DataStax Enterprise (DSE) on Google Container Engine (GKE)

Creating a Cluster

The deploy.sh script wraps both the deploy cluster and deploy DataStax commands. You will need gcloud installed and configured. By default this will deploy a four node cluster. To create a cluster simply clone this repo and run the deploy command:

git clone https://github.com/DSPN/google-container-engine-dse.git
cd google-container-engine-dse
./deploy.sh ben1

A successful deploy should look like this:

Working with a Cluster

With the current configs, our cluster is always called cluster-clustername, so we can get credentials by running:

gcloud container clusters get-credentials cluster-clustername

Now kubectl will be able to connect to the cluster. We can get the names of the nodes with the command:

kubectl get pods

Once you get a pod's name, you can run the following. Previous is especially useful in debugging as it shows the logs from the last container that attempted to run.

kubectl logs datastax-opscenter-rc-5qjv6
kubectl logs --previous datastax-opscenter-rc-5qjv6

You can also check your cluster is running from the command line:

kubectl exec datastax-node-0-n1yqu nodetool status

To get an interactive session to a node you can run:

kubectl exec -it datastax-node-0-8yk87 /bin/bash

To get the external IP of the OpsCenter machine (running on port 8888) use the following command. You can then open a web browser to that ip and port to view OpsCenter.

kubectl get services

Deployment can take up to 15 minutes. Once your DataStax is deployed on top of the Kubernetes cluster, OpsCenter will be accessible via a web browser on port 8888 of external IP for OpsCenter.

Deleting a Cluster

The https://cloud.google.com/ interface currently has an issue deleting a cluster. A workaround is provided in the delete.sh script. To delete the cluster simply run:

./delete.sh