In this tutorial you’ll set-up a Kubernetes cluster on Google Container Engine (GKE) and install the Weave Cloud agents. By the end you’ll be able to visualize and explore your cluster, containers, and infrastructure in an interactive visual map. And start monitoring your applications for performance and reliability with the default metrics, dashboards and alerts.
- You’ll need to have created a Google Container Engine cluster.
Use the GCP Launcher
1. Go to Weave Cloud in Cloud Launcher
To make it easy to install we have a special integration of Weave Cloud in Google Cloud’s Launcher. Go to Weave Cloud in the Launcher. It will show you this page:
2. Subscribe to a plan
Scroll down the page and click
SUBSCRIBE in the table for the plan that you want. You may need to , click
3. Give Weave Cloud permissions
Launcher will start the Weave Cloud sign-up process, creating an account and logging you in. So that Weave Cloud can deploy the management agent on the cluster and to provide usage information we need some permissions. Click Allow to accept the permissions request:
4. Connect to your cluster
In GCP find Kubernetes Engine and find the cluster that you want to manage with Weave Cloud. On the right there is a Connect button select this and copy the first command.
Open a terminal and paste this command in to connect to the cluster.
5. Install the Weave Cloud agent
The Weave Cloud agent runs as a set of microservices on the cluster. Each cluster has a unique key for security this is automatically included in the instructions. Copy the instructions (using the Copy button on the right) and paste them into the terminal. There are two separate commands to run:
You can check their deployment status from the terminal by doing the following:
kubectl get pods -n weave
After a few moments the agents connect and the status page will updates.
Success! You can now use Explore to visualise and explore the services, containers and infrastructure running in the cluster. In the Monitor area there are some default dashboards, metrics and alerts so you can start monitoring the cluster’s performance and reliability. From here you can: