Install Helm in a Namespace
If you’re using Kubernetes you’re probably using Helm. By default Helm installs itself into the kube-system namespace. In our experiences at REVSYS, we find that can get a bit confusing with so many other things in that namespace. We like to put Helm into another separate namespace, often named tiller.
Since I’m not spinning up new clusters on a regular basis I ALWAYS forget how to do this properly and flail around in the dark for a few minutes trying to piece together the answer from blog posts and Stackoverflow.
I’m writing this post for me as much as you so I know I can find it when I need it.
Setup Helm in a namespace
We’re going to install Helm into a namespace tiller. Keep in mind you’ll need to either pass this to helm on
the command line each time you use it or set export TILLER_NAMESPACE="tiller"
in your environment.
The steps are fairly straight forward actually:
And that should do it. After the pod comes up, which can take a few seconds, you can verify your install by running:
You should not get an error after the first command and you should see matching versions for client and server, something like this:
Hope this post helps you as much as it will help me in the future!
NOTE: With the release of Helm 3, which no longer has a tiller process, this problem should just completely go away.
Frank Wiles
Founder of REVSYS and former President of the Django Software Foundation . Expert in building, scaling and maintaining complex web applications. Want to reach out? Contact me here or use the social links below.