Document upgrading a cluster

[robrich]

Now that #5798 is merged, and minikube start doesn't auto-upgrade, how do I upgrade my cluster? Will minikube start --kubernetes-version ... upgrade it or dump the VM and start over? (and consume all the bandwidth as I download base images again.) Is there a minikube start --kubernetes-version stable or minikube start --kubernetes-version latest that will auto-upgrade the cluster?

Perhaps we need a docs page to answer this.

[tstromberg]

minikube start --kubernetes-version=v1.17.0 will upgrade you, without re-downloading the VM image. That said, the base VM image is only ~120MB compared to the ~600MB it takes to download a new Kubernetes release.

We don't yet support 'stable' or 'latest' designators, but it would certainly be a great feature to add. If you or anyone else is interested in making this work, add a check for paramVersion == "stable" here:

minikube/cmd/minikube/cmd/start.go

Line 1127 in 56c72e2

if paramVersion == "" { // if the user did not specify any version then ...

As far as docs, I agree. I'd be happy to review any PR that adds it. Here is where I would probably put it: https://github.com/kubernetes/minikube/tree/master/site/content/en/docs/Tasks

[tstromberg]

Also worth noting, this is the message users should see if there is an upgrade available:

out.T(out.ThumbsUp, "Kubernetes {{.new}} is now available. If you would like to upgrade, specify: --kubernetes-version={{.new}}

That doesn't invalidate the issues you've raised, just showing the guidance we are currently providing, as the documentation should match it.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[sharifelgamal]

We've added the documentation for this, I'm going to go ahead and close this. Feel free to reopen if necessary.

[life5ign]

Where is this in the docs? I can't find it.