Minikube
What is Minikube?
Minikube is a tool that makes it easy to run Kubernetes locally. Minikube runs a single-node Kubernetes cluster inside a VM on your laptop for users looking to try out Kubernetes or develop with it day-to-day.
Features
- Minikube packages and configures a Linux VM, the container runtime, and all Kubernetes components, optimized for local development.
- Minikube supports Kubernetes features such as:
- DNS
- NodePorts
- ConfigMaps and Secrets
- Dashboards
- Container Runtime: Docker, and rkt
- Enabling CNI (Container Network Interface)
Installation
Requirements
- OS X
- Linux
- VT-x/AMD-v virtualization must be enabled in BIOS
kubectl
must be on your path. Minikube currently supports any version of kubectl
greater than 1.0, but we recommend using the most recent version.
You can install kubectl with these steps.
Hyper-v users may need to create a new external network switch as described here. This step may prevent a problem in which minikube start
hangs indefinitely, unable to ssh into the minikube virtual machine. In this add, add the --hyperv-virtual-switch=switch-name
argument to the minikube start
command.
Instructions
See the installation instructions for the latest release.
CI Builds
We publish CI builds of minikube, built at every Pull Request. Builds are available at (substitute in the relevant PR number):
Quickstart
Here's a brief demo of minikube usage.
If you want to change the VM driver add the appropriate --vm-driver=xxx
flag to minikube start
. Minikube Supports
the following drivers:
Note that the IP below is dynamic and can change. It can be retrieved with minikube ip
.
$ minikube start
Starting local Kubernetes cluster...
Running pre-create checks...
Creating machine...
Starting local Kubernetes cluster...
$ kubectl run hello-minikube --image=gcr.io/google_containers/echoserver:1.4 --port=8080
deployment "hello-minikube" created
$ kubectl expose deployment hello-minikube --type=NodePort
service "hello-minikube" exposed
# We have now launched an echoserver pod but we have to wait until the pod is up before curling/accessing it
# via the exposed service.
# To check whether the pod is up and running we can use the following:
$ kubectl get pod
NAME READY STATUS RESTARTS AGE
hello-minikube-3383150820-vctvh 1/1 ContainerCreating 0 3s
# We can see that the pod is still being created from the ContainerCreating status
$ kubectl get pod
NAME READY STATUS RESTARTS AGE
hello-minikube-3383150820-vctvh 1/1 Running 0 13s
# We can see that the pod is now Running and we will now be able to curl it:
$ curl $(minikube service hello-minikube --url)
CLIENT VALUES:
client_address=192.168.99.1
command=GET
real path=/
...
$ minikube stop
Stopping local Kubernetes cluster...
Stopping "minikube"...
Using rkt container engine
To use rkt as the container runtime run:
$ minikube start \
--network-plugin=cni \
--container-runtime=rkt \
--iso-url=http://storage.googleapis.com/minikube/iso/buildroot/minikube-v0.0.6.iso
This will use an alternative minikube ISO image containing both rkt, and Docker, and enable CNI networking.
Driver plugins
See DRIVERS for details on supported drivers and how to install
plugins, if required.
Reusing the Docker daemon
When using a single VM of kubernetes its really handy to reuse the Docker daemon inside the VM; as this means you don't have to build on your host machine and push the image into a docker registry - you can just build inside the same docker daemon as minikube which speeds up local experiments.
To be able to work with the docker daemon on your mac/linux host use the docker-env command in your shell:
eval $(minikube docker-env)
you should now be able to use docker on the command line on your host mac/linux machine talking to the docker daemon inside the minikube VM:
docker ps
On Centos 7, docker may report the following error:
Could not read CA certificate "/etc/docker/ca.pem": open /etc/docker/ca.pem: no such file or directory
The fix is to update /etc/sysconfig/docker to ensure that minikube's environment changes are respected:
< DOCKER_CERT_PATH=/etc/docker
---
> if [ -z "${DOCKER_CERT_PATH}" ]; then
> DOCKER_CERT_PATH=/etc/docker
> fi
Remember to turn off the imagePullPolicy:Always, as otherwise kubernetes won't use images you built locally.
Managing your Cluster
Starting a Cluster
The minikube start command can be used to start your cluster.
This command creates and configures a virtual machine that runs a single-node Kubernetes cluster.
This command also configures your kubectl installation to communicate with this cluster.
Configuring Kubernetes
Minikube has a "configurator" feature that allows users to configure the Kubernetes components with arbitrary values.
To use this feature, you can use the --extra-config
flag on the minikube start
command.
This flag is repeated, so you can pass it several times with several different values to set multiple options.
This flag takes a string of the form component.key=value
, where component
is one of the strings from the list below, key
is a value on the
configuration struct and value
is the value to set.
Valid key
s can be found by examining the documentation for the Kubernetes componentconfigs
for each component.
Here is the documentation for each supported configuration:
Examples
To change the MaxPods
setting to 5 on the Kubelet, pass this flag: --extra-config=kubelet.MaxPods=5
.
This feature also supports nested structs. To change the LeaderElection.LeaderElect
setting to true
on the scheduler, pass this flag: --extra-config=scheduler.LeaderElection.LeaderElect=true
.
To set the AuthorizationMode
on the apiserver
to RBAC
, you can use: --extra-config=apiserver.AuthorizationMode=RBAC
.
Stopping a Cluster
The minikube stop command can be used to stop your cluster.
This command shuts down the minikube virtual machine, but preserves all cluster state and data.
Starting the cluster again will restore it to it's previous state.
Deleting a Cluster
The minikube delete command can be used to delete your cluster.
This command shuts down and deletes the minikube virtual machine. No data or state is preserved.
Interacting With your Cluster
Kubectl
The minikube start
command creates a "kubectl context" called "minikube".
This context contains the configuration to communicate with your minikube cluster.
Minikube sets this context to default automatically, but if you need to switch back to it in the future, run:
kubectl config use-context minikube
,
or pass the context on each command like this: kubectl get pods --context=minikube
.
Dashboard
To access the Kubernetes Dashboard, run this command in a shell after starting minikube to get the address:
minikube dashboard
Services
To access a service exposed via a node port, run this command in a shell after starting minikube to get the address:
minikube service [-n NAMESPACE] [--url] NAME
Networking
The minikube VM is exposed to the host system via a host-only IP address, that can be obtained with the minikube ip
command.
Any services of type NodePort
can be accessed over that IP address, on the NodePort.
To determine the NodePort for your service, you can use a kubectl
command like this:
kubectl get service $SERVICE --output='jsonpath="{.spec.ports[0].NodePort}"'
Persistent Volumes
Minikube supports PersistentVolumes of type hostPath
.
These PersistentVolumes are mapped to a directory inside the minikube VM.
The Minikube VM boots into a tmpfs, so most directories will not be persisted across reboots (minikube stop
).
However, Minikube is configured to persist files stored under the following host directories:
/data
/var/lib/localkube
/var/lib/docker
Here is an example PersistentVolume config to persist data in the '/data' directory:
apiVersion: v1
kind: PersistentVolume
metadata:
name: pv0001
spec:
accessModes:
- ReadWriteOnce
capacity:
storage: 5Gi
hostPath:
path: /data/pv0001/
Mounted Host Folders
Some drivers will mount a host folder within the VM so that you can easily share files between the VM and host. These are not configurable at the moment and different for the driver and OS you are using. Note: Host folder sharing is not implemented on Linux yet.
Driver |
OS |
HostFolder |
VM |
Virtualbox |
OSX |
/Users |
/Users |
Virtualbox |
Windows |
C://Users |
/c/Users |
VMWare Fusion |
OSX |
/Users |
/Users |
Xhyve |
OSX |
/Users |
/Users |
Private Container Registries
To access a private container registry, follow the steps on this page.
We recommend you use ImagePullSecrets, but if you would like to configure access on the minikube VM you can place the .dockercfg
in the /home/docker
directory or the config.json
in the /home/docker/.docker
directory.
Add-ons
Minikube has a set of built in addons that can be used enabled, disabled, and opened inside of the local k8s environment. Below is an exampe of this functionality for the heapster
addon:
$ minikube addons list
- addon-manager: enabled
- dashboard: enabled
- kube-dns: enabled
- heapster: disabled
# minikube must be running for these commands to take effect
$ minikube addons enable heapster
heapster was successfully enabled
$ minikube addons open heapster # This will open grafana (interacting w/ heapster) in the browser
Waiting, endpoint for service is not ready yet...
Waiting, endpoint for service is not ready yet...
Created new window in existing browser session.
The currently supported addons include:
If you would like to have minikube properly start/restart custom addons, place the addon(s) you wish to be launched with minikube in the .minikube/addons
directory. Addons in this folder will be moved to the minikubeVM and launched each time minikube is started/restarted.
If you have a request for an addon in minikube, please open an issue with the name and preferably a link to the addon with a description of its purpose and why it should be added. You can also attempt to add the addon to minikube by following the guide at ADD_ADDON.md
Documentation
For a list of minikube's available commands see the full CLI docs.
Using Minikube with an HTTP Proxy
Minikube creates a Virtual Machine that includes Kubernetes and a Docker daemon.
When Kubernetes attempts to schedule containers using Docker, the Docker daemon may require external network access to pull containers.
If you are behind an HTTP proxy, you may need to supply Docker with the proxy settings.
To do this, pass the required environment variables as flags during minikube start
.
For example:
$ minikube start --docker-env HTTP_PROXY=http://$YOURPROXY:PORT \
--docker-env HTTPS_PROXY=https://$YOURPROXY:PORT
Known Issues
- Features that require a Cloud Provider will not work in Minikube. These include:
- Features that require multiple nodes. These include:
- Advanced scheduling policies
Design
Minikube uses libmachine for provisioning VMs, and localkube (originally written and donated to this project by RedSpread) for running the cluster.
For more information about minikube, see the proposal.
Additional Links:
- Goals and Non-Goals: For the goals and non-goals of the minikube project, please see our roadmap.
- Development Guide: See CONTRIBUTING.md for an overview of how to send pull requests.
- Building Minikube: For instructions on how to build/test minikube from source, see the build guide
- Adding a New Dependency: For instructions on how to add a new dependency to minikube see the adding dependencies guide
- Updating Kubernetes: For instructions on how to add a new dependency to minikube see the updating kubernetes guide
- Steps to Release Minikube: For instructions on how to release a new version of minikube see the release guide
- Steps to Release Localkube: For instructions on how to release a new version of localkube see the localkube release guide
Contributions, questions, and comments are all welcomed and encouraged! minkube developers hang out on Slack in the #minikube channel (get an invitation here). We also have the kubernetes-dev Google Groups mailing list. If you are posting to the list please prefix your subject with "minikube: ".