For mac:
- Install Docker for Desktop if you have not already: https://www.docker.com/products/docker-desktop
- Set up kubernetes locally: In docker for desktop, go to Preferences > Kubernetes > Enable Kubernetes. Once you see the green dot and "Kubernetes is running", you can switch contexts to docker-desktop
- Install kubectl
brew install kubernetes-cli- Switch and confirm you are in your local k8s cluster
kubectl config use-context docker-desktop
kubectl config current-context
docker-desktop- Install helm
brew install kubernetes-helm- Initialize helm for this context
helm init- Assumes general knowledge of Docker containers.
- Install redis-cli (brew install redis)
- kubernetes: Greek for "helmsman"/"captain" - container orchestration for automating deployments, scaling, and management.
- node: worker machine that contains services necessary to run pods managed by master components
- cluster: A group of kubernetes nodes
- container: a light weight version of a virtual machine with an isolated piece of filesystem, cpu, memory, etc...
- pod: group of containerized components. Guaranteed to be co-located on a node and can share resources.
- replica set: grouping mechanism that maintains declared number of instances of a pod
- deployment: higher level mechanism that manages what happens to a replica set (updates, roll backs, etc).
- stateful set: like a deployment, only the state of the pods are preserved if they are restarted or destroyed.
- service: groups pods by labels and load balances traffic to them in round-robin manner.
- ingress: expose http(s) routes from outside the cluster to services within the cluster.
- volume: external/persistent storage that can be mountd onto a pod and shared between pods.
- namespace: used for partitioning and organization of resources (e.g. 'prod' and 'dev' namespaces can live in the same k8s cluster)
- config map: configuration, like environment variables, that can be attached to a pod or set of pods.
- secret: like config maps, but base64 encoded.
- manifest files: kubernetes files (yaml or json) that define your resources
Set up a python flask webserver with a redis backend that keeps track of page views.
- Create a deployment for redis
$ kubectl apply -f k8s/redis-deployment.yaml
deployment.apps/redis created
$ kubectl get deployment
NAME READY UP-TO-DATE AVAILABLE AGE
redis 1/1 1 1 23s
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
redis-d8f776cc6-29bvn 1/1 Running 0 7s
You will see that the deployment has created 1 out of 1 replicas of the redis pods.
- Create a service for redis
$ redis-cli ping
Could not connect to Redis at 127.0.0.1:6379: Connection refused
$ kubectl apply -f k8s/redis-svc.yaml
service/redis created
$ kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
redis ClusterIP 10.98.49.143 <none> 6379/TCP 3s
$ kubectl port-forward service/redis 6379:6379
$ redis-cli ping
PONG
- Deploy web app:
Note that the svc and deployment can go in the same yaml file!
$ kubectl apply -f k8s/web.yaml
service/web created
deployment.apps/web created
$ kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
redis ClusterIP 10.98.49.143 <none> 6379/TCP 25m
web ClusterIP 10.101.20.137 <none> 5000/TCP 18m
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
redis-d8f776cc6-29bvn 1/1 Running 0 11m
web-69cdcbddb-b4577 1/1 Running 0 93s
web-69cdcbddb-s7t8f 1/1 Running 0 93s
web-69cdcbddb-tt8tg 1/1 Running 0 93s
- Port forward and see the magic happen at localhost:5000
$ kubectl port-forward service/web 5000:5000
- Delete redis pod and see what happens
$ kubectl delete pod redis-d8f776cc6-29bvn
pod "redis-d8f776cc6-29bvn" deleted
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
redis-d8f776cc6-879lr 1/1 Running 0 10s
the old pod is gone, but the deployment noticed and spun up a new one.
refresh localhost:5000 - the page counts have started back over :( should have set up a persistent volume... maybe next time
- helm: (apparatus for steering a ship) - the package manager for kubernetes to configure/deploy.
- chart: package which consists of templated yaml files that can be rendered into k8s manifest files.
- values: parameters that can be used in the chart templates.
- release: specific configuration and deployment of a chart
- tiller: in-cluster component of helm that interacts with the k8s api to manage the k8s resources.
Create a template to work from
$ helm create mychart
mychart/
├── Chart.yaml
├── charts
├── templates
│ ├── NOTES.txt
│ ├── _helpers.tpl
│ ├── deployment.yaml
│ ├── ingress.yaml
│ └── service.yaml
└── values.yaml
$ head mychart/templates/deployment.yaml
apiVersion: apps/v1beta2
kind: Deployment
metadata:
name: {{ include "mychart.fullname" . }}
labels:
app.kubernetes.io/name: {{ include "mychart.name" . }}
helm.sh/chart: {{ include "mychart.chart" . }}
app.kubernetes.io/instance: {{ .Release.Name }}
app.kubernetes.io/managed-by: {{ .Release.Service }}
spec:
-
Edit to make it look like our k8s objects.
-
Create a namespace to keep things separate:
kubectl create namespace mychart
- Deploy the first release of this chart
$ helm upgrade mychart mychart --install --namespace mychart --wait
Release "mychart" does not exist. Installing it now.
NAME: mychart
LAST DEPLOYED: Thu Oct 10 17:58:59 2019
NAMESPACE: mychart
STATUS: DEPLOYED
RESOURCES:
==> v1/Service
NAME AGE
redis 4s
web 4s
==> v1/Deployment
redis 4s
==> v1beta2/Deployment
web 4s
==> v1/Pod(related)
NAME READY STATUS RESTARTS AGE
redis-7bf554988d-h896s 1/1 Running 0 4s
web-56989b5654-fl8sv 1/1 Running 0 4s
- Port forward from this namespace and we can see we are directed to this pod.
kubectl port-forward -n mychart service/web 5000:5000
- Change a value
helm upgrade mychart mychart --install --namespace mychart --wait --set web.replicaCount=2
...
==> v1/Pod(related)
NAME READY STATUS RESTARTS AGE
redis-98d85bc77-45d76 1/1 Running 0 7m
web-c795c955f-j7qwg 1/1 Running 0 4s
web-c795c955f-stvln 1/1 Running 0 7m
We've got another web pod.
Notice we are on revision 2
$ helm list
NAME REVISION UPDATED STATUS CHART APP VERSION NAMESPACE
mychart 2 Thu Oct 10 18:12:10 2019 DEPLOYED mychart-0.1.0 1.0 mychart