Kubenetes and Azure notes

Microsft Training - Developing Applications with Containers
============================================================

aka.ms/wscontainers

docker run 

Get an Image
===================
from Dockerhub
docker pull tutum/wordpress  

docker run -d -p 90:80  tutum/wordpress
-d detach mode  (let it run)]

docker search tutum
docker search tutum/wordpress

docker search microsoft/aspnet --limit 5   (best match)

--no-trunc (full description)


Remove Image
================
docker -rmi  imagename
 
 
Tip: How to stop and remove with part of the Container ID
=========================================================== 
docker stop  aa  
docker rm aa

-f force


How to pipe into the remove
============================
docker rm -f $(docker ps -aq) 


See the layers of a Docker Container 
========================================
docker inspect
docker inspect 11 | findstr "IPAddress" or docker inspect 11 | grep "IPAddress"


Dockerfile
================
LABEL
RUN
WORKDIR
ADD - get stuff from url
COPY 
CMD  - 
ENTRYPOINT  - Keeps the containers running/alive such as Service WatchDog  use /dev/null 
EXPOSE - ports


Docker Tags
==================
company/project - should use
use tags for different version/history etc..
default is always latest
-t mccontainer:v1

Docker Build
=================
docker build -tmycontainer:v1 .   (dot is the context)



Publishing dotNEt
====================
dotnet 
dotnet restore 
TIP: add nuget.cfg to your project for http Proxy
<?xml version="1.0" encoding="utf-8"?>
<configuration>
    <config>
         <add key="dependencyVersion" value="Highest" />
        <add key="https_proxy" value="https://proxy:80800" />
    </config>
</configuration>
dotnet build

dotnet publish -o published  (this will do the restore and build altogether)



Dockerfile for dotNEt
=====================
FROM microsoft/aspnetcore:2.0
WORKDIR /app
COPY published ./
ENTRYPOINT ["dotnet", "mywebapp.dll"]

Windows Containers (Node.js)
====================
2GB  - Nano  <1 seconds
11GB  - ServerCore <2 seconds  

Hyper-V Containers (it has own Kernel)   Nano 5s and Server core 10s


Microsoft SQL on Windows/Linux Docker=
-----------------------------------------
Connecting to a SQL Server container use the ContainerID:portnumber such as 1433


Logging into Private Registries
======================================
docker login URIRegistry -u <username> -p <password>
docker pull URIRegistry/dockerimage
eg docker pull dockerreg.i386.com/webappp 


Find out CPU/Memory usage
===============================
docker stats
docker top <container>  (like unix)
docker events 


Networking
===================
docker inspect nat - to see all the containers in the networks

docker network create -d transpatent tlan -  to make a transpatent network

Now specific a network for a container  docker run 

K8S
====
 Install kubectl tool which allows you to run commands against Kubernetes cluster. Skip this step if it is already installed on your machine (run “kubectl” to see if the command is recognized).

Windows:

     choco install -y kubernetes-cli

Linux:

sudo apt-get -y update && sudo apt-get -y upgrade  

curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl

chmod +x ./kubectl

sudo mv ./kubectl /usr/local/bin/kubectl

In Linux, due to current bug with azure-cli you need to revert to 2.0.23-1 version.

sudo apt-get -y purge azure-cli
sudo apt-get -y install azure-cli=2.0.23-1
 
 K8S
 ===
 https://kubernetes.io/docs/tutorials/stateless-application/expose-external-ip-address/
 kubectl run hello-world --replicas=5 --labels="run=load-balancer-example" --image=gcr.io/google-samples/node-hello:1.0  --port=8080
 
 
 

 Important to know and change to the relevant context.
kubectl config get-contexts  - List Clusters available
kubectl config current-context - Displays the current context of Pods and Nodes


 Create a deployment and name it “nginx” 
 
 kubectl run nginx --image=nginx --port=80  


kubectl delete deployment nginx

kubectl get deployment

kubectl expose deployment nginx --type=NodePort  (Services )

kubectl get services - Networking for your pod deployments

kubectl delete service nginx

kubectl expose deployment nginx --type=LoadBalancer (Services as LoadBalancer)
kubectl scale --replicas=3 deployment/nginx

kubectl get pods  - list the pod and show the number of services eg. Nginx running/alive
kubectl get nodes - list nodes (i.e computers)#
kubectl get pods -o wide
kubectl proxy (allows to create a tunnel)


K8S Labels - use them to manage updates/removes etc...


Minikube

Fix dos files on unix  apt-get / yum dos2unix
=================
Use to develop locally, cut down version of k8s.  Single node on the computer 


Minikube on Windows
========================
Installing see - https://github.com/kubernetes/minikube

https://rominirani.com/tutorial-getting-started-with-kubernetes-on-your-windows-laptop-with-minikube-3269b54a226

https://blog.alexellis.io/minikube-behind-proxy/

Azure
=====
 CLI - https://docs.microsoft.com/en-us/cli/azure/install-azure-cli-windows?view=azure-cli-latest
 az login
 

 
 Kubenetes under Docker for Windows
 ====================================
 Run under Powershell (Administrator)
 [Environment]::SetEnvironmentVariable("KUBECONFIG", $HOME + ".kube\config", [EnvironmentVariableTarget]::Machine)
 
Install the Kubenetes Dashboard on port 8001
================================================
kubectl create -f https://raw.githubusercontent.com/kubernetes/dashboard/master/src/deploy/recommended/kubernetes-dashboard.yaml

Local download the yaml file
kubectl create -f kubernetes-dashboard.yaml

Remember to run kubectl proxy
https://www.ntweekly.com/2018/05/25/deploy-kubernetes-web-ui-dashboard-docker-windows/