wvides/crud-k8s-deployment

K8s Crud application deployment

This repository will contain the files that will describe the deployment of the crud application

The namespaces at the cluster will be

• staging
• production

The services will be

• Backend application (an RESTful API written in node)
• Frontend application (a web ui)

How to recreate the cluster this repositorie uses

To create the cluster, the project crud-challenge must exist

gcloud beta container --project "crud-challenge" clusters create "crud-challenge-cluster" --zone "us-central1-a" --username "admin" --cluster-version "1.10.9-gke.5" --machine-type "n1-standard-1" --image-type "COS" --disk-type "pd-standard" --disk-size "100" --scopes "https://www.googleapis.com/auth/devstorage.read_only","https://www.googleapis.com/auth/logging.write","https://www.googleapis.com/auth/monitoring","https://www.googleapis.com/auth/servicecontrol","https://www.googleapis.com/auth/service.management.readonly","https://www.googleapis.com/auth/trace.append" --preemptible --num-nodes "3" --enable-cloud-logging --enable-cloud-monitoring --no-enable-ip-alias --network "projects/crud-challenge/global/networks/default" --subnetwork "projects/crud-challenge/regions/us-central1/subnetworks/default" --addons HorizontalPodAutoscaling,HttpLoadBalancing --enable-autoupgrade --enable-autorepair

To configure kubectl to use that cluster run

gcloud container clusters get-credentials crud-challenge-cluster --zone us-central1-a --project crud-challenge

How to use this repository

Requirements

1. You will need a gcloud account configured and linked to a valid and existing Google Cloud Platform account
2. You will need to create a projecdt called crud-challenge
3. At this poit you should already ran the steps described at "How to recreate the cluster this repositorie uses" in this document

Steps

1. Run sh common/initscript.sh to reserve two static IPv4 that will be used by the deployment (Please notice that this requires a gcloud cli and an existing google cloud account)

• WARNING Please notice that the initscript.sh reserves two static IPv4 on google cloud, if reserve an static IPv4 and don't assign it to a resource Google cloud will charge you per hour it spends not assigned, as shown here, under the Unused IPv4 address pricing section, at the moment of writing of this document the price for unused Static IPv4 is 0.010 US$ per hour, that would mean 14.6 US$ a month if both those IPs are not used

2. Run kubectl apply -f common/namespaces.yml to create the namespaces in the cluster

3. Create a secrets.yml file under a directory called secrets on this repository or in some place secure on your computer, with the secrets for both your staging and production database and ssl certificates, you can use the template below

NOTE: To encode your secrets use echo -n "yoursecrethere" | base64 in a unix system

WARNING: THIS REPOSITORY WILL IGNORE ANYTHING YOU PUT UNDER secrets FOLDER, DO NOT COMMIT ANY SECRETS, DOING SO WOULD COMPROMISE YOUR CREDENTIALS EVEN WHEN THEY ARE ENCODED WITH BASE64, THAT IS NO ENCRYPTION METHOD

apiVersion: v1
kind: Secret
metadata:
  name: staging-db
  namespace: staging
type: Opaque
data:
  dbhost: YOUR_ENCODED_BASE64_SECRET_HERE
  dbuser: YOUR_ENCODED_BASE64_SECRET_HERE
  dbpassword: YOUR_ENCODED_BASE64_SECRET_HERE
  dbport: YOUR_ENCODED_BASE64_SECRET_HERE
  database: YOUR_ENCODED_BASE64_SECRET_HERE
---
apiVersion: v1
kind: Secret
metadata:
  name: prod-db
  namespace: prod
type: Opaque
data:
  dbhost: YOUR_ENCODED_BASE64_SECRET_HERE
  dbuser: YOUR_ENCODED_BASE64_SECRET_HERE
  dbpassword: YOUR_ENCODED_BASE64_SECRET_HERE
  dbport: YOUR_ENCODED_BASE64_SECRET_HERE
  database: YOUR_ENCODED_BASE64_SECRET_HERE
---
apiVersion: v1
data:
  tls.crt: YOUR_ENCODED_BASE64_SECRET_HERE
  tls.key: YOUR_ENCODED_BASE64_SECRET_HERE
kind: Secret
metadata:
  name: tls-secret
  namespace: prod
type: Opaque
---
apiVersion: v1
data:
  tls.crt: YOUR_ENCODED_BASE64_SECRET_HERE
  tls.key: YOUR_ENCODED_BASE64_SECRET_HERE
kind: Secret
metadata:
  name: tls-secret
  namespace: staging
type: Opaque

To deploy to staging

• Run

$ cd /staging
$ kubectl apply -f .

To deploy to production

• Run

$ cd /production
$ kubectl apply -f .

crud-k8s-deployment's Issues

Create depployments and services for the application

• Backend application kubernetes deployment
• Backend application kubernetes service
• Frontend application kubernetes deployment
• Frontend application kubernetes service
• staging ingress
• production ingress