AWS - EKS - TASK
EKS - Elastic Kubernetes Service , it is one of the great service offered by AWS , it helps in
running K8S on AWS Cloud. It is fully managed by Amazon. It automatically
detects the and restart the instances which have gone down / terminated.
Let's get Started :
Creating K8S cluster using AWS EKS :
We
will be creating the cluster using a terraform file , because it is the best
way as it creates or destroys the cluster in 1 click and also as it is a file
we can share it or use it across multiple systems.
First
of create a file named clusterdep with file extension as yml , so the file name
should look like clusterdep.yml and then write the code given below.
The terraform code used is given below :
apiVersion:
eksctl.io/v1alpha5
kind:
ClusterConfig
metadata:
name: lwcluster
region: us-east-1
nodeGroups:
- name: ng1
describedCapacity: 1
instanceType: t2.micro
ssh:
publicKeyName: myredhatkey
- name: ng2
desiredCapacity: 1
instanceType: t2.small
ssh:
publicKeyName: myredhatkey
Now
lets deploy our terraform code using the following command:
eksctl
create cluster -f clusterdep.yml
Here
we can see that our cluster is ready and is in active state
Now that we have created our cluster lets make some more changes
to it and deploy our Wordpress on it , its going to be very interesting !
Lets
modify the terraform file made earlier i.e clusterdep.yml
apiVersion:
eksctl.io/v1alpha5
kind:
ClusterConfig
metadata:
name: lwcluster
region: us-east-1
nodeGroups:
- name: ng1
describedCapacity: 1
instanceType: t2.micro
ssh:
publicKeyName: myredhatkey
- name: ng2
desiredCapacity: 1
instanceType: t2.small
ssh:
publicKeyName: myredhatkey
- name: ngmixed
minSize: 2
maxSize: 3
instanceDistribution:
maxPrice: 0.017
instanceTypes:
["t3.small","t3.medium"]
onDemandBaseCapacity: 0
onDemandPercentageAboveBaseCapacity: 50
spotInstancePools: 2
ssh:
publicKeyName: myredhatkey
Now
let's create a web-st-pvc.yml file :
apiVersion:
v1
kind:
PersistentVolumeClaim
metadata:
name: lwpvc1
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 2Gi
Now
once our cluster is created, we will check for dns and will paste it in
browser.
It
is worth mentioning that our website works on type LB - load balancer so
traffic remains distributed and hence reduce the load on the servers and also
we are using pvc so our data is persistent.
we
will use kustomization
file with secret keys.
kubectl
create -k
Now
we can launch our wordpress
Here
we can see that our wordpress is configured.
Now let's use Helm package to launch Grafana and Prometheus
:
HELM: It is a program for installing
packages. It is package manager/ chart manager. Helm hub gives us k8s ready
applications . The server side used for Helm is tiller.
For
initialising Helm we will give the command :
helm
init
Then
we are required to run the following commands:
helm
repo add stable https://kubernetes-charts.storage.googleapis.com/
helm
repo list
helm
repo update
Now
proceeding further we will create tiller, for doing that we will run the
following commands:
kubectl
-n kube-system create service account tiller
kubectl
create clusterrolebinding tiller --clusterrole cluster-admin
--serviceaccount=kube-system:tiller
helm
init --service-account tiller
kubectl
get pods --namespace kube-system
helm
init --service-account tiller --upgrade
Now
we will install Prometheus with the help of Helm, for doing that we will run
the following commands:
kubectl
create namespace prometheus
helm
install stable/prometheus --namespace
prometheus --set
alertmanager.presistentVolume.storageClass="gp2" --set
server.presistentVolume,storageClass="gp2"
kubectl
get svc -n prometheus
kubectl
-n prometheus port-forward svc/whimsical-buffalo-prometheus-server 8888:80
We
can see that prometheus is successfully configured
Now
lets move further and install grafana, but for doing that we will need to run
some specific commands as follows:
kubectl
create namespace grafana
helm
install stable/grafana --namespace
grafana --set
presistence.storageClassName="gp2" --set adminPassword='GrafanaAdmin'
--set datasources."datasources\.yaml".apiVersion=1 --set
datasources."datasources\.yaml".datasources[0].name=Prometheus --set
datasources."datasources\.yaml".datasources[0].type=Prometheus --set
datasources."datasources\.yaml".datasources[0].url=http://promethus-server.prometheus.svc.cluster.local --set
datasources."datasources\.yaml".datasources[0].access=proxy --set
datasources."datasources\.yaml".datasources[0].isDefault=true
--set service.type="LoadBalancer"
kubectl
get sercret fair-numbat-grafana
--namespace grafana -o ymal
we
can see that Grafana has been configured successfully.
Now lets configure the Fargate Cluster:
So
for creating a fargate we just need to write a simple script.
Let the name of the script be farcluster.yml
apiVersion:
eksctl.io/v1alpha5
kind:
ClusterConfig
metadata:
name: fargate-lwcluster
region: ap-southeast-1
fargateProfiles:
-name: fargate-default
selectors:
- namespace: kube-system
- namespace: default
So as we have created the file now its time to
run it , but for doing this as usual we will be running some commands given
below :
kubectl
get ns
kubectl
get pods -n kube-system -o wide
eksctl
create cluster -f farcluster.yml
eksctl
get cluster --region ap-southeast-1
After
running the above commands we will create our cluster , yay!!
You can find all the scripts here:
