Kubernetes Engine

Question 1

Kubernetes Engine Cluster Modes

Answer 1

1) Standard: user-self managed cluster

2) Autopilot: GKE managed cluster, reduce operational cost

Question 2

Gcloud container clusters create

Answer 2

Create Kubernetes cluster

Question 3

gcloud container clusters get-credentials

Answer 3

connecting to Kubernetes cluster

Question 4

kubectl create deployment

Answer 4

deploying microservice to the cluster

Question 5

kubectl get deployment

Answer 5

reveal status of the microservice

Question 6

kubectl expose deployment

Answer 6

expose microservice external IP

Question 7

kubectl get service

Answer 7

checking status of the service, gives out

1) cluster internal IP
2) external IP
3) port used for the service
4) type expose of the service

Question 8

kubectl scale deployment

Answer 8

scale number of instances service in the cluster

Question 9

gcloud container clusters resize

Answer 9

increase number of nodes in Kubernetes cluster

Question 10

Command to autoscale microservice

Answer 10

kubectl autoscale deployment –max = –cpu-percent = 70

Question 11

Command to autoscale Kubernetes existing cluster

Answer 11

gcloud containers clusters update –enable-autoscaling –min-nodes= 1 –max-nodes=10

Question 12

command to create autoscale cluster

Answer 12

gcloud containers clusters create –enable-autoscaling –min-nodes=1 –max-nodes=10

Question 13

kubectl create configmap

Answer 13

create additional configuration to microservice

Question 14

kubectl create secret generic

Answer 14

create password configuration

Question 15

Master node

Answer 15

Manage cluster. Have 4 components:

1) API server
2) Scheduler
3) Control manager
4) etcd

Question 16

API server

Answer 16

Handle communication for clusters from outside and nodes by exposing the API

Question 17

Scheduler

Answer 17

Decide placement of the pods

Question 18

Control manager

Answer 18

Manage deployment and replica sets

Question 19

etcd

Answer 19

distributed database storing cluster configuration

Question 20

Worker node

Answer 20

Run the workloads and the pods

Question 21

Kubelet

Answer 21

Component of the worker node. Handle communication with the master nodes

Question 22

Pods

Answer 22

Contain one or more containers. All containers in a pod share same network, storage, ports and volumes

Question 23

Deployment

Answer 23

Created for each microservice

Question 24

Replica set

Answer 24

Managing specific number of pods running for specific microservice version

Question 25

Services

Answer 25

Help to expose application to the outside world using three types:

1) ClusterIP: for intra network communication, expose service only between internal IP
2) LoadBalancer: expose service externally using GCP Load Balancer. One load balancer is used for one microservice deployed
3) NodePort: expose service on each node’s IP by creating Ingress component to load balance multiple microservice

Question 26

Stateful deployments

Answer 26

Use Stateful set where each set of pods come with unique, persistent identities and stable hostnames

Question 27

Daemon Set

Answer 27

Run services on nodes for logging and monitoring

Question 28

kubectl apply -f deployment.yaml

Answer 28

Update the deployment

Question 29

kubectl rollout undo deployment

Answer 29

rolling back the deployment to previous version

Question 30

Pod “PENDING” state

Answer 30

Pod cannot be schedule onto a node due to insufficient resources

Question 31

Pod “WAITING” state

Answer 31

Fail to pull image