Tweet
Tweet

Slide 1

Slide 1 text

Anubhav Mishra Senior Developer - Production Operations and Delivery @anubhavm Hootsuite’s Journey to a Production Grade Cluster Kubernetes

Slide 2

Slide 2 text

No content

Slide 3

Slide 3 text

Agenda ● Fundamentals ● Architecture ● AWS deployment ● Other options for deployment

Slide 4

Slide 4 text

Fundamentals

Slide 5

Slide 5 text

Fundamentals ● Pods ● Nodes

Slide 6

Slide 6 text

Fundamentals - Pods

Slide 7

Slide 7 text

Fundamentals - Nodes

Slide 8

Slide 8 text

Architecture

Slide 9

Slide 9 text

No content

Slide 10

Slide 10 text

API

Slide 11

Slide 11 text

API kubelet

Slide 12

Slide 12 text

scheduler API kubelet

Slide 13

Slide 13 text

scheduler API kubelet controller Manager

Slide 14

Slide 14 text

scheduler API kubelet controller Manager kube proxy

Slide 15

Slide 15 text

scheduler API kubelet controller Manager kube proxy kube DNS

Slide 16

Slide 16 text

scheduler API kubelet controller Manager kube proxy kube DNS etcd

Slide 17

Slide 17 text

scheduler API kubelet controller Manager kube proxy kube DNS etcd

Slide 18

Slide 18 text

Computer

Slide 19

Slide 19 text

Computer The journey begins….

Slide 20

Slide 20 text

“Kubernetes The hard Way” Kelsey Hightower Developer Advocate - Google

Slide 21

Slide 21 text

No content

Slide 22

Slide 22 text

No content

Slide 23

Slide 23 text

pod1 pod2

Slide 24

Slide 24 text

No content

Slide 25

Slide 25 text

No content

Slide 26

Slide 26 text

No content

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

No content

Slide 29

Slide 29 text

“Source/Destination Checks” whaaaaat?

Slide 30

Slide 30 text

No content

Slide 31

Slide 31 text

No content

Slide 32

Slide 32 text

No content

Slide 33

Slide 33 text

Advantages ● Using layer 3 (AWS Routing Table) ● Simpler implementation Limitations ● Routing table (50 entries max) = 50 nodes ● Complex orchestration*

Slide 34

Slide 34 text

Our Production Criteria ● ~1000s of nodes / cluster ● Operationally friendly ● Highly available ● Easily reproducible ○ Multi region ○ Multiple clusters / environment ● Secure

Slide 35

Slide 35 text

Our Production Criteria ● ~1000s of nodes / cluster ● Operationally friendly ● Highly available ● Easily reproducible ○ Multi region ○ Multiple clusters / environment ● Secure

Slide 36

Slide 36 text

Lets not use the AWS Routing Table? But how? pod1 pod2

Slide 37

Slide 37 text

Overlay Network (Flannel, Calico)

Slide 38

Slide 38 text

Overlay Network (Flannel, Calico)

Slide 39

Slide 39 text

Overlay Network (Flannel, Calico)

Slide 40

Slide 40 text

Our Production Criteria ● ~1000s of nodes / cluster ● Operationally friendly ● Highly available ● Easily reproducible ○ Multi region ○ Multiple clusters / environment ● Secure

Slide 41

Slide 41 text

Our Production Criteria ● ~1000s of nodes / cluster ● Operationally friendly ● Highly available ● Easily reproducible ○ Multi region ○ Multiple clusters / environment ● Secure

Slide 42

Slide 42 text

AWS Deployment

Slide 43

Slide 43 text

No content

Slide 44

Slide 44 text

No content

Slide 45

Slide 45 text

No content

Slide 46

Slide 46 text

Terraform

Slide 47

Slide 47 text

Vault

Slide 48

Slide 48 text

4 months x

Slide 49

Slide 49 text

No content

Slide 50

Slide 50 text

No content

Slide 51

Slide 51 text

But wait isn’t there something called “kops”?

Slide 52

Slide 52 text

kops (Kubernetes Operations) ● Easy to get a production grade cluster ● Create/Upgrade/Delete a cluster ● Reconfigure the cluster components ● Manage multiple clusters ● Export terraform

Slide 53

Slide 53 text

Kelsey Hightower

Slide 54

Slide 54 text

Yay! We made it through. Yes! We are hiring :D Senior Developer - Production Operations and Delivery @anubhavm Anubhav Mishra