Cloud Native Infrastructure with Azure: Building and Managing Cloud Native Applications

by Nishant Singh, Michael Kehoe

Overview

The cloud is becoming the de facto home for companies ranging from enterprises to startups. Moving to the cloud means moving your applications from monolith to microservices. But once you do, running and maintaining these services brings its own level of complexity. The answer? Modularity, deployability, observability, and self-healing capacity through cloud native development.

With this practical book, Nishant Singh and Michael Kehoe show you how to build a true cloud native infrastructure using Microsoft Azure or another cloud computing solution by following guidelines from the Cloud Native Computing Foundation (CNCF). DevOps and site reliability engineers will learn how adapting applications to cloud native early in the design phase helps you fully utilize the elasticity and distributed nature of the cloud.

This book helps you explore:

• Why go cloud native?
• How to use infrastructure as code
• What it takes to containerize an application
• Why and how Kubernetes is the "grand orchestrator"
• How to create a Kubernetes cluster on Azure
• How observability complements monitoring
• How to use service discovery and a service mesh to find new territories
• How networking and policy management serve as gatekeepers
• How distributed databases and storage work

Product Details

ISBN-13:	9781492090960
Publisher:	O'Reilly Media, Incorporated
Publication date:	03/22/2022
Pages:	322
Product dimensions:	7.00(w) x 9.19(h) x 0.68(d)

About the Author

Nishant Singh is a senior site reliability engineer at LinkedIn, where he works to improve the reliability of the site with a focus on reducing the mean time to detect (MTTD) and mean time to respond (MTTR) to incidents. Prior to joining LinkedIn, he worked at Paytm and Gemalto as a DevOps engineer, spending his time building custom solutions for clients and managing and maintaining services over the public cloud. Nishant has a keen interest in site reliability engineering and in building distributed systems.

Michael Kehoe is a senior staff security engineer at Confluent. Prior to this, he worked on incident response, disaster recovery, visibility engineering, and reliability principles as a senior staff site reliability engineer at LinkedIn. During his time at LinkedIn, he led the company's efforts to automate the migration to Microsoft Azure. Michael specializes in maintaining large system infrastructure as demonstrated by his work at LinkedIn (applications, automation, and infrastructure) and at the University of Queensland (networks). He has also spent time building small satellites at NASA and writing thermal environments software at Rio Tinto.

Table of Contents

Preface xi

1 Introduction: Why Cloud Native? 1

The Journey to the Cloud 1

Challenges in the Cloud 2

Cloud Native Computing Foundation 4

Adopting a Cloud Native Infrastructure with Azure 4

Summary 5

2 Infrastructure as Code: Setting Up the Gateway 7

Infrastructure as Code and Its Importance in the Cloud Native World 8

Getting Started with Azure and Setting Up the Environment 11

Azure Fundamentals and Preparing Your Azure Environment 11

Creating an Azure Account 12

Installing the Azure CLI 13

Prominent IaC Tools 13

Terraform 14

Packer 29

Ansible 31

Azure DevOps and Infrastructure as Code 33

Summary 33

3 Containerizing Your Application: More Than Boxes 35

Why Containers? 35

Isolation 36

Security 36

Packaging and Deployment 37

Basic Container Primitives 37

Cgroups 38

Namespaces 39

Copy on Write 40

Capabilities 40

Seccomp-BPF 40

Components of Running a Container 40

Container Orchestrators 41

Container Software 41

Container Runtimes 42

Containers 43

Operating System 43

Open Container Initiative (OCI) Specification 43

OCI Image Specification 44

OCI Runtime Specification 45

Docker 46

Building Your First Docker Image 46

Best Practices While Using Docker 48

Other Container Platforms 49

Kata Containers 49

LXC and LXD 50

Container Registries 50

Securely Storing Images with Harbor 51

Securely Storing Images with Azure Container Registry 55

Storing Docker Images in a Registry 59

Running Docker on Azure 60

Azure Container Instances 60

Deploying an Azure Container Instance 61

Running a Docker Container Engine 65

Summary 66

4 Kubernetes: The Grand Orchestrator 67

Kubernetes Components 69

Control Plane 70

Worker Nodes 71

Kubernetes API Server Objects 72

Pods 72

ReplicaSets 73

Deployments 73

Services 73

Namespaces 74

Labels and Selectors 74

Annotations 74

Ingress Controller 74

StatefulSets 75

DaemonSets 75

Jobs 75

Observe, Operate, and Manage Kubernetes Clusters with kubectl 76

General Cluster Information and Commands 76

Managing Pods 78

Kubernetes in Production 85

Summary 98

5 Creating a Kubernetes Cluster in Azure 99

Creating a Kubernetes Cluster from Scratch 99

Creating the Resource Group 100

Creating the Machine Images for the Worker and Controller Machines 100

Creating a Storage Account Backend 101

Creating an Azure Virtual Network 102

Creating Public IPs for the Load Balancer 102

Creating Worker and Controller Instances 104

Using Ansible to Deploy and Configure the Kubernetes Controller Nodes 106

Using Ansible to Deploy and Configure the Kubernetes Worker Nodes 109

Setting Up Pod Networking and Routing 109

Generating the kubeconfig File for Remote Access and Cluster Validation 110

Azure Kubernetes Service 111

Deploying Applications and Services Using Helm: A Package Manager for Kubernetes 113

Helm Basics 114

Installing and Managing Helm 114

Managing Helm Releases 117

Creating Charts for Your Applications 118

Summary 120

6 Observability: Following the Breadcrumbs 121

Introduction to Observability 121

Observability: More Than Three Pillars 122

Observability: A Superset of Monitoring 123

Observability-Driven Development 124

Monitoring Metrics with Prometheus in a Cloud Native World 125

Prometheus Components and Architecture 125

Installing and Configuring Prometheus 127

Node_exporter 129

Instrumentation of Applications 130

Finding Hosts 134

Prometheus on Kubernetes 135

Logging in the Cloud Native World 138

Logging with Fluentd 138

Fluentd on Kubernetes 146

Distributed Tracing in the Cloud Native World 150

Tracing: Key Concepts 151

General Tracing System Architecture and Trace Assembly 153

Tracing Standards, Tools, and Code Instrumentation 154

Azure Monitor 159

Summary 161

7 Service Discovery and Service Mesh: Finding New Territories and Crossing Borders 163

Service Discovery 164

Introduction to CoreDNS 165

Installing and Configuring CoreDNS 167

Kubernetes Service Discovery with CoreDNS 169

Azure DNS 171

The Service Mesh 172

Introduction to Istio 174

Installing Istio on Azure Kubernetes Service 175

Automatically Injecting the Sidecar Proxy (Envoy Proxy) 177

Managing Istio Service Meshes Using Kiali 179

Summary 187

8 Networking and Policy Management: Behold the Gatekeepers 189

The Container Network Interface (CNI) 190

Why Would You Use a CNI? 191

How Does CNI Work with Azure? 191

Various CNI Projects 192

Calico 193

Why Would You Use Calico? 193

Basic Architecture 194

Deploying Calico 195

A Calico Deep Dive 197

Implementing Calico Security Policy 198

Cilium 200

Deploying Cilium 201

Integrating Cilium with Your Cloud 204

Flannel 207

Deploying Flannel 207

A Flannel Deep Dive 208

Azure Policy 210

Azure Policy Quickstart 210

Creating Your Own Azure Policy 212

Azure Policy for Kubernetes 213

Open Policy Agent 214

Deploying OPA on Kubernetes 215

Deploying Policy with OPA 216

Summary 218

9 Distributed Databases and Storage: The Central Bank 219

The Need for Distributed Databases in Cloud Native Architecture 219

Azure Storage and Database Options 220

Introduction to Vitess: Distributed and Sharded MySQL 221

Why Run Vitess? 221

The Vitess Architecture 222

Deploying Vitess on Kubernetes 223

Introduction to Rook: Storage Orchestrator for Kubernetes 224

The Rook Architecture 224

Deploying Rook on Kubernetes 225

Introduction to TiKV 226

Why Use TiKV? 226

The TiKV Architecture 226

Deploying TiKV on Kubernetes 228

More on etcd 229

Hardware Platform 230

Autoscaling and Auto-remediation 230

Availability and Security 231

Summary 231

10 Getting the Message 233

The Need for Messaging 233

A Sample Messaging Use Case: Log Ingestion and Analytics 235

Generation 1: Without Queues 235

Generation 2: With Cloud Queues and Object Storage 236

Generation 3: With Memory-Based Pub/Sub Queuing 237

The Basics of Messaging Platforms 238

Messaging Versus Streaming 238

Messaging Fundamentals 238

Producers and Consumers 239

Brokers and Clustering 240

Durability and Persistence 241

Message Delivery 241

Security 242

Common Messaging Patterns 242

Simple Queue 242

Publish and Subscribe 242

Durable Queue 242

An Overview of Popular Cloud Native Messaging Platforms 243

RabbitMQ 243

Apache Kafka 243

CNCF CloudEvents 244

Cloud Messaging Deep Dive with NATS 244

NATS Protocol Architecture 244

NATS Persistence with JetStream 249

NATS Security 249

Deploying NATS on Kubernetes 251

Azure Messaging Services 253

Azure Service Bus 253

Azure Event Hubs 258

Azure Event Grid 261

Summary 263

11 Serverless 265

Introduction to Serverless Computing 265

What Is Serverless? 265

What Is a Serverless Function? 266

The Serverless Landscape 266

Benefits of Serverless 267

Potential Downsides of Serverless 268

Azure Function Apps 268

Function App Architecture 269

Creating a Function App 270

Knative 272

Knative Architecture 272

Installing and Running Knative Serving on Kubernetes 272

Installing and Running Knative Eventing on Kubernetes 274

KEDA 276

KEDA Architecture 276

Installing KEDA on Kubernetes 277

OpenFaaS 281

OpenFaaS Architecture 281

Installing OpenFaaS 281

Writing Your First OpenFaaS Function 282

Summary 283

12 Conclusion 285

What's Next? 287

Index 289