Don’t fly blind. Observability gives you actionable insights into your cloud native systems—from pinpointing errors, to increasing developer productivity, to tracking compliance.
Observability is the difference between an error message and an error explanation with a recipe how to resolve the error! You know exactly which service is affected, who’s responsible for its repair, and even how it can be optimized in the future. Cloud Observability in Action teaches you how to set up an observability system that learns from a cloud application’s signals, logging, and monitoring, all using free and open source tools.
In Cloud Observability in Action you will learn how to:
• Apply observability in cloud native systems
• Understand observability signals, including their costs and benefits
• Apply good practices around instrumentation and signal collection
• Deliver dashboarding, alerting, and SLOs/SLIs at scale
• Choose the correct signal types for given roles or tasks
• Pick the right observability tool for any given function
• Communicate the benefits of observability to management
A well-designed observability system provides insight into bugs and performance issues in cloud native applications. They help your development team understand the impact of code changes, measure optimizations, and track user experience. Best of all, observability can even automate your error handling so that machine users apply their own fixes—no more 3AM calls for emergency outages.
About the technology
Cloud native systems are made up of hundreds of moving parts. When something goes wrong, it’s not enough to know there is a problem—you need to know where it is, what it is, and how to fix it. This book takes you beyond traditional monitoring, explaining observability systems that turn application telemetry into actionable insights.
About the book
Cloud Observability in Action gives you the background and techniques you need to successfully introduce observability into cloud-based serverless and Kubernetes environments. In it, you’ll learn to use open standards and tools like OpenTelemetry, Prometheus, and Grafana to build your own observability system and end reliance on proprietary software. You’ll discover insights from different telemetry signals, including logs, metrics, traces, and profiles. Plus, the book’s rigorous cost-benefit analysis ensures you’re getting a real return on your observability investment.
What's inside
• Observability in and of cloud native systems
• Dashboarding, alerting, and SLOs/SLIs at scale
• Signal types for any role or task
• State-of-the-art open source observability tools
About the reader
For application developers, platform owners, DevOps, and SREs.
About the author
Michael Hausenblas is a Product Owner in the AWS open source observability team.
Author(s): Michael Hausenblas
Edition: 1
Publisher: Manning Publications
Year: 2023
Language: English
Commentary: Publisher's PDF
Pages: 264
City: Shelter Island, NY
Tags: DevOps; Cloud Computing; Logging; Microservices; Incident Response; Tracing; Grafana; Prometheus; Alerting; OpenTelemetry; Jaeger; Distributed Tracing; Observability; Service Level Objectives; Metrics; Continuous Profiling
Cloud Observability in Action
brief contents
contents
preface
acknowledgments
about this book
Who should read this book
How this book is organized
About the code
liveBook discussion forum
Online resources
about the author
about the cover illustration
1 End-to-end observability
1.1 What is observability?
1.2 Observability use cases
1.3 Roles and goals
1.4 Example microservices app
1.5 Challenges and how observability helps
1.5.1 Return on investment
1.5.2 Signal correlation
1.5.3 Portability
Summary
2 Signal types
2.1 Reference example
2.2 Assessing instrumentation costs
2.3 Logs
2.3.1 Instrumentation
2.3.2 Telemetry
2.3.3 Costs and benefits
2.3.4 Observability with logs
2.4 Metrics
2.4.1 Instrumentation
2.4.2 Telemetry
2.4.3 Costs and benefits
2.4.4 Observability with metrics
2.5 Traces
2.5.1 Instrumentation
2.5.2 Telemetry
2.5.3 Costs and benefits
2.5.4 Observability with traces
2.6 Selecting signals
Summary
3 Sources
3.1 Selecting sources
3.2 Compute-related sources
3.2.1 Basics
3.2.2 Containers
3.2.3 Kubernetes
3.2.4 Serverless compute
3.3 Storage-related sources
3.3.1 Relational databases and NoSQL data stores
3.3.2 File systems and object stores
3.4 Network-related sources
3.4.1 Network interfaces
3.4.2 Higher-level network sources
3.5 Your code
3.5.1 Instrumentation
3.5.2 Proxy sources
Summary
4 Agents and instrumentation
4.1 Log routers
4.1.1 Fluentd and Fluent Bit
4.1.2 Other log routers
4.2 Metrics collection
4.2.1 Prometheus
4.2.2 Other metrics agents
4.3 OpenTelemetry
4.3.1 Instrumentation
4.3.2 Collector
4.4 Other agents
4.5 Selecting an agent
4.5.1 Security for and of the agent
4.5.2 Agent performance and resource usage
4.5.3 Agent nonfunctional requirements
Summary
5 Backend destinations
5.1 Backend destination terminology
5.2 Backend destinations for logs
5.2.1 Cloud providers
5.2.2 Open source log backends
5.2.3 Commercial offerings for log backends
5.3 Backend destinations for metrics
5.3.1 Cloud providers
5.3.2 Open source metrics backends
5.3.3 Commercial offerings for metrics backends
5.4 Backend destinations for traces
5.4.1 Cloud providers
5.4.2 Open source traces backends
5.4.3 Commercial offerings for trace backends
5.5 Columnar data stores
5.6 Selecting backend destinations
5.6.1 Costs
5.6.2 Open standards
5.6.3 Back pressure
5.6.4 Cardinality and queries
Summary
6 Frontend destinations
6.1 Frontends
6.1.1 Grafana
6.1.2 Kibana and OpenSearch Dashboards
6.1.3 Other open source frontends
6.1.4 Cloud providers and commercial frontends
6.2 All-in-ones
6.2.1 CNCF Jaeger
6.2.2 CNCF Pixie
6.2.3 Zipkin
6.2.4 Apache SkyWalking
6.2.5 SigNoz
6.2.6 Uptrace
6.2.7 Commercial offerings
6.3 Selecting frontends and all-in-ones
Summary
7 Cloud operations
7.1 Incident management
7.1.1 Health and performance monitoring
7.1.2 Handling the incident
7.1.3 Learning from the incident after the fact
7.2 Alerting
7.2.1 Prometheus alerting
7.2.2 Using Grafana for alerting
7.2.3 Cloud providers
7.3 Usage tracking
7.3.1 Users
7.3.2 Costs
Summary
8 Distributed tracing
8.1 Intro and terminology
8.1.1 Motivational example
8.1.2 Terminology
8.1.3 Use cases
8.2 Using distributed tracing in a microservices app
8.2.1 Example app overview
8.2.2 Implementing the example app
8.2.3 The “happy path”
8.2.4 Exploring a failure in the example app
8.3 Practical considerations
8.3.1 Sampling
8.3.2 Observability tax
8.3.3 Traces vs. metrics vs. logs
Summary
9 Developer observability
9.1 Continuous profiling
9.1.1 The humble beginnings
9.1.2 Common technologies
9.1.3 Open source CP tooling
9.1.4 Commercial continuous profiling offerings
9.1.5 Using continuous profiling to assess continuous profiling
9.2 Developer productivity
9.2.1 Challenges
9.2.2 Tooling
9.3 Tooling considerations
9.3.1 Symbolization
9.3.2 Storing profiles
9.3.3 Querying profiles
9.3.4 Correlation
9.3.5 Standards
9.3.6 Using tooling in production
Summary
10 Service level objectives
10.1 The fundamentals of SLOs
10.1.1 Types of services
10.1.2 Service level indicator
10.1.3 Service level objective
10.1.4 Service level agreement
10.2 Implementing SLOs
10.2.1 High-level example
10.2.2 Using Prometheus to implement SLOs
10.2.3 Commercial SLO offerings
10.3 Considerations
Summary
11 Signal correlation
11.1 Correlation fundamentals
11.1.1 Correlation with OpenTelemetry
11.1.2 Correlating traces
11.1.3 Correlating metrics
11.1.4 Correlating logs
11.1.5 Correlating profiles
11.2 Using Prometheus, Jaeger, and Grafana for correlation
11.2.1 Metrics–traces correlation example setup
11.2.2 Using metrics–traces correlation
11.3 Signal correlation support in commercial offerings
11.4 Considerations
11.4.1 Early days
11.4.2 Signals
11.4.3 User experience
11.5 Conclusion
Summary
appendix—A Kubernetes end-to-end example
A.1 Overview
A.2 Prerequisites
A.3 Demo walk-through
A.3.1 Installing the demo
A.3.2 Using the demo
index
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
Z
