Operability in Go

Explore techniques for improving Go service operations in this GopherCon 2017 talk. Delve into resources from the standard library and runtime nuances to enhance instrumentation, stack inspection, and memory tracking. Learn about operability concepts, including failing gracefully, maintenance strategies, and diagnosing failures using the 5 Whys method. Discover how to gather crucial information for troubleshooting, manage logging effectively, and expose key data like environment variables, flags, and stack traces. Gain insights on using expvar for metrics, creating specialized monitoring endpoints, and developing libraries with operability in mind. Enhance your ability to design, maintain, and troubleshoot Go services for improved operational efficiency.

Intro
Preface
What are Operations
Software Design Life Cycle (SDLC)
Failing Well - Panic
Maintenance - Failure
Diagnosing Failure - 5 Whys
Diagnosing Failure We need a lot of information!
Killing a Stuck Process
Sources of Information
A Note on Errors
Logging Context
Logging Anxiety
MemStats
Exposing Environment
Exposing Secrets
Exposing Flags
Exposing a Stack Trace
expvar ALL the things!
Caution
Use Verbose Names
Specialized Endpoints - Monitoring
Library Developers
Recap