/Go Engineering/Phase 6

Phase 6Days 26-30

Testing, Debugging, Performance & Security

Table tests, fuzzing, race detection, benchmarks, pprof, and govulncheck

Phase Goal

Prove correctness and improve performance using evidence from Go’s standard toolchain.

Progress

Notes

Open the written lectures for this course before checking off the phase topics.

Projects

Untrusted Telemetry Decoder Lab

Day 26: Table-Driven Tests & Testable Design

Define testing.T, subtests, helpers, and failure messages in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when testing.T, subtests, helpers, and failure messages is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for testing.T, subtests, helpers, and failure messages and fix it deliberately

Define Table-driven tests and boundary selection in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Table-driven tests and boundary selection is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Table-driven tests and boundary selection and fix it deliberately

Define Fakes, stubs, and interface seams in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Fakes, stubs, and interface seams is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Fakes, stubs, and interface seams and fix it deliberately

Define httptest and filesystem test isolation in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when httptest and filesystem test isolation is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for httptest and filesystem test isolation and fix it deliberately

Day 27: Fuzzing & Property Thinking

Define Seed corpus and FuzzXxx structure in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Seed corpus and FuzzXxx structure is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Seed corpus and FuzzXxx structure and fix it deliberately

Define Properties, invariants, and deterministic targets in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Properties, invariants, and deterministic targets is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Properties, invariants, and deterministic targets and fix it deliberately

Define Reproducing and promoting failures to regression tests in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Reproducing and promoting failures to regression tests is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Reproducing and promoting failures to regression tests and fix it deliberately

Define Where fuzzing helps and where it wastes time in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Where fuzzing helps and where it wastes time is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Where fuzzing helps and where it wastes time and fix it deliberately

Day 28: Benchmarks, Allocations & pprof

Define Benchmark loops and avoiding benchmark lies in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Benchmark loops and avoiding benchmark lies is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Benchmark loops and avoiding benchmark lies and fix it deliberately

Define b.ReportAllocs and allocation reasoning in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when b.ReportAllocs and allocation reasoning is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for b.ReportAllocs and allocation reasoning and fix it deliberately

Define CPU and heap profiles with pprof in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when CPU and heap profiles with pprof is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for CPU and heap profiles with pprof and fix it deliberately

Define Optimize measured hot paths, then re-measure in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Optimize measured hot paths, then re-measure is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Optimize measured hot paths, then re-measure and fix it deliberately

Day 29: Debugging, Race Detector & Security

Define Compiler, logs, delve awareness, and minimal reproductions in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Compiler, logs, delve awareness, and minimal reproductions is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Compiler, logs, delve awareness, and minimal reproductions and fix it deliberately

Define go test -race coverage and overhead in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when go test -race coverage and overhead is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for go test -race coverage and overhead and fix it deliberately

Define govulncheck call-graph-aware findings in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when govulncheck call-graph-aware findings is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for govulncheck call-graph-aware findings and fix it deliberately

Define Input limits, timeouts, secrets, and secure defaults in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Input limits, timeouts, secrets, and secure defaults is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Input limits, timeouts, secrets, and secure defaults and fix it deliberately

Day 30: Mini Project — Untrusted Telemetry Decoder Lab

Define Decode a deliberately inconsistent telemetry envelope in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Decode a deliberately inconsistent telemetry envelope is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Decode a deliberately inconsistent telemetry envelope and fix it deliberately

Define Table tests plus structural and semantic fuzz properties in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Table tests plus structural and semantic fuzz properties is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Table tests plus structural and semantic fuzz properties and fix it deliberately

Define Benchmark and profile before optimization in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Benchmark and profile before optimization is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Benchmark and profile before optimization and fix it deliberately

Define Project requirements, govulncheck, and threat notes in plain language and explain the problem it solvesTrace what the Go compiler or runtime does when Project requirements, govulncheck, and threat notes is usedWrite a tiny example, then apply it to a realistic backend taskDiagnose the beginner failure mode for Project requirements, govulncheck, and threat notes and fix it deliberately

Capstone

Untrusted Telemetry Decoder Lab

Harden a binary-and-text telemetry decoder against truncation, corrupt lengths, duplicate fields, oversized payloads, and pathological nesting.

Minimized fuzz failures promoted to regression tests
Before/after CPU and allocation evidence
Documented resource ceilings and vulnerability scan

Phase Complete!

After this phase, you'll be able to:

testing.T, subtests, helpers, and failure messages
Table-driven tests and boundary selection
Fakes, stubs, and interface seams
httptest and filesystem test isolation
Seed corpus and FuzzXxx structure
Properties, invariants, and deterministic targets
Reproducing and promoting failures to regression tests
Where fuzzing helps and where it wastes time

You can apply Phase 6 skills in a finished, explainable Go project.

Back to all phases

Phase Goal

Day 26: Table-Driven Tests & Testable Design

testing.T, subtests, helpers, and failure messages

Table-driven tests and boundary selection

Fakes, stubs, and interface seams

httptest and filesystem test isolation

Day 27: Fuzzing & Property Thinking

Seed corpus and FuzzXxx structure

Properties, invariants, and deterministic targets

Reproducing and promoting failures to regression tests

Where fuzzing helps and where it wastes time

Day 28: Benchmarks, Allocations & pprof

Benchmark loops and avoiding benchmark lies

b.ReportAllocs and allocation reasoning

CPU and heap profiles with pprof

Optimize measured hot paths, then re-measure

Day 29: Debugging, Race Detector & Security

Compiler, logs, delve awareness, and minimal reproductions

go test -race coverage and overhead

govulncheck call-graph-aware findings

Input limits, timeouts, secrets, and secure defaults

Day 30: Mini Project — Untrusted Telemetry Decoder Lab

Decode a deliberately inconsistent telemetry envelope

Table tests plus structural and semantic fuzz properties

Benchmark and profile before optimization

Project requirements, constraints & review

Untrusted Telemetry Decoder Lab

Phase Complete!