golang-troubleshooting

Read the stack trace to find the exact line, then trace backward to where the nil value originated rather than adding a nil check at the crash site. Common causes are unchecked error returns, map lookups returning zero values, and uninitialized struct fields.

Run tests or the binary with the race detector: go test -race ./... or go run -race main.go. It slows execution roughly 10x but reliably reports data races with full stack traces of the conflicting accesses.

A common cause is a select statement with a default case inside a for loop, which busy-spins when no channel is ready. Another cause is reading from a closed channel in select, which fires continuously; fix it with the comma-ok idiom and nil-ing the channel.

No. recover() only catches panics in the same goroutine where it is deferred. A panic in a child goroutine crashes the entire process, so each goroutine needs its own deferred recover.

Not without protection. pprof leaks goroutine stacks and memory contents and can be abused for DoS. Gate it behind an environment variable, require basic auth, and bind it to a separate port or localhost.

Start with fmt.Println or structured logging for simple, reproducible bugs. Escalate to Delve breakpoints when you need to inspect live state, step through execution, or attach to a running process that simpler instrumentation cannot explain.