stack-traces-2.md

Use Case #stack-traces-2

Name: Async stack traces cannot be controlled by the user - can't log errors in production

Burton wrote a production Node.js application and is currently debugging code that uses promises in production.

// model file.
module.exports.someAsyncFn = async function someAsyncFn(ms) {
  await require('util').promisify(setTimeout)(ms);
  b();
};
async function b() {
  await require('util').promisify(setTimeout)(100);
  c();
}
async function c() {
  await require('util').promisify(setTimeout)(100);
  throw new Error(); // in practice - a real error case given input
}

// server file
app.get((req, res) => 
  let result = await someAsyncFn(5); // this throws an exception sometimes
  res.json(result); 
});

What happens

Burton gets a stack trace that breaks on async actions unlike when debugging in development with the inspector

Why it happens

Because async stack traces are currently not exposed outside of the inspector.

What can we maybe do better

Node.js could expose async stack traces sometimes somehow to achieve parity with promise libraries which would make users' lives easier. At least - it should be users' call.