How to deal with timeout issue when debugging Phoenix app
Subscribe to receive Elixir news to your inbox every two weeks.
Expand your skills
Download free e-books, watch expert tech talks, and explore open-source projects. Everything you need to grow as a developer - completely free.
There is a common scenario: You'd like to debug your Phoenix app with break!/4 or IEx.pry/0. Everything works fine, until... Phoenix server throws a timeout error statement.
** (EXIT from #PID<0.5227.0>) shell process exited with reason: shutdownIt happens after about 60 seconds. Why?
The process has a certain amount of time to send a response. If it takes more - then the server says: "It takes too long. The process is probably suspended and in the meanwhile, the Client/user is raging on the other side. Let's kill him... (the process, not the user)".
Usually, it's a very nice, and helpful logic. But in the case of debugging it might be a real pain in the neck.
As soon, as we know the reason, we're able to find the solution. And here we are. Just set the idle_timeout option at your endpoint config. Generally at config/dev.exs, or so.
config :appname, Appname.Endpoint,
http: [
protocol_options: [idle_timeout: 5_000_000_000],
...
]Work with a team that keeps learning and building better software every day.
Related posts
Dive deeper into this topic with these related posts
You might also like
Discover more content from this category
Hey! Have you ever wondered about tests running inside the IEx shell? For a long time, I was convinced that it’s not really possible. And as it turns out - that’s not really straightforward. You won’t easily find information about that in the documentation.
Today's Advent of Code puzzle inspired me to create this TIL. It may sound trivial, but in fact, it's tricky if you are unfamiliar with the nuances of guards' functioning.
Ensuring that GraphQL mutations properly update your Apollo client's cache can be a bit tricky - here's how to manually control that.