(Bandit.TransportError) Unrecoverable error: timeout

Phoenix Forum Questions / Help
,
1

Been getting these errors after upgrading to Phoenix 1.7.19, and Phoenix Liveview 1.0.4

I’m using Bandit 1.6.7 in my mix.lock file.

** (Bandit.TransportError) Unrecoverable error: timeout

Backtrace:

lib/bandit/http1/socket.ex:440 Bandit.HTTPTransport.Bandit.HTTP1.Socket.socket_error!/1	
lib/bandit/http1/socket.ex:50 Bandit.HTTPTransport.Bandit.HTTP1.Socket.read_headers/1	
lib/bandit/pipeline.ex:33 Bandit.Pipeline.run/4	
lib/bandit/http1/handler.ex:12 Bandit.HTTP1.Handler.handle_data/3	
lib/bandit/delegating_handler.ex:18 Bandit.DelegatingHandler.handle_data/3	
lib/bandit/delegating_handler.ex:8 Bandit.DelegatingHandler.handle_continue/2	
gen_server.erl:1085 :gen_server.try_handle_continue/3	
gen_server.erl:995 :gen_server.loop/7

That’s all the context information I have unfortunately, all the errors on AppSignal look like this one here.

Happening on the background namespace. Potential namespace values are: background, live_view, oban, web

image
image403×136 5.11 KB

Anyone experience this error before?

3 Likes
2

Noob question, how did you identify the namespace where this occurring ?

We are using Phoenix and also standing up our own websocket server (using cowboy) and are also seeing

(Bandit.TransportError) Unrecoverable error: timeout

but no stack trace beyond that so we haven’t been able to determine where exactly the error is coming from.

3

Wondering if anyone discovered any more on this?

I’ve seeing the same, every now and then:

[error] ** (Bandit.TransportError) Unrecoverable error: timeout

No further details.

Currently on Bandit 1.8.0.

Each time it happens, the app (page) goes dead. Have to do a forced reload, or navigate to another page, to get a reconnect. So far… not at all sure what the pattern is – and this particular app is very simple. No long running queries, processes, etc… pretty basic CRUD-oriented feature set.