zachallaun
Hi friends,
I’m working on a contribution to Finch and am having a hell of a time debugging an unexpected process exit that’s causing a test to fail.
The code being tested is rather unfinished and very much a work-in-progress, but I wanted to have a baseline before I start refactoring and moving things around. When the test reaches the call to cancel the async request, the test fails (not immediately, but after a very small period), with the following message:
Things I’ve tried so far:
- Re-creating the issue without using Finch (I couldn’t)
- Running the code outside of the context of a test (it works as expected)
- Creating the process with a Task supervisor (what’s currently in the code)
- Creating the process with
Task.async - Creating the process with
spawn - Various exit trapping
I have a hunch this has something to do with ExUnit’s GenServer, but I’m really struggling to find more information and I’m at a loss for how to debug further.
Trending in Questions
Hello!
Suppose you are building workflow (order / task / payment) processing system with the following requirements:
Each workflow con...
New
I’m aware of 2 such projects, are there others?
git: GitHub - team-alembic/realworld: A fullstack Phoenix LiveView application with back...
New
Hi everyone.
My team and I have been working on a fairly modest app based around video streaming and chat, but we’ve landed a customer t...
New
I really like the adapter patterns that ecto, nebulex, waffle, etc. use and would love find something similar for a key management servic...
New
Hello folks!
So at work, we are seeing some situations where we have to define some “fixed” strings that are used across the codebase in...
New
How Can I Optimise Compile Time Dependencies
I have been building an elixir application for about 2 years now. Many modules and files ha...
New
Hi,
we run different LLM models across a cluster, with each node loading a configurable subset. Model-specific queues let nodes consume ...
New
Other Trending Topics
I fully migrated to my own harness from Anthropic/Gemini and I think it’s time to share it. Welcome DSH, the DeepSeek Harness, fully writ...
New
Hey. Is there anyone here who creates agents in their apps? Not talking about using agents, but creating them. I’m finding it pretty diff...
New
Beam Bots (or just BB for short) is a framework for building fault-tolerant robotics applications in Elixir using familiar OTP patterns. ...
New
There has been a thread to discuss the Stack Overflow Developer Survey on this forum every year since 2018, so here’s yet another one for...
New
With AI doing more of the implementation work, I’ve been wondering how much coding I should deliberately keep doing myself.
My main conc...
New
I love Elixir. It’s one of 2 programming languages I’ve ever fallen in love with.
But I don’t use it anymore.
Serverless was the promis...
New
Categories:
Sub Categories:
Forums
Popular Tags
- #ecto
- #liveview
- #troubleshooting
- #learning-elixir
- #library
- #deployment
- #erlang
- #testing
- #genserver
- #mix
- #absinthe
- #remote-other
- #otp
- #plug
- #how-to-question
- #macros
- #postgres
- #elixirconf
- #channels
- #exunit
- #discussion
- #code-sync
- #podcasts
- #javascript
- #onsite
- #dialyzer
- #docker
- #authentication
- #ai
- #umbrella
- #full-time-contract
- #podcasts-by-brainlid
- #ecto-query
- #blog-post
- #elixirconf-us
- #elixir-ls
- #phoenix_html
- #iex
- #graphql
- #genstage
- #websockets
- #supervisor
- #advent-of-code
- #distillery
- #processes
- #elixirconf-eu
- #api
- #forms
- #security
- #hex











Showing Posts 1 to 6- Show Best Posts
- Show All (oldest first)
- Show All (newest first)
al2o3cr
The message looks a lot like the symptom from this issue:
https://github.com/PSPDFKit-labs/bypass/issues/120
This could be happening when the
Taskterminates while still waiting for a reply from Bypass, causing the Bypass process to crash and taking the test process down along with it.You could try trapping exits in the test and calling
receiveexplicitly to see what’s bubbling up.Another thing to try: add a sleep for longer than the sleep in the
Bypass.expectblock to the very end of the test, so that the Bypass process is definitely finished before the test ends.zachallaun
That looks really promising. I cloned Bypass and changed the dep to a local path-based one and then tried fixing the line referenced in that issue and removing it altogether, both to no avail. It does seem likely that it’s taking down the Bypass server somehow.
hkrutzer
Maybe
sysdebug can help? Debug tracing · Elixir Recipeszachallaun
That was a good idea.
So it looks like Bypass is sending an exit shutdown message to what I presume to be the test server. I’m not entirely sure why or what I can do to prevent it.
Edit: Looking at the code, it’s Bypass that’s exiting here.
Edit 2: And, indeed, if I modify Bypass to not exit at that line, the test works.
pcruz
I know it’s a bit late, but I have also encountered this problem, and I was able to resolve it by calling the
Bypass.downfunction before the end of the test:I hope this helps someone in the future.
RudManusachi
Bypass.down(bypass)for us didn’t work.However, what works is “overriding” the
on_exitfor the bypass so that it doesn’t try to verify the expectations.Some notes on my investigation:
In order to reproduce the issue we could do the following:
Basically the idea is to simulate that in the code, despite being asynchronous, actually makes the request that reaches out Bypass server.. But the test exits prior to the Bypass sending the response and. And since the test exists it shuts down Bypass!
However,
Bypass.open()prematurely setson_exitcallback where it makes a synchronousGenServer.callto Bypass to verify all its expectations. And the state of Bypass at that moment is erroneous:exit, :shutdownthat’s being raised.Knowing those details, I didn’t really want to comment that line in the Bypass source, since I think it’s still valuable for
Bypass.expect(when the request was made, but test finishes prior to Bypass responding.. though the error could have been better, but in that case the developer might want explicitly synchronize the test with Bypass maybe by some message passing andassert_receive)I made this patch but I thing it’s not as reliable as overriding
on_exitcallback.. since forBypass.stubthere should be no verification of expectations.. at lieast not on the number of requests