Timeout using CLI

[Larsschoen]

I'm trying to integrate a simple tolgee compare command into our CI/CD pipeline. Running that command on my local machine works fine, however inside the pipeline I'm running into a timeout:

(we are running a self hosted version of Tolgee)

> tolgee compare --verbose
Analyzing code...   ✓ Success
⚪ [HTTP] Requesting: GET http://xxx/v2/projects/2/all-keys
🔴 An unexpected error occurred while running the command.
🔴 Please report this to our issue tracker: https://github.com/tolgee/tolgee-cli/issues
🔴 fetch failed
TypeError: fetch failed
    at node:internal/deps/undici/undici:13178:13
    at async coreFetch (file:///builds/xxx/node_modules/openapi-fetch/dist/index.js:129:18)
[cause] Error: read ETIMEDOUT
    at TLSWrap.onStreamRead (node:internal/stream_base_commons:218:20)

At first I thought it's something network related but I can do a simple curl at the endpoint before the actual tolgee compare command and it will correctly respond with 403 (since I'm not sending any authentication headers for this simple test)

node: 22.5.1
@tolgee/cli: 2.14.0
tolgee-platform: 3.133.1

I'm grateful for any help :)

[Anty0]

Hi! Thanks for the report! ^^
Ah, the worst kind of bugs – the ones that only happen in the CI/CD pipeline. I definitely never spent hours trying to figure these out :3

I don't know if I'll be able to help much. Without a reproducible example or anything to go on, I still suspect something's wrong with the pipeline environment. One thing I'd try is making the curl request succeed by adding the authentication headers from the same environment variable the Tolgee CLI uses. If you manage to reproduce the issue—request ending up in a timeout error— that would be great!

Hope to hear back soon! Also, I'm sorry for not replying earlier. We parted ways with a colleague responsible for these parts of Tolgee, and it took us some time to realize nobody was responding to issues in this repository anymore.