nymtech
Add peer-controller timeouts and lock tweaks
Ad a 5 s timeout around every PeerManager call into the WireGuard peer controller (recv_with_timeout). Instead of blocking forever when the controller is stuck after suspend, the authenticator now surfaces PeerInteractionTimeout and keeps processing other clients.
This change is 
The latest updates on your projects. Learn more about Vercel for GitHub.
| Project | Deployment | Preview | Comments | Updated (UTC) |
|---|---|---|---|---|
| nym-explorer-v2 |  Error |
Dec 11, 2025 1:54pm |
2 Skipped Deployments
| Project | Deployment | Preview | Comments | Updated (UTC) |
|---|---|---|---|---|
| docs-nextra |  Ignored |
Preview | Dec 11, 2025 1:54pm | |
| nym-node-status | Ignored |
Preview | Dec 11, 2025 1:54pm |
![vercel[bot] avatar](https://avatars.githubusercontent.com/in/8329?v=4 "Published by a pub.dev verified publisher"){kind=small}
Instead of blocking forever when the controller is stuck after suspend, the authenticator now surfaces PeerInteractionTimeout and keeps processing other clients.
What suspend are you referring to exactly? I'm curious about the logs that caused this bug, could you reference them in this PR as well please?
Also, if the controller is stuck for some reason, there's not really much reason to process other clients on the mixnet listener, because there won't be any working controller that can process the requests. In that case, we should probably just stop the whole authenticator, because otherwise we would just get continuous error logs.
Added comments in the code explaining this. We don't have specific log snippets in this PR, but I'll try to source them again from testing.
Stopping the authenticator - Agreed. This PR adds the timeout to prevent blocking. Failure tracking and shutdown on repeated failures.
A follow up PR is probably needed to address the shutdown of the authenticator. A good thing though is that the timeout logs warnings so operators can see the issue.