Automatic Restart by AI Agent and zombie processes
I am a beginner. I use Linux Mint. I understand that the Wasp server restarts with every change. I also know about the wasp clean command, but I’ve been running into a problem: when AI agents reboot the server, it seems to leave zombie processes alive, and the server never properly reboots—it hangs with an error.
What is the fix? It requires logging out and logging back in. In short, this removes all zombie processes. However, this is not a practical solution.
I have two dangerous commands that can stop the Wasp servers because Ctrl-C is disabled and won't stop the server manually:
1) pkill -f wasp || pkill -f node // Not recommended
2) lsof -i:3001 -t | xargs -r kill -9 // Not recommended
I need a manual way to reboot the server:
1) Do you have a proper way to manually reboot the Wasp server?
2) Is it possible to enable Ctrl-C for development?
3) Is there a way to make wasp start work again without logging out of my Linux?
No, I don’t have any errors right now. I’m just sharing my experience and the problems I’m facing.
Pls help me solve this problem because it has become a huge problem. (as of Oct 2025. )
2 Replies
Hi @Sky Blue. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
----------------------
Thank you for sharing your experience—this is a known pain point for some Wasp users, especially when zombie processes are left behind after restarts or crashes.
1) Proper Way to Manually Reboot the Wasp Server
The recommended approach is to ensure all Wasp-related processes are stopped before restarting. If you encounter a situation where ports (like 3000 or 3001) are still in use, you can manually kill the processes occupying those ports. The Wasp community suggests the following commands:
This will kill any process using those ports, which are typically the frontend (3000) and backend (3001) servers for Wasp. This is considered safer than using
- Use
pkill -f wasp
or kill -9
on all node processes, as it targets only the relevant ports and avoids killing unrelated processes. After running these commands, you should be able to run wasp start
again without needing to log out or reboot your machine [Github Issue: Process is already running on port 3000].
2) Enabling Ctrl-C for Development
Normally, Ctrl-C should work to stop the Wasp development server. However, there are rare cases where it becomes unresponsive due to how child processes are managed. This is a known issue, and the Wasp team has worked on improvements to ensure that all child processes are terminated when the main process is killed. If Ctrl-C is not working, using the fuser
commands above is the next best manual solution [Github Issue: Process is already running on port 3000].
3) Making wasp start
Work Again Without Logging Out
You do not need to log out or reboot your system. Instead, use the fuser
commands to free up the ports, then run wasp clean
to remove any generated code and cached artifacts:
Then, start your app again:
This sequence should resolve most issues with zombie processes and allow you to continue development without a full logout or reboot [Docs: CLI Reference].
---
Summary:- Use
fuser -k 3000/tcp
and fuser -k 3001/tcp
to kill stuck processes.
- Run wasp clean
before restarting with wasp start
.
- Ctrl-C should work, but if not, use the above commands.
- No need to log out or reboot—these steps should resolve the issue.
If you continue to experience problems, please provide more details or error messages, and the community can help further!
----------------------
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
----------------------
Sorry @Sky Blue, I seem to be having a bit of trouble. Don't worry though, I've pinged my support and they will try to fix me up! :technologist:
Perhaps try again in 15mins?In my case Ctr-C has not worked. It has failed 100% of times.