I’m facing an issue with my self-hosted n8n setup (running in Docker). For the past week, my instance has become very slow and unresponsive.
Whenever I open any node in a workflow, it takes a long time to load (10–15 seconds), and after that the browser becomes unresponsive and shows the “Wait” or “Page Unresponsive” dialog.
Because of this, I’m unable to debug or even change a single line in my workflows.
I noticed in the canvas that it was prompting for a critical update, so I upgraded to the latest version ([email protected]). Unfortunately, the problem still persists even after updating.
I’ve been struggling with this for 4 days and couldn’t figure out the root cause.
My setup:
n8n version: 1.107.4
Deployment: Self-hosted, using Docker
Has anyone else faced this issue? Could it be related to memory limits, browser cache, or something in the Docker setup? Any guidance would be really helpful.
You’re not alone — I’ve seen similar UI sluggishness after upgrading to 1.107.4 in Docker environments.
Besides enabling task runners as @Denisf88 suggested (by setting N8N_RUNNERS_ENABLED=true), here are a couple more troubleshooting steps that helped in my case:
Increase available memory or CPU: In Docker, check if your n8n container has enough resources. A constrained limit can make the UI lag.
Clear browser cache or try Incognito mode: Sometimes browser caching issues cause UI freezes; testing in incognito can isolate that.
Use a lighter browser or disable dev tools: Heavy manipulation like Console or Network panels open can slow the UI significantly.
Check Docker logs for errors: Run docker logs <container> to see if there are memory warnings, failure loops, or retry storms.
If none of that helps, consider rolling back to v1.107.3 temporarily to verify if performance is restored. That way we can determine if this is a regression in 1.107.4.
Let us know how it goes — happy to keep troubleshooting with you!
I just opened the workflow after changing the environment variable and restarting the Docker container.
Now, when I try to open Telegram or any other node, it throws an error saying “unresponsive, waiting, or exited.”
The issue is still the same. I also tried using the n8n Cloud free version, but it still gives me this error sometimes.
Task execution aborted because runner became unresponsive
The task runner failed to respond as expected, so it was considered unresponsive, and the task was aborted. You can try the following:
Optimize your script to prevent CPU-intensive operations, e.g. by breaking them down into smaller chunks or batch processing.
Ensure that all paths in your script are able to terminate, i.e. no infinite loops.
If your task can reasonably keep the task runner busy for more than 30 seconds, increase the heartbeat interval using the N8N_RUNNERS_HEARTBEAT_INTERVAL environment variable.
I’m also with the same issue in v1.106.3. Did you see that the v1.108.0 made performance improvements to the editor? I don’t think it’s related with our problem but i’ll test and see.
I encountered the same issue as you, and I found that nodes copied from older versions cannot be edited or modified on the latest versions. I had to completely recreate the nodes from scratch to avoid the unresponsive error. However, the UI on these versions still feels very slow.
