Reoccurring issue with canvas not rendering nodes

Questions
1

Describe the problem/error/question

I keep having issues with loading workflows in the canvas. All nodes display as “?” and indicate they are not installed (see screenshots)
The nodes continue to function and run it appears, but i can no longer interact with them or the workflows via the canvas until i restart n8n.
After a restart everything seems fine for a day or two maybe.. Then the issue resurfaces.

What is the error message (if any)?

Please share your workflow

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

  • n8n version: 2.6.3
  • Database (default: SQLite): postgresql
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app): GCP cloudrun
  • Operating system:

image
image897×236 16.1 KB

image
image1886×913 42.2 KB

2

@spyder this is a known issue with n8n where nodes fail to load in the canvas but workflows still execute.

Here are fixesyou can try:

1. Clear Browser Cache

2. Check Browser Console for Errors

3. Memory Issues

If you’re running n8n in Docker with limited resources:

Increase memory allocation:

# docker-compose.yml
services:
  n8n:
    image: n8nio/n8n
    environment:
      - NODE_OPTIONS=--max-old-space-size=4096
    mem_limit: 4g

Or if using Docker run:

docker run -e NODE_OPTIONS=--max-old-space-size=4096 ...

4. Update n8n

5. Database Connection Issues

If using PostgreSQL/MySQL, connection pool exhaustion can cause this:

Increase connection pool:

# Environment variable
DB_POSTGRESDB_POOL_SIZE=20

6. Reverse Proxy/CDN Issues

If you’re behind a reverse proxy (Nginx, Cloudflare):

  • Check proxy timeout settings

  • Ensure WebSocket connections aren’t being blocked

  • Verify static assets are being served correctly

Nginx example:

location / {
    proxy_pass http://localhost:5678;
    proxy_http_version 1.1;
    proxy_set_header Upgrade $http_upgrade;
    proxy_set_header Connection 'upgrade';
    proxy_cache_bypass $http_upgrade;
    proxy_read_timeout 300s;
    proxy_connect_timeout 75s;
}

8. Check Disk Space

Low disk space can cause asset loading failures:

df -h  # Check available space

n8n needs space for:

  • Workflow data

  • Execution logs

  • Temporary files

  • Node modules

Does this help!

3

That’s a lot of things to list.

Let’s address them in order:

1.) Browser cache is not responsible. I have tried incognito windows, other browsers, other users on their own computers have the same issue. Only a reboot of the container running the main process resolves it

2.) No errors in browser console

3.) Memory is fine. Well within limits on both client and server

4.) There should be no reason to increase max-old-space-size. I will not try this approach for this issue

4.) I am only one minor release behind and the minor big fixes in the release notes from 2.6.3 to 2.6.4 are not remotely related to this issue. No reason to think this would solve anything

5.) I already have the DB POOL_SIZE=20 and no DB connection errors in logs and if there were DB connection issues it would render differenly with no access to workflows at all or credential store and readiness checks would fail.

6.) I use GCP Load Balancer. Been running as is with websocket support for 2 years in same ioc configuration. No reason to beleive it is the load balancer (especially since a reboot of the server corrects it for a few days.

8.) I use a GCP bucket for storage. It has unlimited capacity. Also all the items listed are stored in a postgres database (not disk) and the postgres database also has plenty of free space.

Sorry, but none of those solutions apply and in many cases re not diagnostically relevent.