Frontend still shows v1.81.4 after upgrading backend to v1.95.3

Thank you in advance for any insights or suggestions you can share—I’d greatly appreciate your help!

I have self-hosted n8n with Docker Compose on /opt/n8n. After pulling the latest image and restarting, the CLI inside the container reports 1.95.3, but the UI still shows Version 1.81.4 and displays the “You’re on 1.81.4” banner. I have already:

  1. Corrected the volume mount to /home/node/.n8n.
  2. Cleared Service Worker and browser cache.
  3. Verified the n8n-editor-ui package version in the container (it was 1.95.2).
  4. Updated versionCli and versionNotifications.lastVersionSeen in SQLite via an external container.

However, the frontend banner persists.

What I’ve tried

  • Ensured volume mapping matches the official directory (n8n_data:/home/node/.n8n) community.n8n.io
  • Forced pull of docker.n8n.io/n8nio/n8n:1.95.3 and pruned old images
  • Cleared Service Worker (“Unregister”) and site data in Chrome DevTools n8n.io

**Expected behaviour
The UI should detect versionCli = 1.95.3 and no longer show the outdated version banner.

**Actual behaviour
Banner still reads “You’re on 1.81.4, which was released 3 months ago…” despite backend and database being updated.

**Environment

  • n8n Docker image: docker.n8n.io/n8nio/n8n:1.95.3
  • Volumes: n8n_data:/home/node/.n8n
  • Browser: Chrome v114 (Cleared cache, Service Worker unregistered)

Hi,
I’m getting a similar result. Did you figure this one out?
Thanks,
David

Hi David,

Thanks for reaching out.
I haven’t figured out the exact root cause yet, but I encountered a similar issue and managed to temporarily resolve it by trying a few things.

I suspect it might be related to the storage volume or some kind of cache issue. Before trying anything, I strongly recommend backing up your workflow JSON file just to be safe.

If you’re using Docker for deployment, one method that worked for me was to completely stop and remove the existing container, then spin up a new one using the latest n8n image. Just make sure your data is persisted using a properly configured volume, so nothing important is lost during the process.

You can refer to the following article for guidance on setting up persistent volumes and mounting directories correctly:
:backhand_index_pointing_right: How to deploy n8n on a server

This way, you should get the latest n8n web interface up and running.
Hope this helps!

If my suggestion doesn’t fully resolve the issue or things remain unclear, feel free to reach out again — I’ll try reproducing it in my own n8n environment and can provide a more detailed SOP if needed.

Best,
Kylzen

1 Like

Hi Kylzen,
This took a while to untangle…
I have 2 VMs on which I run containers and I was trying to update an instance of n8n on the wrong docker container.
I discovered the instance I was actually using had a different IP address than what Traefik said it had. I only checked this when I completely removed the volume I thought I was using and when I restarted the container it still said I was on 1.95.x…
I eventually spotted that the correct container was on my 2nd VM. The normal steps (when run on the correct VM) upgraded my instance. Happy days.

Now I’m going to go and remove the instances/containers that confused me and that I don’t need.

Many thanks for your reply and your kind offer of help.

Best regards,
David

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.