Issues when moving workflows from could to self-hosted version

I have an issue with n8n self hosted version.

when i download a workflow json from the cloud version and try to import it to the self hosted version, lots of the nodes are missing details, ai agent missing tool connections, merge node does not have input connections. and all types of nodes that are not working properly, is this a known issue? is there any way to fix this so that the workflow is working in the self hostef version as the cloud version?

this is thre command i use to run n8n on docker:

sudo docker run -d \ --name n8n \ --network circular-app_circular-network \ -p 5678:5678 \ -v /opt/n8n/data:/home/node/.n8n \ -e N8N_HOST=n8n.cirklehub.com \ -e N8N_PORT=5678 \ -e N8N_PROTOCOL=https \ -e WEBHOOK_URL=https://n8n.cirklehub.com/ \ -e N8N_ENFORCE_SETTINGS_FILE_PERMISSIONS=true \ -e N8N_RUNNERS_ENABLED=true \ --restart unless-stopped \ n8nio/n8n:latest

1 Like

Hello @HadiAbu,

So this could be a few reasons, the first thing I would check, did you merge to same version? While there are some difference from cloud version, if importing into a different version of n8n you could be referensing old packages maybe, and connections aren’t showing.

So you had cloud which version was this running? and the dockers selfhosted which version please?

Best regards,

Samuel

Same versions 1.93 (currently the latest stable build)

Sure got it thanks, and how are you copying them into, just by selecting all and selecting copy? then pasting into your local enviroment? I’ve not seen this before so may need someone else to chime in on this… I guess we could check if there are any community packages you installed or was it just a base n8n cloud install.

If possiable it might be worth trying to share the workflow here too if okay.

n8n attach workflow

Applogies I can’t help further atm, if I think of anything else I’ll be sure to reach out

I tried copy paste between the n8n envs without luck. also downloading and importing.

Also i can’t copy past the workflow because here it works!
so it would not show any issues.

hm, I would try a restart / redeploy in that case seems maybe the local instance may have issues? Hope you get resolved soon