@luckow thanks for reporting, I think https://git.cloudron.io/packages/n8n-app/-/merge_requests/64 should fix it

I just found the reason, a bad setting in my env.sh file which prevented the normal functioning of the code nodes. Regards

Yes, I've been doing this for years, you just have to make /app/data friendly enough for apps to find what they need. This generally involves setting a few env variables: HOME PATH (including NODE_PATH avoiding npm i -g) LD_LIBRARY_PATH (less common) other app specific ENVs All of these changes do not affect the running app since it's already running with its own ENVs. So these are great for additional services. Nice work @djxx

@basiliolp the forum search is a powerfull tool Welcome to the Cloudron Forum https://forum.cloudron.io/topic/13654/could-not-connect-to-your-mcp-server/

Deleting all executions still didn't move the database down from 500mb, but after running VACUUM FULL execution_data; it's now 13mb. I''ll take it.

@girish Works thank you. I see you already updated the docs!

@BrutalBirdie said in I turned off cloudron and turned it on again and my workflows disappeared and now I can no longer use n8n: if we continue in spanish wow - cloudron forum knows no bounds

Solved here: https://community.n8n.io/t/upload-a-local-file-to-an-s3bucket-via-a-form/96480/13 gDrive example [image: 1743781721625-fb0c02e7-a51f-4b94-be95-29debb33330c-image-resized.png]

In order to help you, please provide app logs and as much other information as possible. Which credentials are you referring to and also where is the logo replaced? In the Cloudron dashboard or somewhere else?

Yes, I think so. because 1.82.2 (latest) > 1.82.0, i expect it to include everything in 1.82.0

You can often install things to the /app/data location manually if you can set the right env variables. try: export HOME=/app/data then run the pip install

Good catch @nebulon . @zonzonzon https://docs.cloudron.io/apps/n8n/#custom-node-modules . Just edit /app/data/env.sh and set EXTRA_NODE_MODULES to empty string and then restart the app, see if that helps.

My mistake with setting in firewall in AWS, everything back to normal, it works just fine

Okay so for those with the same problem.. it's possible to fix on 1.74.3 (Current version of Cloudron) by following these steps: Set N8N_PROXY_HOPS to 1 in your ENV, then restart your n8n instance. DELETED the credential and recreate it (Modifying it don't work, you need to delete and recreate!) Update your workflow And it will works.

I managed to install by it by removing some un-used community nodes, not sure which one was causing this but by cleaning up I got it to install

Adding max_old_space_size to my .env solved the issue yay 🥳

New package should fix the issue

Hi @joseph pandoc is also part of Cloudron n8n package You're right I didn't realize this. It is a good surprise. I have other workflow requiring GDAL binaries for example (https://gdal.org/en/latest/). This one is mode specific and is not here. I guess I am good to make a custom packaging... I found some shady npm packages including binaries but it does not seem properly maintained.