Lost workflows in n8n after an update at night
Anonymous
HOBBYOP

5 months ago

Hello,

I worked with my workflows in N8N, but I had a few days off. When i finally come back and tried to log in, there were "registration" page. I thought that is strange, but I created an account with the same data i had. When i logged in, I saw, that i lost all of my workflows.

Then i logged to Railway and saw, that there were an update (screenshots attached). Is it possible, that my data is somewhere, and during the update, n8n accidentally was connected to the empty database? How can I stop automatic updates, to won't get similar problems in the future?

$10 Bounty

6 Replies

brody
EMPLOYEE

5 months ago

This thread has been marked as public for community involvement, as it does not contain any sensitive or personal information. Any further activity in this thread will be visible to everyone.

Status changed to Open brody 5 months ago


idiegea21
HOBBY

5 months ago

Did you initiate any update 4 days ago or was that auto?

why i am asking this is: it is possible that railway redeployed your N8N service (possibly from a GitHub push or automatic update). and a new db was created or the n8n variables change or something.


Anonymous
HOBBYOP

5 months ago

No, that update was automatic. Can I, or support check it, if i can get any Access to my old database? How can I prevent that kind of situation in the future?


sim
FREE

5 months ago

Can you check if your volume is correctly mounted? If the path is set wrong then you are not writing data to persistent storage and you will lose data on restart. Here is documentation on mount path: https://docs.railway.com/guides/volumes#using-the-volume


No, that update was automatic. Can I, or support check it, if i can get any Access to my old database? How can I prevent that kind of situation in the future?

idiegea21
HOBBY

5 months ago

To prevent this from happening in future, disable auto deploys and pin N8N to a specific version, by using a version tag like n8n:1.39.1 instead of n8n:latest in your Dockerfile or Railway settings. this will prevents auto-updates when redeploying. You can find versions here: Docker Hub(https://hub.docker.com/r/n8nio/n8n/tags) .

then to solve the problem at and, can you confirm that either the was variable change or a new db was created during the redeploy?


Anonymous
HOBBYOP

5 months ago

Okay, I've checked my n8n instance on Railway and discovered I didn't have any volumes added. Gemini helped me realize that my workflows were likely stored directly within the container, which meant they were lost after the service restarted. I've now added a volume and redeployed the instance. I'm hopeful that all future workflows will be persistently stored, but unfortunately, it seems my old ones are unrecoverable. I'm quite new to DevOps and this area feels a bit like magic to me, which probably led to this oversight. Do you have any other suggestions I should implement now?"


sim
FREE

5 months ago

Do you have any other suggestions I should implement now?"

Turn on backups and check this https://docs.railway.com/reference/production-readiness-checklist


Loading...