Hey there! We've found the following might help you get unblocked faster:

• 🧵 Workflow Recovery on [Railway.com](Railway.com) - Lost Access to n8n Instance

• 🧵 How to correctly customize the n8n Dockerfile on Railway (without breaking login or DB migrations)?

If you find the answer from one of these, please let us know by solving the thread!

did you make a backup?

Unfortunately not, I saw the update available and thought it would work!

Hey there!
Unless you are using Railway's backup feature or a custom backup system, Railway is unable to provide a fix for lost, deleted, or corrupted data. Please note that the backup feature is available only for the pro and enterprise plan not the hobby.

For restoring the actual data, I am unsure the depth that it was corrupted. I suggest reaching out to the n8n forums! They might have people there who deal with these types of issues more frequently then us.

The same problem happening to me. Has anyone solved it?

I have de same problem: Problem logging in column User.role does not exist.
does anybody noks how to solve this? pls

I can't go back to previous versions either. I'm desperate... can anyone help?

What is Railway's support channel? They've only been answering me so far. It's been half a day and nothing. Terrible service.

Please can you help us with a roadmap, so far no one has given an answer at least to get out of the problem.

Does anyone have a link to a deployment that is failing with this error? We want to make sure the auto updates feature is not breaking services, but whether or not an update is supported is based on the actual service being deployed. Railway does not have any control over n8n or how it handles updates

i just redeploy this two and make some hard runs, i think its ok for now.
It would have been helpful if someone had at least provided some test cases, especially for those new to railway. It's been almost 10 hours since the incident, and I haven't received a direct response.

Railway does not control n8n, so for any application specific issues (like the column User.role does not exist error) we rely on community support. Apologizes that it took so long to get a proper fix. If anyone else comes across this issue, can you explain what you did to fix. Then we can share your response with other users experiencing similar problems with n8n.

Sure, I'll send you a log.

1. I woke up to a version update that was generating random errors in various areas, from database queries to Open AI nodes, all very random.

2. Go to "primary", select the three points, and select Redeploy.

3. Go to "postgres" and do the same.

4. Test with database queries in N8N and AI agents.

its happening again, i send some info of the logs

I'm a bit stuck, I don't have the technical skills for any of these solutions.

I am by no means an n8n export, but I see that your worker and primary n8n services are different versions. Can you try redeploying the worker. I think they need to be on the same version

done, ill make some test

Hi, I updated the database and redid it. I also redid it for Redis, Workers, and Primary in the same way. I didn't get the role error. Could you try it?

its working, thkns!