It worked yesterday, but today suddenly this error came out.
As you can see, the Input is clear and I’ve searched “role” column all over the nods but there are no such things.
In the image, it appears on Message a model nod,
but time to time it happens Google Sheet nod or Send a mail node, too.
Problem in node ‘Message a model‘
column 9827483463398fbb7d6b96916fe451d8bdefa0e8.role does not exist
Same issue. I updated yesterday to version 1.110.1. Did you find a solution?
same issue here.
I have same issue. column 9827483463398fbb7d6b96916fe451d8bdefa0e8.role does not exist
Just updated self hosted n8n.
I have the same problem. Yesterday (9.09.25) everything worked pretty cool. Now - it’s a nightmare. I thoguht that this is problem with LLMs directly from OpenAI but it crash on every LLM from OpenRouter. Whatsmore sometimes this error happens on MySQL node. Someone have any idea how to fix it? I have also 1.110.1 version
ZERO @n8n support here? This is obviously a bug (somewhere). how do we fix this?
This solved my issue. My Primary was the only service I was redeploying to update n8n, but I also needed to redeploy my worker. Keep in mind guys, if you have n8nio/n8n as the source image, it is already pulling the latest version so you just have to redeploy without editing the source image to n8nio/n8n:1.110.1
you statement is incorrect on my issue. I was updating 1M times, redeploying. didn’t work until I actually named the current version. Do I need to change it back so the future updates doen’t break it?
Same Here.
Redeploying (at least in Railway) n8nio/n8n keeps bringing the .110 version.
For anyone else using n8nio/n8n (which is the popular n8n with workers template in Railway) you need to manually type the docker image for the .111 version. Redeploy and you will fix this problem.
NOTE: you will also need to to this in workers instance to solve the column problem
I managed to solve it.
I use Self-Hosted. In my case, I had updated the editor but not the workers or the webhook.
In short, I kept them all on the same version: latest, and I’m no longer experiencing the issue.
I updated through Portainer.
Como eu resolveria isso usando o railway por exemplo? dando redeploy nos workers?
(respondi em portugues pq vi que é br kkk)
Sem problemas qto ao idioma rss.
Meu querido eu não uso o railway mas tenho quase certeza que sim. Se vc tem a possibilidade de atualizar o workers, faz isso que eu acredito que vá resolver.
Resumindo, é importante que ambos estejam na mesma versão.
Obrigado meu querido!
I faced with the same issue. I managed to solve it by re-choosing the model of the ai again. Becasue the model I used before is gone from the dropdown. That’s it. the new version just update the model list. You just have to choose the new one in the list.
Funcionou dando redeploy no worker (self hosted)!!!
Ah man. You have no idea how happy i am that i came across this comment. I have been trying to figure workarounds since a week. I updated Postgres, redis, worker & primary and everything is working smoothly.
Question though, how often do you end up redeploying all 4. Till now i was only redeployng my promary instance and that eventually lead to all the issues. Do you update all 4 everytime there’s an update?
