N8n to baserow integration

Technical Help
, ,
1

Are you using our SaaS platform (Baserow.io) or self-hosting Baserow?

Self-hosted

If you are self-hosting, what version of Baserow are you running?

my version is 1.31.1

If you are self-hosting, which installation method do you use to run Baserow?

baserow & n8n on OpenShift with yaml file, both of them are running in same namespace so they should be able to hit each other using internal service address. it’s using the same docker image baserow/baserow:1.31.1

What are the exact steps to reproduce this issue?

  1. n8n > new node > baserow
  2. when using baserow URL https://br.example.com it works fine but my issue is it’s very slow, one minute to run a simple query where I am specifying, Database Name, Table Name and Row ID, I think because of this it doesn’t even display my list of as dropdown selection
  3. so I thought switching to local service address will help it going out to internet so when using
    baserow.baserow.svc.cluster.local it fails
    I have tried following variations they all fail
    http://baserow.baserow.svc.cluster.local
    https://baserow.baserow.svc.cluster.local:443
    http://baserow.baserow.svc.cluster.local:8000
    http://baserow.baserow.svc.cluster.local:58000

I added the following environment variables
- name: PRIVATE_BACKEND_URL
value: “https://baserow.baserow.svc.cluster.local:443”
- name: BACKEND_PORT
value: “0.0.0.0:443”
- name: BASEROW_ALLOWED_HOSTS
value: “br.example.com, baserow.baserow.svc.cluster.local”

2

Hi @seanclt,

I’m not familiar enough with this exact set up so I’ve done a bit of googling and LLMing. The conclusion is that this is likely some kind of openshift/kubernetes cluster networking issue.

Could you verify which ports Baserow actually exposes within the cluster?
oc get service baserow -n <namespace>

Then ensure that n8n can resolve the Baserow service:
nslookup baserow.baserow.svc.cluster.local

Other than that, perhaps openshift has some network policies that might be blocking internal traffic?

If these steps don’t get you naywhere, I’ll escalate this internally to see if someone from the dev team can take a closer look.