Graylog Cluster Node Status is wrong

Hi,

I’m setting up my new Graylog Cluster. The Node Status Page is definitely showing a wrong information. All Nodes are processing exactly the same amount of messages, but this is wrong because I send all logs to node 001 at the moment.

Also if I stop processing messages for example on Node 003, this action is performed on all nodes.

I have no idea how to debug this and find the issue. I can imagine that this may belong to some wrong data in the mongodb?

Anyone some advices?

Are you using the same rest_transport_uri for all nodes?

Yes. currently it is set on all nodes to the value my reverse proxy for the webinterface (X-Graylog-Server-URL) is exposing.

While I am writing this, I realize that this can not be right…