[rabbitmq-discuss] How to remove a node from cluster without loosing data

Francesco Mazzoli francesco at rabbitmq.com
Wed Apr 25 16:00:18 BST 2012


Hi Nithya,

> I am using 2.7.1. I have a two node cluster and I want to remove the
> second node. In the clustering guide it says that this can be done by
> running stop_app, reset and start_app on the second node. In our setup,
> however, we no longer have access to the second node and have access
> only to the first node. We use automatic config file for clustering.

Currently removing "problematic" nodes from your cluster is kind of tricky.

If your node is simply down, then the quickest way to do this is to set 
up an impostor node on the same host with the same name and remove that.

If for some reason you can't reach your node at all, then the situation 
is a lot uglier and I don't think there is a safe way to proceed.

We are aware that this is a problem and we're working on this, along 
with making clustering operations more user friendly in general.

Francesco.


More information about the rabbitmq-discuss mailing list