[rabbitmq-discuss] RabbitMQ 2.8.0 Clustering Problem

Michael Bridgen mikeb at rabbitmq.com
Thu Mar 29 13:13:33 BST 2012


Hi Reza,

> I'm using RMQ 2.8.0 and am trying to create a cluster between two (disk)
> nodes rabbit1 and rabbit2.
>
> I'm getting the error "Error:
> {no_running_cluster_nodes,[rabbit at rabbit1],[rabbit at rabbit1]}".
>
> I can telnet to rabbit1 5672 from rabbit2 and vice-versa from rabbit2,
> and I can do a rabbitmqctl -n rabbit1 status for rabbit2 and vice-versa
> as well (so I think the cookie works). The one difference is that both
> RMQ servers are running within a chroot env, due to requirements in my
> deployment environment.

You can test the connectivity definitively by getting a shell into each 
Erlang node and pinging the other. You need to do this while rabbit is 
stopped.

E.g., on host rabbit1:

     $ rabbitmqctl stop_app

and on host rabbit2:

     $ rabbitmqctl stop_app

then on rabbit1 again:

     $ erl -remsh rabbit at rabbit1 -sname r1
     > net_adm:ping(rabbit at rabbit2).

which should give you the result 'pong';
then, on rabbit2

     $ erl -remsh rabbit at rabbit2 -sname r2
     > net_adm:ping(rabbit at rabbit1).

(and again expect 'pong').


-Michael


More information about the rabbitmq-discuss mailing list