Today we installed RabbitMQ in a brand new server running Scientific Linux (SL) 6.2. SL is a relatively new Linux distribution based on RedHat Enterprise Linux, similar to CentOS, that lately has been gaining a lot of popularity.

The RabbitMQ installation on RHEL based servers is quite easy and well documented. So it was a surprise when RabbitMQ failed to start.

[root@dev]# /etc/init.d/rabbitmq-server start
Starting rabbitmq-server: FAILED – check /var/log/rabbitmq/startup_{log, _err}
rabbitmq-server.

The startup_err file included this error:

Crash dump was written to: erl_crash.dump
Kernel pid terminated (application_controller) ({application_start_failure,rabbit,{bad_return,{{rabbit,start,[normal,[]]},{‘EXIT’,{rabbit,failure_during_boot}}}}})

The message is quite cryptic. After looking around, some comments suggested issues with the hostname setup of our server, and others said it was an indication that something else was using the TCP port 5672 that RabbitMQ uses. The latter was the issue. A program called qpidd (Qpid AMQP Message Broker Daemon) was running and using the port. After stopping it, RabbitMQ started happily.