RabbitMQ and CentOS 7

[Lewis Franklin]

I was just wondering if anyone has successfully installed RabbitMQ on CentOS 7. I have tried, using both EPEL Erlang and Erlang Solutions, and regardless of the version of Erlang used, the install succeeds without issue but then actually running the server becomes a problem. If I run "service rabbitmq-server start", I get an the following log:

rabbitmq-server.service - LSB: Enable AMQP service provided by RabbitMQ broker

   Loaded: loaded (/etc/rc.d/init.d/rabbitmq-server)

   Active: failed (Result: exit-code) since Tue 2014-09-02 08:40:03 CDT; 11s ago

  Process: 3278 ExecStart=/etc/rc.d/init.d/rabbitmq-server start (code=exited, status=1/FAILURE)

Sep 02 08:40:02 localhost.localdomain systemd[1]: Starting LSB: Enable AMQP service provided by RabbitMQ broker...

Sep 02 08:40:02 localhost.localdomain su[3288]: (to rabbitmq) root on none

Sep 02 08:40:03 localhost.localdomain su[3332]: (to rabbitmq) root on none

Sep 02 08:40:03 localhost.localdomain su[3343]: (to rabbitmq) root on none

Sep 02 08:40:03 localhost.localdomain rabbitmq-server[3278]: Starting rabbitmq-server: FAILED - check /var/log/rabbitmq/startup_{log, _err}

Sep 02 08:40:03 localhost.localdomain rabbitmq-server[3278]: rabbitmq-server.

Sep 02 08:40:03 localhost.localdomain systemd[1]: rabbitmq-server.service: control process exited, code=exited status=1

Sep 02 08:40:03 localhost.localdomain systemd[1]: Failed to start LSB: Enable AMQP service provided by RabbitMQ broker.

Sep 02 08:40:03 localhost.localdomain systemd[1]: Unit rabbitmq-server.service entered failed state.

The output for /var/log/rabbitmq/startup_log is:

ERROR: distribution port 25672 in use on localhost (by non-Erlang process?)

[FAILED]

However, after multiple checks, there is nothing listening on 25672. I am able to successfully start rabbitmq-server by directly running "/usr/sbin/rabbitmq-server", but want to be able to take advantage of systemd for server control.

Lewis

[Michael Klishin]

On 2 September 2014 at 18:48:54, Lewis Franklin (lewis.f...@gmail.com) wrote:
> The output for /var/log/rabbitmq/startup_log is:
>
> ERROR: distribution port 25672 in use on localhost (by non-Erlang
> process?)
> [FAILED]
>
> However, after multiple checks, there is nothing listening
> on 25672. I am able to successfully start rabbitmq-server by
> directly running "/usr/sbin/rabbitmq-server", but want to
> be able to take advantage of systemd for server control.

This suggests it may be a package issue. 
--
MK

Staff Software Engineer, Pivotal/RabbitMQ

[Lewis Franklin]

That may be true, but I'm using the package from http://www.rabbitmq.com/install-rpm.html. I'm happy to do whatever testing is necessary as well.

Lewis

[Vignesh Kumar]

I had the same issue and I found out from SO that disabling and stopping firewalld service / or opening required ports and disabling SELINUX this issue is resolved in centos7. In /etc/selinux/config file make SELINUX=disabled.

[Lewis Franklin]

Thanks. Disabling SELinux did the trick (had already opened the correct ports).

--
You received this message because you are subscribed to a topic in the Google Groups "rabbitmq-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/rabbitmq-users/iK3q4GLpHXY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to rabbitmq-user...@googlegroups.com.
To post to this group, send email to rabbitm...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jauncy Riley]

I have disabled SELinux, disabled firewalld, iptables are open. I still am getting the 

Jan 15 20:02:06 node systemd[1]: Starting LSB: Enable AMQP service provided by RabbitMQ broker...

Jan 15 20:02:06 node su[1470]: (to rabbitmq) root on none

Jan 15 20:02:06 node su[1472]: (to rabbitmq) root on none

Jan 15 20:07:06 node systemd[1]: rabbitmq-server.service operation timed out. Terminating.

Jan 15 20:07:06 node rabbitmq-server[1372]: Starting rabbitmq-server:

Jan 15 20:07:06 node systemd[1]: Failed to start LSB: Enable AMQP service provided by RabbitMQ broker.

Jan 15 20:07:06 node systemd[1]: Unit rabbitmq-server.service entered failed state.

I've tried switching from EPEL erlang to the erlang repo version 17.4. I'm using the rabbitmq provided rpm 3.4.3-1

Centos7 64bit. Any ideas?

[Jean-Sébastien Pédron]

On 15.01.2015 21:16, Jauncy Riley wrote:
> I have disabled SELinux, disabled firewalld, iptables are open. I still
> am getting the
>

> Jan 15 20:07:06 node systemd[1]: rabbitmq-server.service operation timed
> out. Terminating.

Hi!

Could you please post RabbitMQ logs? Both the "normal" log and the sasl log.

--
Jean-Sébastien Pédron
Pivotal / RabbitMQ

[Karthik Suresh]

I ran into the same issue!  even after disabling firewalld and selinux on centos 7 'service rabbitmq-server start' was failing. 

I used the 'rabbitmq-server-3.5.6-1.noarch.rpm' from rabbitmq webpage. 

I should have checked this earlier but, the /var/log/rabbitmq/startup_err file pointed to /etc/init.d/rabbitmq-server file line 64 'startproc' having the error. 

I noticed that the daemon had the correct path to the executable /usr/sbin/rabbitmq-server, so I altered the line from

START_PROG="startproc"  == To ==> START_PROC=""

so that startproc is not used to call the binary, and that fixed the issue. 

is the issue a remnant of a suse rpm?