Join GitHub today
GitHub is home to over 36 million developers working together to host and review code, manage projects, and build software together.Sign up
Alternate restarts fail on CentOS #264
I noticed an issue with SELinux enforced CentOS.
Since the "rabbitmq.upstart.conf.erb" template has the PID_FILE hardcoded to "/var/run/rabbitmq/pid". The startup script "/etc/init.d/rabbitmq-server" is not able to create the directory "/var/run/rabbitmq" and I am getting an error when I run "service rabbitmq-server restart" every other time. It turned out to be that it could not find which process to kill.
This seems to be happening because of the security context of the "/var/run" folder restricting the startup script to be able to create that folder.
Ideally it would be nice if that value where we wanna store the PID_FILE is an attribute.