As I’ve mentioned in previous posts, I’ve had problems with resque, as it’s currently set up, running as its supposed to on Centos 6.5. I can see the application active if I check for it:
[root@srv-git-01 ~]# ps -ef | grep "resque" root 27821 1 0 13:14 ? 00:00:00 /bin/bash -i -l -c chruby ruby- && /var/www/gitorious/app/bin/rake resque:work QUEUE=* git 27850 27821 77 13:14 ? 00:00:02 ruby /var/www/gitorious/app/bin/rake resque:work QUEUE=* root 27853 7112 0 13:14 pts/0 00:00:00 grep resque
But it isn’t doing the work it’s supposed to: I manually have to run the command for new repositories to be created, and updates from users don’t show up on the website until I run resque manually.
I now have a way to make it work. I kill the above processes that are running. I su to the Gitorious user, git, in our case, and run the command from the Upstart file from the command line:
-bash-4.1$ chruby-exec ruby- — bin/rake resque:work QUEUE=*
I again ‘ps’ to find the running process:
root@srv-git-01 ~]# ps -ef | grep "resque" git 33861 33601 0 13:22 pts/0 00:00:00 /bin/bash -i -l -c chruby ruby- && bin/rake resque:work QUEUE=* git 33889 33861 56 13:22 pts/0 00:00:13 ruby bin/rake resque:work QUEUE=* root 33931 33903 0 13:22 pts/1 00:00:00 grep resque
I kill the parent process, in this case 33861:
[root@srv-git-01 ~]# kill -9 33861 [root@srv-git-01 ~]# ps -ef | grep "resque" git 33889 1 62 13:22 pts/0 00:00:26 resque-1.24.1: Waiting for * root 34055 33903 0 13:23 pts/1 00:00:00 grep resque
Now the child process is running and, as you can see from the second check, it’s in a ‘waiting’ state. And, in this state, it is working as I had expected it to.
I still haven’t figured out how to do this to run correctly at start. That’s the next step.
As an aside, when doing my latest search on this topic, I discovered that Upstart is going away:
http://unix.stackexchange.com/questions/118454/upstart-documentation-for-centos-6
I haven’t done more research to find out why. But maybe the alternative will solve my problem.
