How to kill/stop openshift server process to free the service port? Error: "Address already in use"

Question

My openshift python-2.7 app was working ok few days ago but suddenly it got broken. I don't know the exact reason why. It might be a change of server configuration and my app was not probably following the proper way of WSGI deployment. Nevertheless I fixed it and pushed the change. But the service cannot be started because of the following error.

remote: deploy script running

remote: Starting Python 2.7 cartridge (Apache+mod_wsgi)

remote: Application directory "/" selected as DocumentRoot

remote: Application "wsgi.py" selected as default WSGI entry point

remote: (98)Address already in use: make_sock: could not bind to address 127.11.54.xxx:8080

remote: no listening sockets available, shutting down

remote: Unable to open logs

remote: -------------------------

remote: Git Post-Receive Result: failure

remote: Activation status: failure

remote: Activation failed for the following gears:

remote: 52814cb4500446c785000126 (Error activating gear: CLIENT_ERROR: Failed to execute: 'control start' for /var/lib/openshift/xxxxxcb4500446c7xxxxxxxx/python)

remote: Deployment completed with status: failure

remote: postreceive failed

I tried to kill the apache process but do not have access to do so. I tried to use rhc app stop -a myapp -a. But nothing paid off. What to do now to start the process? And how did it get into this situation in the first place?

Answer 1

This bug started started march 13 or so. It has affected my python 3.3 app also. Others and myself have posted about it with no response from openshift. I wish they would fix it. It the meanwhile, I have has luck using the 'rhc app-force-stop ' cmd before I do a push.

Answer 2

I think your apps is based on a server which is shared by other users also. So any one can login to SSH and kill httpd and mysql. When httpd is not running, it starts giving 503 error. This could be a security whole which still exist (in March 2014)