Hello Linux Admins. I need some help. I have a "customer" that runs a LAMP stack/jquery server on my virtual infrastructure. They have been having intermittent problems with the box. There is some runaway process that chews up all of the memory/cpu for several minutes and the box gets unresponsive. They usually end up bouncing it before it comes back to life but it never truly hard locks.
They insist that VMWare is the smoking gun, which doesn't make sense. What can I do on the box (RHEL 6) to monitor specific applications and processes so I can find out what is causing this?
Edit: User/Customer provided this from one of their external monitoring tools:
http://i.imgur.com/cCZYsDb.jpg
Edit2: messages data that seems important: http://paste.openstack.org/raw/75780/
[link][16 comments]