The status of container always "starting" after upgrade

Hi everybody, I’m happy to use rancher. I want to use persistent volume with rancher but rancher-nfs is not good for performance. My company is using ScaleIO so I have setup rexray work as volume driver to connect to ScaleIO. Everything is OK except one thing. When I upgrade my service using a volume from rexray driver (or flocker driver), Rancher can’t check health the container. Althought I saw that container was up.
Does anybody use rancher with rexray or flocker. Please help me. Sorry for my bad english. I’m vietnamese. Thanks