The typical end-to-end (E2E) encryption method used for secure cloud computing is not enough anymore because there are new inference/frequency analysis attacks that can be performed by the cloud providers. We need oblivious computation together with E2E encryption for safe and scalable cloud computing. My Master Capstone Project with Professors Jose Renau and Owen Arden of University of California, Santa Cruz and Baskin Engineering at UCSC (called Locker 2.0), adds oblivious computation to Kubernetes and Etcd that can be used with Kubernetes Secrets to enable almost-perfect security (except for client side-channel and currently-unknown attacks). The article submitted describes Locker 2.0 in detail.