Openshift Private Cloud

Published
Length
3 min read

Luis Felipe Dias

All posts

You set up an openshift private cloud by defining roles on your own hardware. These roles control who can run apps, view logs, and enter running pods. Many teams start by giving engineers broad cluster permissions to move fast. Broad roles let one bad command break live apps or expose private data. To keep your cluster safe, you must scope permissions down to single projects. You also need clear rules for who can open a shell inside a pod. Let us walk through how native role rules work and wher

Your coding agents run with your access.

Claude Code reaches prod with a developer's credentials, and approval prompts get clicked through. hoop.dev checks every query, masks sensitive data, and holds risky changes for a human.

You set up an openshift private cloud by defining roles on your own hardware. These roles control who can run apps, view logs, and enter running pods. Many teams start by giving engineers broad cluster permissions to move fast. Broad roles let one bad command break live apps or expose private data.

To keep your cluster safe, you must scope permissions down to single projects. You also need clear rules for who can open a shell inside a pod. Let us walk through how native role rules work and where gaps remain.

How roles work in an openshift private cloud

An openshift private cloud uses Kubernetes role-based rules to manage user actions. Every rule in this system is additive, so there are no deny rules. You grant rights by creating a role and binding it to a user.

A standard role applies only to one namespace, which keeps teams isolated. A cluster role covers the whole cluster, including nodes and all namespaces. You use a role binding to give a user permissions inside one namespace. A role binding can point to a local role or a cluster role. If a user needs rights everywhere, you use a cluster role binding instead.

Permissions also cover smaller targets called subresources, such as pod logs or pod shells. In the API, these subresources attach to the main resource name. You must grant specific rights for a user to read pod logs. You must also grant separate rights before a user can run commands in pods.

What native roles solve and what they miss

Native roles are great at deciding who can reach the cluster API. They stop a developer in one namespace from deleting pods in another namespace. They also let you separate read-only viewers from admins who change cluster state.

OpenShift RBAC + Virtual Private Database: architecture patterns and best practices

Continue reading? Get the full guide.

Free. No spam. Unsubscribe anytime.

Still, native role checks stop the moment a pod shell session starts. Once RBAC allows a user into a pod, the API stops checking commands. The cluster knows a session started, but not what the user typed inside. If the pod holds database keys, the user can read or copy them.

Native roles also cannot inspect or filter the output of a running command. If an engineer queries a database from a pod, raw data prints out. Emails, phone numbers, card numbers, and national IDs show up on the screen. Also, static kubeconfig files often sit on laptops for months without expiring. You need a way to check single commands and mask sensitive fields in flight.

Govern kubectl and API calls in flight

Identity providers and Kubernetes RBAC decide who logs into a cluster. They cannot redact keys inside returned JSON objects or hold a risky delete call for review. By design, hoop.dev sits inline as an open-source proxy, a sidecar for your Kubernetes API, without replacing RBAC. On every production call from an engineer or AI agent, hoop.dev decides to pass, mask, block, or hold it for human approval.

First, hoop.dev guardrails refuse banned methods, paths, or headers before they run or hold risky deletes for review. Next, hoop.dev data masking rewrites sensitive keys in returned Secret, ConfigMap, and log responses in memory, and writes each decision to the audit trail. Explore the repo on GitHub.

Frequently asked questions

How do you limit user permissions to a single namespace?

You create a role or pick an existing cluster role with the needed rules. Then you link that role to the user with a namespace role binding. This setup keeps the user's permissions inside that single namespace.

Can you add a deny rule to block one specific action?

No, native Kubernetes role permissions are purely additive and have no deny rules. To block an action, you must remove the rule that grants that permission. Start with zero permissions and add only the exact verbs each team needs.

Why is pod shell access risky even with strict role bindings?

Role bindings only check if a user is allowed to open the pod shell. They do not check the shell commands or mask sensitive data like card numbers. A user with shell rights can run any tool installed inside that container.

The query runs. The emails stay hidden.