Openshift Ceph Storage

Published
Length
3 min read

Luis Felipe Dias

All posts

You set up openshift ceph storage by deploying storage pods and storage classes in your cluster. Then you grant your team roles so apps and admins can claim those volumes. Most platform teams start by binding broad cluster roles to get storage running fast. Overly broad roles let users open shells on storage pods and read raw disks. How openshift ceph storage permissions work Storage in this platform runs as a set of pods inside a dedicated namespace. Kubernetes role-based access control gove

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 openshift ceph storage by deploying storage pods and storage classes in your cluster. Then you grant your team roles so apps and admins can claim those volumes. Most platform teams start by binding broad cluster roles to get storage running fast. Overly broad roles let users open shells on storage pods and read raw disks.

How openshift ceph storage permissions work

Storage in this platform runs as a set of pods inside a dedicated namespace. Kubernetes role-based access control governs who can view or change these storage resources. A Role sets permissions inside one namespace, while a ClusterRole covers the whole cluster. Rules in these roles are purely additive, which means there are no deny rules.

You attach permissions to users or service accounts with a RoleBinding or a ClusterRoleBinding. A RoleBinding grants access inside one namespace, and it can reference a ClusterRole. A ClusterRoleBinding grants access across every namespace in the cluster at once. Admins need permissions on storage classes, persistent volumes, and the storage namespace itself.

App teams only need to create persistent volume claims inside their own app namespaces. Still, day-to-day storage fixes often require someone to check storage health directly. Admins use toolbox pods and subresources like pod logs or pod exec to run checks. In Kubernetes RBAC, you grant access to these subresources using a slash in the rule.

Why native roles leave a gap during live fixes

Native cluster roles do a great job of deciding who can reach the API server. They control which user can list volumes, read pod logs, or start an exec session. Once a user opens an exec session on a storage pod, Kubernetes steps back. The API server checks the initial request to open the shell, then stops watching.

Inside that storage toolbox pod, every command runs with the full privileges of the container. A user who only needs to check disk status can also run destructive storage commands. Kubernetes RBAC cannot inspect or block individual shell commands inside an active pod session. It also cannot mask sensitive fields like emails or card numbers in command output.

OpenShift RBAC: architecture patterns and best practices

Continue reading? Get the full guide.

Free. No spam. Unsubscribe anytime.

Shared service accounts and static kubeconfig files make this problem even harder to track. When three admins share one cluster role to fix a volume, audit logs lack detail. You see that someone opened a shell on the storage pod, but nothing after that. To run safe storage operations, teams need control over what happens inside the session itself.

Tightening storage access across namespaces

Start by splitting daily app work from cluster-wide storage tasks in your role bindings. Give app developers a standard RoleBinding that only covers persistent volume claims in their namespace. Keep them out of the core storage namespace where the underlying Ceph pods run. Remove pod exec rules from general developer roles so users cannot jump into storage containers.

For platform admins who must run toolbox commands, avoid long-lived cluster admin bindings. Scope their Role to the exact storage namespace and restrict which resources they can modify. Keep in mind that Kubernetes does not let you restrict subresource creation by specific resource names. Because of that limit, native roles alone cannot lock down which commands run inside a pod.

Check Kubernetes API requests and responses on the wire

Kubernetes RBAC decides who may call the API server. RBAC does not check response payloads or split listing Secret names from reading Secret values when a role grants get access. With hoop.dev, you run an open-source HTTP proxy, a sidecar for your Kubernetes API. The kubectl client changes only the server address it talks to while RBAC stays in place.

On each call, hoop.dev guardrails read the method, normalized path, and headers to refuse bad deletes or full Secret reads with a 403 error. On the way back, hoop.dev data masking walks JSON responses by key and masks matched values in Secrets, ConfigMaps, and pod logs. Every request and verdict is logged without storing bearer tokens. View the project on GitHub.

Frequently asked questions

How do I let developers claim storage without giving them cluster admin rights?

Create a RoleBinding inside the developer namespace that only grants permissions on persistent volume claims. The cluster handles the underlying volume creation by default through the storage class.

Can Kubernetes RBAC block specific Ceph CLI commands inside a toolbox pod?

No, Kubernetes RBAC only checks if a user has permission to start the pod exec session. Once the shell opens, Kubernetes does not read or filter the commands the user types.

Why can I not use deny rules to block access to the storage namespace?

Kubernetes RBAC permissions are purely additive and do not support deny rules at all. To block access, you must avoid binding broad cluster roles and grant namespace roles instead.

The query runs. The emails stay hidden.