How to Configure OpenShift Break Glass Access
You set up openshift break glass access by binding emergency cluster roles to temporary users. When the incident ends, your team removes those role bindings right away. Many teams start by giving a shared admin role to an on-call group. This broad setup creates big risks when an emergency hits. Anyone with the role can touch every app and secret in the cluster. A small mistake during a high-stress outage can delete live workloads. Teams need a clear path to grant and revoke emergency power fast.
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 break glass access by binding emergency cluster roles to temporary users. When the incident ends, your team removes those role bindings right away. Many teams start by giving a shared admin role to an on-call group. This broad setup creates big risks when an emergency hits. Anyone with the role can touch every app and secret in the cluster. A small mistake during a high-stress outage can delete live workloads. Teams need a clear path to grant and revoke emergency power fast.
How openshift break glass access works in native RBAC
OpenShift uses the standard Kubernetes role system to control who can do what. A role holds a list of rules that grant specific permissions. In this system, permissions only add power and never use deny rules. A standard role sets rules inside one single namespace. A cluster role works across the whole cluster and covers nodes and all namespaces.
To grant emergency access, you link a role to a user or group. A role binding gives those permissions inside one specific namespace. A cluster role binding gives those permissions across the entire cluster at once. For a single app outage, a team should use a namespace role binding. This keeps the emergency user inside the broken project.
Some emergency tasks need access to pod logs or shell commands. The role system treats these actions as subresources under the main pod resource. You must add specific rules for log reading and command execution. Once you apply the binding, the on-call engineer can log in and fix the issue. When the fix is done, an admin deletes the binding to end the session.
What native roles solve and where gaps remain
Native role bindings solve the basic problem of who can enter the cluster. They let you keep daily accounts locked down to read-only views. You only hand out write power when an outage demands fast action. By using namespace bindings, you also limit how many apps an engineer can touch.
Still, native role bindings stop working once the user is inside the pod. The role system only checks if a user can start an execution session. It does not check the individual shell commands that run inside that session. An engineer with pod execution rights can run any command they want. They can read local files, change app state, or dump database tables.
Break-Glass Access Procedures + Customer Support Access to Production: architecture patterns and best practices
Continue reading? Get the full guide.
Free. No spam. Unsubscribe anytime.
Native roles also cannot filter the data that comes back to the screen. When an engineer reads logs or queries a store, raw data flows back. Customer emails, phone numbers, and card numbers appear in plain text. Finally, someone must remember to delete the role binding after the incident. If the team forgets, that emergency account keeps the session active for days or weeks.
Managing emergency sessions with better controls
A strong emergency process needs more than static role bindings. First, teams should require a peer approval step before any break glass role starts. Second, every emergency role should have a hard timer that deletes the access. This stops standing privileges from piling up after busy nights.
Teams also need visibility into what happens after the pod shell opens. Logging the start of a session is not enough for a security review. You need a clear record of every command run during the break glass window. Redacting sensitive fields like national IDs in output results also lowers data risk. Combining tight role scope with command-level checks makes emergency fixes much safer.
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 proxy, a sidecar for your Kubernetes API, inline on the API connection. On every kubectl or API call an AI agent or engineer runs against production, hoop.dev decides to pass, mask, block, or hold it for human review.
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, or hold risky calls for approval. On the way back, hoop.dev data masking walks JSON responses by key and masks matched values in Secrets, ConfigMaps, and pod logs in memory. View the project on GitHub.
Frequently asked questions
What is the difference between a role binding and a cluster role binding?
A role binding grants permissions inside one specific namespace only. A cluster role binding grants permissions across all namespaces and cluster nodes.
Can I write a deny rule to block one secret during break glass?
No, the native role system is purely additive and has no deny rules. You must build roles that include only the exact resources the user needs.
How do I limit break glass users from running pod commands?
Leave the pod execution subresource out of the emergency role definition. Give them the pod log subresource instead so they can still read output.
Related guides
In this post