Skip to content
NewCybercrime Report 2026 · Eight Chapters · A Post Is Not An Incident → Explore The ChaptersCybercrime Report 2026 → Explore
Security NewsNews Explainer

Rackspace Spot Breach: Rotate Kubernetes Secrets and Tokens

Rackspace Spot breach response: replace tokens and kubeconfig files, rotate Kubernetes Secrets, and review cluster activity from November 17.

By
Brandefense Research Team
Reviewed by
Brandefense Analyst Team
Published
22 November 2026
Reading time
5 min read
A Kubernetes cluster protected by layered credential rotation

The Rackspace Spot breach requires a credential response from affected Kubernetes customers. Rackspace said leaked credentials gave an unauthorized party access to parts of Spot on November 17, 2026. Its customer notice, as relayed to BleepingComputer on November 21, says the party could create pods in some customer clusters and potentially read Secrets and Spot tokens.

The notice does not establish how many Secrets were read or how many customers were affected. Rackspace instructed affected customers to reissue Spot API tokens, replace automation refresh tokens, obtain fresh kubeconfig files, and rotate credentials in cluster Secrets. The sequence below turns those instructions into a scoped response plan; local evidence determines each organization’s actual exposure.

Rackspace Spot Breach: Which Access Should You Contain First?

Start a response record for every notified Spot organization and cluster. Preserve the notification, current token inventory, administrator changes, Kubernetes audit events, and downstream service logs before cleanup. Use November 17 as the initial investigation start because Rackspace dates the unauthorized access to that day; extend the window if your own evidence requires it.

Prioritize identities that can create pods or read Secrets, then suspend suspect sessions through the provider’s supported controls. Treat newly downloaded kubeconfig files as sensitive credentials and distribute them only through approved channels. This is a supplier access problem, so include the affected account and recovery owner in your third-party risk management platform record.

Brandefense Blackhole counted 364 dark-chat posts matching the broad term “Rackspace” between August 24 and November 22, 2026. The unit is matching posts, not distinct actors, customers, or breach reports. This broad count cannot establish discussion of this Spot incident or whether any customer Secret was read.

Spot credential response order
  1. Preserve the access recordKeep provider notice, audit events, token inventory, and service logs.
  2. Replace Spot accessReissue API tokens, refresh tokens, and kubeconfig files.
  3. Rotate reachable credentialsChange source-system values before updating Kubernetes Secrets.
  4. Hunt cluster changesReview pod creation and Secret access in the affected window.
  5. Validate downstream useCheck linked services and test old credential revocation.

How Do You Replace Tokens Without Leaving Old Automation Active?

Reissue each Spot API token and replace refresh tokens used by automation, following the Rackspace customer notice. Map each old token to its owner, job, permission set, and last use. Update the consuming job with its replacement, test the job, and then verify that the old credential can no longer authenticate.

Download fresh kubeconfig files for authorized operators and service workflows, as Rackspace advised. Inventory copies in build systems, deployment jobs, and administrator workstations before retiring old files. A kubeconfig can reference separate credentials, so review those referenced identities rather than treating a new file alone as proof of revocation.

Which Kubernetes Secrets Need Rotation?

List Secret names, namespaces, types, and owners without exporting their values. An authorized operator can use kubectl get secrets --all-namespaces for this inventory; Kubernetes documents the command and the Secret resource. Cross-reference each Secret with the identities and namespaces reachable by the affected Spot user.

Rotate the underlying database passwords, registry credentials, cloud keys, and service tokens at their issuing systems first. Then update the corresponding Kubernetes Secret and restart workloads that consume values as environment variables. Kubernetes notes that mounted Secret updates can propagate, while environment variables do not update in an existing process; check each workload’s delivery method.

Do not paste Secret values into incident tickets or command output shared with broad teams. Record only a secret’s identifier, owner, issuing system, rotation time, and verification result. Kubernetes warns that base64 encoding is not encryption and recommends tight access controls for Secret objects.

What Evidence Shows Whether Cluster Access Was Used?

Review Kubernetes audit events for unexpected pod creation and Secret read requests by the affected identity from November 17 onward. Compare the actor, namespace, object, time, and response status with approved deployments. Kubernetes auditing can record who made an API request and what resource it targeted, provided that the cluster’s audit policy captured the event.

Inspect pod inventory with kubectl get pods --all-namespaces and compare creation times and owners against deployment records. Preserve object details before removing unauthorized workloads through your incident procedure. Check service account bindings, node listings, and volume activity because the Rackspace notice describes those as possible capabilities, not proven actions in every cluster.

The cloud account access technique provides a separate lens for reviewing authorized identity use. A successful login under a legitimate account does not establish that all subsequent activity was approved. An audit gap also does not establish that a Secret was untouched.

How Do You Close the Response?

Rackspace advised examining logs for databases, container registries, Git repositories, cloud storage, and third-party services from November 17. Match access under rotated credentials to expected jobs and business hours, and investigate unexplained reads or changes. If a downstream system shows suspicious use, widen the response to that system’s owner and incident process.

For each cluster, record which Spot credentials were replaced, which Secret-backed credentials were rotated, and whether old values were rejected. Confirm applications work with new values before closing the incident record. Use the incident response plan template to track owners, evidence, decisions, and unresolved scope.

After immediate containment, reduce standing Secret access with role-based permissions and review long-lived service account tokens. Kubernetes recommends short-lived, automatically rotating TokenRequest tokens instead of legacy long-lived ServiceAccount token Secrets. Those improvements reduce future exposure, while the incident finding still depends on the preserved provider and cluster records.

Frequently Asked Questions

Does the Rackspace Spot notice prove a customer Secret was read?

It describes possible Secret access in affected clusters. The notice does not quantify actual Secret reads; use your Kubernetes audit and service logs to assess local scope.

Is downloading a new kubeconfig enough?

No. Review the credentials referenced by old files, replace them through supported controls, and verify that old access fails.

Where should the investigation begin?

Start with November 17, 2026, the date Rackspace gave for unauthorized access. Preserve cluster audit events and linked service logs, then extend the window if local evidence warrants.

Sources

  1. Kubernetes, Secrets
  2. Kubernetes, Good practices for Kubernetes Secrets
  3. Kubernetes, Auditing
  4. Kubernetes, Managing Service Accounts
  5. Kubernetes, kubectl get
Share this articleLinkedInXEmail
Brandefense logo

Stay Ahead of Digital Threats.

Take control of your digital security with an exclusive demo of our AI-driven threat intelligence platform.

Open The Full Demo Form →