🐛 fixed stale secret whiteout and SA reference - #147
Conversation
|
Warning Review limit reached
Next review available in: 38 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughThe Kubernetes plugin now whiteouts service-account-token and dockercfg Secrets and removes existing ChangesKubernetes resource handling
Estimated code review effort: 2 (Simple) | ~10 minutes Possibly related issues
Suggested reviewers: 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
transform/kubernetes/kubernetes_test.go (1)
1024-1104: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick winCover the
DisableWhiteoutOwnedoverride path.These cases use ownerless Secrets and leave
DisableWhiteoutOwneddisabled, so they do not lock in the “always whiteout” behavior. Add a case with owner references andDisableWhiteoutOwned: true, expectingIsWhiteOut: true.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@transform/kubernetes/kubernetes_test.go` around lines 1024 - 1104, Add a table-driven test case alongside the existing Secret cases, using an owner reference and setting DisableWhiteoutOwned to true, with the expected PluginResponse asserting IsWhiteOut: true. Ensure the test exercises the override path rather than the ownerless-secret behavior covered by the existing cases.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@transform/kubernetes/kubernetes_test.go`:
- Around line 1128-1144: Update the table-driven test runner around the
ServiceAccount case to assert that resp.Patches is empty whenever
PatchResponseJson is unspecified or empty. Preserve the existing expected-patch
comparison for cases that provide PatchResponseJson, and ensure
ServiceAccountWithoutSecretsFieldUntouched fails if any patch is emitted.
---
Nitpick comments:
In `@transform/kubernetes/kubernetes_test.go`:
- Around line 1024-1104: Add a table-driven test case alongside the existing
Secret cases, using an owner reference and setting DisableWhiteoutOwned to true,
with the expected PluginResponse asserting IsWhiteOut: true. Ensure the test
exercises the override path rather than the ownerless-secret behavior covered by
the existing cases.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro Plus
Run ID: d05b7ab7-506f-4734-9d52-a24985111bf0
📒 Files selected for processing (2)
transform/kubernetes/kubernetes.gotransform/kubernetes/kubernetes_test.go
|
Tested with OCP successfully |
2 participants
Summary
Whiteout auto-generated SA token Secrets (
kubernetes.io/service-account-token) and OCP dockercfg Secrets (kubernetes.io/dockercfg) by Secret type, regardless of ownerReferences. Also strip thesecretsfield fromServiceAccounts to remove dangling references to source-cluster token Secrets.
Problem
When migrating from Kubernetes < 1.24, auto-generated ServiceAccount token Secrets have no ownerReferences. The KubernetesPlugin only whiteouts resources that have ownerReferences, so these Secrets pass through the
entire pipeline and are applied to the target with stale source-cluster credentials (JWT signed by the wrong key, wrong CA cert, wrong SA UID).
On the target:
secretsreference.Same issue applies to OCP dockercfg Secrets on older OCP versions where they lack ownerReferences.
Changes
1. Whiteout Secrets by type (
getWhiteOuts())Added a type-based check before the ownerReferences check. Secrets of type
kubernetes.io/service-account-tokenandkubernetes.io/dockercfgare always whiteout-ed. These are cluster-generated credentials that arenever valid on a different cluster. The target's controllers regenerate them automatically.
User-created Secrets (
Opaque,kubernetes.io/tls, etc.) are unaffected.2. Strip SA
secretsfield (getKubernetesTransforms())Added a JSONPatch
remove /secretsfor ServiceAccounts that have asecretsfield. This field carries source-specific token Secret names that don't exist on the target. The target cluster's token controllerrepopulates it with fresh references.
Migration matrix
Fixes: crane#484
Summary by CodeRabbit