Skip to content

[PR-20] Selective Server-Side Apply Pilot #388

Closed as not planned
Closed as not planned

Description

@amuraru

Delivery bundle: C: Convergence, APIs, and gated scalability. See #395 and #368 for the four-PR stack.

Parent: #393
Roadmap: #368
Plan item: PR-20
Priority: P3
Dependencies: PR-07

Kubernetes support: Rolling upstream-maintained minors; use the oldest/newest boundaries pinned in config/kubernetes-versions.env. See #368 for the current policy. Unrun jobs are not passed validation.

Problem Statement

Evidence: pkg/k8sutil/resource.go:46-160 combines objectmatcher annotations, full updates, Service field preservation, and broker-config status side effects.

Proposed Solution

Files: That helper and tests, the selected ConfigMap or Service generator/callers, and compatibility tests/docs. Do not remove objectmatcher globally.

Interface contract: Pick one low-risk resource family. Use modern Client.Apply with explicit field ownership and only owned intent. Preserve defaulted/allocated fields, user metadata, and existing broker-config drift/rolling-upgrade behavior.

  • Add tests for server defaulting, ClusterIP/allocated fields if applicable, unrelated user annotations, field conflicts, field removal, no-op apply, and migration from existing last-applied annotations.
  • Explicitly test ordered storage/log-directory configuration and PVC retention remain unchanged by the pilot.
  • Define ownership and takeover policy; do not force ownership of unrelated fields or change CRD list topology as incidental cleanup.
  • Compare API traffic, controller CPU, server-side cost, and convergence with the existing implementation.
  • Validate focused resource tests, envtest against real apply semantics, and relevant Kind rollout tests.

Done when: A small resource family has proven compatible apply behavior and measured justification, or the pilot is deferred.

Delivery: Four consolidated PRs, approved November 16, 2026: A -> B -> C; D branches from A. Foundation is #395. The 22 PR-XX identifiers are work-item IDs, not separate PRs. Preserve every work-item acceptance criterion and native dependency. No gated experiment is automatically enabled, and no issue is complete merely because one PR references it. D must satisfy its native #375 gate, including lifecycle-domain signals assigned to B; branching from A does not waive that merge prerequisite. Write focused regression tests first; record exact commands/results and run make check plus this issue's real-cluster/measurement gates.

Alternatives Considered

Twenty-two individual PRs add unnecessary integration overhead, while one omnibus PR obscures correctness and rollout boundaries. Deliver this scope in its approved dependency-aligned bundle. Do not expand into deferred consistency, storage, or concurrency changes without their own decision gates.

Additional Context

Baseline: 95138e0d; detailed roadmap scope: #368, section PR-20. The roadmap #368 and its native sub-issues are the planning source of truth. No planning document is committed in the final PR diffs. Re-read affected source and overlapping work before implementation. Historical baseline test passes are not validation of this change.

Activity

  1. added theissue type on Nov 16, 2026
  2. amuraru commented on Nov 17, 2026

    @amuraru
    Author

    SSA experimentation is removed from active scope under #368: no demonstrated benefit justifies ownership/migration complexity. Measured efficiency remains in #393 without a new apply architecture.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions