Skip to content

Restore replica scheduling headroom for the next declared V1 claim #6282

Closed

Description

@devantler

🤖 Generated by the Agentic Engineer

Problem

The existing bounded V1 reservation repair handles only its original pair of claims. A later legitimate claim can exhaust a disk's logical replica budget and remain below its requested redundancy even when the physical free-space safeguards permit another copy. The V2 migration must hold until the source volumes have their required healthy copies.

Outcome

Restore the missing copy through the existing protected production workflow, with a narrowly reviewed allocation plan for the declared affected claim. Preserve the complete disk ledger, physical free-space floor, original data and all existing recovery gates.

Acceptance

  • Reproduce the refused current case and distinguish it from the historical repair.
  • Bind any reservation adjustment to complete current inventory, unchanged source and identities, two intact peer copies, the exact missing allocation, and the existing free-space margin.
  • Keep unrelated or ambiguous degraded volumes, competing maintenance, changed identities and uncertain writes closed.
  • Verify the original affected volume regains all requested healthy replicas before accepting recovery or continuing migration.

Part of #5582.

Activity

  1. added theissue type on Nov 14, 2026
  2. coderabbitai commented on Nov 14, 2026

    @coderabbitai
    Contributor
    🔗 Related PRs

    #5969 - fix(storage): restore capacity for missing V1 replicas [merged]


    📝 Issue Planner

    Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.

    • Create Plan

    🧪 Issue enrichment is currently in open beta.

    You can configure auto-planning by selecting labels in the issue_enrichment configuration.

    To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

    issue_enrichment:
      auto_enrich:
        enabled: false

    💬 Have feedback or questions? Drop into our discord!

  3. self-assigned this
    on Nov 14, 2026
  4. moved this to 🏃🏻‍♂️ In Progress in 🌊 Project Boardon Nov 14, 2026
  5. added 2 commits that reference this issue on Nov 14, 2026
  6. devantler commented on Nov 15, 2026

    @devantler
    ContributorAuthor

    🤖 Generated by the Agentic Engineer

    The original affected encrypted source volume now has its full requested redundancy: the backend is healthy, three replicas run on distinct workers, and the engine reports all three as RW. Fresh read-only native inventories confirm the same original volume identity. Normal scheduling recovered the missing copy after the first retained recovery copy was removed through #6041.

    No reservation repair was needed or dispatched. The unused draft #6289 is closed. Closing this outcome as recovered; #5582 remains open for the full V2 migration and the remaining cleanup. Detailed native receipts remain private.

  7. moved this from 🏃🏻‍♂️ In Progress to ✅ Done in 🌊 Project Boardon Nov 15, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

Fields

Priority

None yet

Effort

None yet

Projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions