Skip to content

Prove temporary migration workers are empty before reusing them for V2 #5782

Open

Description

@devantler

🤖 Generated by the Agentic Engineer

The maintainer directed on 2026-11-02 that production volumes move to released V2 without waiting for upstream fixes. The current temporary workers are Ready but their Longhorn storage remains fenced. Reusing that unused capacity for V2 could let the source workers and their historical data remain intact during cutover; Kubernetes metadata alone does not prove the temporary disks are empty.

Extend the existing protected, read-only native replica inventory to the exact configured temporary migration pool. Resolve targets from the reviewed pool configuration and complete native node/storage/replica inventories, preserve UID, hardware and boot checks, and report only counts and commitments. Do not accept caller-supplied host addresses or broaden the permanent-worker reset operation.

Acceptance: confirmed first-attempt main dispatch can inventory each fixed temporary worker; incomplete pool membership, foreign identity, unfenced storage, API drift, native listing errors and unexpected on-disk content hold the observation. Existing permanent-worker observations and mutation authorization remain intact. Use the live observation to decide whether temporary capacity can safely be repurposed; a read-only receipt is not reset or data-migration clearance.

Part of #5582.

Activity

  1. added theissue type on Nov 2, 2026
  2. moved this to 🏃🏻‍♂️ In Progress in 🌊 Project Boardon Nov 2, 2026
  3. coderabbitai commented on Nov 2, 2026

    @coderabbitai
    Contributor
    🔗 Related PRs

    #5713 - feat(storage): inventory native V1 replica directories [closed]


    📝 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!

  4. self-assigned this
    on Nov 2, 2026
  5. added a commit that references this issue on Nov 2, 2026
  6. added 3 commits that reference this issue on Nov 2, 2026
  7. devantler commented on Nov 3, 2026

    @devantler
    ContributorAuthor

    🤖 Generated by the Agentic Engineer

    The three fixed temporary selections completed their first protected native storage-root observations at reviewed main 0d5b20a:

    Each observation found the four recognized cache directories, no files or unrecognized immediate entries, and no replicas entry. Every receipt is observation-only with reset clearance false. These observations do not prove an empty EPHEMERAL partition, the raw allocation, historical preservation or permission to erase data.

    The separately confirmed, read-only block-layout extension is being delivered with the host prerequisite diagnostic in #5792. The unused-worker reset implementation remains private; the native block facts and clean V1 baseline remain unmet.

  8. added a commit that references this issue on Nov 3, 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

No fields configured for Spike.

Projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions