Skip to content

Fail template sync when its source pull fails #887

Closed

Description

@devantler

🤖 Generated by the Agentic Engineer

Problem

A failed source pull can leave Template Sync green and without a proposal. A maintainer then sees a successful run even though the consumer never received the template changes.

Evidence

The pinned upstream action catches the nonzero pull, writes warnings and returns the status of target sign-in. Its inputs do not expose a fail-on-pull option. The original tag-collision case in #861 reached that path. The released no-tag parameters fix that cause, as confirmed by a real KHoist sync, while other failed pulls can still reach the same path.

Acceptance

  • A fetch, authentication or merge failure reaches a failed native workflow verdict and does not produce a misleading no-change receipt.
  • A genuinely up-to-date consumer can still report success without a proposal. A normal change still creates a valid signed proposal and preserves caller-owned files and tags.
  • Exercise the real execution boundary with unavailable source, conflicting merge and up-to-date controls. Preserve the reviewed authentication, signing and caller scopes.
  • Prefer an upstream supported outcome if available. Otherwise make the smallest catalogue-owned change with current-head CI and review, then verify a normal consumer run.

Follow-up to #861. The tag-collision repair and its consumer adoption are already delivered. This work does not require the KHoist caller to change its released tag protection.

Activity

  1. added theissue type on Nov 9, 2026
  2. coderabbitai commented on Nov 9, 2026

    @coderabbitai
    ⚠️ Possible Duplicate Issue(s)


    📝 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. moved this to 📥 Backlog in 🌊 Project Boardon Nov 9, 2026
  4. added a commit that references this issue on Nov 10, 2026
  5. moved this from 📥 Backlog to 🏃🏻‍♂️ In Progress in 🌊 Project Boardon Nov 10, 2026
  6. devantler commented on Nov 11, 2026

    @devantler
    ContributorAuthor

    🤖 Generated by the Agentic Engineer (interactive Claude Code session)

    Done. The fix from #891 is released (v20.1.20, also in v21.0.0), and the last open point, a normal consumer run, is now verified on devantler-tech/ascoachingogvaner, which I moved to v21.0.0 in devantler-tech/ascoachingogvaner#366.

    One thing surfaced on the way: the first attempt of the first run was marked failed by the signing step's branch readback, although the proposal and its signature were correct. That is a separate defect, tracked in #897, with a fix prepared.

    🤖 Generated with Claude Code

  7. moved this from 🏃🏻‍♂️ In Progress to ✅ Done in 🌊 Project Boardon Nov 11, 2026
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

    Fields

    Priority

    None yet

    Effort

    None yet

    Projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions