Skip to content

🐛 Adopted conflicting PR sits on Rebasing attempt 1/2 indefinitely while the fixer is busy #16023

Closed

Description

@clubanderson

What happened

#15998 (clanker-authored, author offline) was adopted in place after operator approval at 06:47 ET with mergeable_state=dirty. The tracker row read Adopted by hive · Rebasing · attempt 1/2 for 25+ minutes with no push to the branch.

Logs (grep 15998) show PR routed to fixer for rebase … "attempt":1,"max":2,"fixer":"scanner" at 07:02:03 and again at 07:11:26 — every sweep re-routes the same attempt to scanner, which is busy with other work.

Root cause

  • pkg/rebasemerge.Attempts only advances an attempt when a new dirty head is observed (Observe). If the fixer never pushes, the episode never leaves attempt 1, so Exhausted() never fires and nothing escalates.
  • writeMergeEligible lists the row in ci-failing.json every sweep (that is how the fixer's kick overlay sees it) and logs "routed to fixer" every time; nothing records that the attempt was already dispatched or how long it has waited.
  • The in-place adoption's only escape to a superseding PR (AdoptReasonRebaseRehome) requires the 074 dwell (3h) via adoptionStale, so a conflict the fixer never picks up stalls silently for hours.

Expected

  • Per-attempt deadline (default 15m, governor.labels.stage.adoption.rebase_attempt_timeout); an attempt with no push past it counts as failed and attempt 2 is dispatched with one INFO line.
  • The attempt is announced once; later sweeps keep the row listed but do not re-route it.
  • Past the cap, an in-place adoption re-homes to a hive-owned superseding PR (rebase-rehome) at once rather than after the 3h dwell.
  • The row shows the stall: "Rebasing · attempt 1/2 · waiting for scanner 12m".

Follow-up (not in scope): a dedicated / alternate fixer for attempt 2 (the fallback fixer is hard-coded to scanner in routeConflictFixAgent), or an in-process git rebase for trivial conflicts.

Activity

  1. added
    help wantedDenotes an issue that needs help from a contributor. Must meet "help wanted" guidelines.
    on Oct 31, 2026
  2. hivecommons-hive commented on Oct 31, 2026

    @hivecommons-hive
    Contributor

    🔒 Claimed by agent scanner (agent on hive hosted-available-oke-11-placeholder-r05x). The claim lasts 2h0m0s without activity and renews while the work continues. Reply /unclaim to release, or /claim to take it over if you outrank the holder.

  3. github-actions commented on Oct 31, 2026

    @github-actions
    Contributor

    @clubanderson this issue was closed by @hivecommons-hive[bot] as completed.

    If this was closed by a merged PR, the fix only takes effect once it ships in a tagged release and your deployment updates to it — check your running version against the fix before concluding the fix failed.

    If you are still seeing the problem, comment /reopen and it will be reopened for you — you do not need write access. Please add what you are still observing (running version, logs, steps) so it is not closed the same way again.

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

    bugSomething isn't workinghelp wantedDenotes an issue that needs help from a contributor. Must meet "help wanted" guidelines.

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions