Skip to content

[Bug]: Subagent completion silently lost — no retry, no notification, no auto-restart on timeout #44925

@IIIyban

Description

@IIIyban

Environment

  • Version: OpenClaw 2026.3.11
  • OS: Ubuntu 24.04, Telegram forum bot
  • Channel: Telegram (forum mode)

Describe the bug

Subagent task orchestration has multiple failure modes where results are silently lost:

Pattern 1: Completion announce fails (E31, E42, E45, E79)

Subagent completes the task successfully, but the announce-back to parent session fails. Result is lost. No error shown to user. This happened at least 9 times in production.

Typical log:

Subagent completion direct announce failed
Subagent announce give up (retry-limit)
Subagent orphan run pruned ... reason=missing-session-entry

Pattern 2: Subagent timeout with no auto-restart (E44, E110)

Subagent hits timeout. System does nothing — no retry, no restart, no notification. User waits indefinitely until they manually ask "what happened?"

Pattern 3: Subagent claims file created, but file does not exist (E61)

Subagent completion says "File created at /path/to/file.md" but the file is not on disk. No verification step.

Expected behavior

  1. If announce fails → retry with backoff, then surface error to user
  2. If subagent times out → auto-restart or notify user immediately
  3. File creation claims should be verified before announcing completion

Our incident IDs

  • E31 — Subagent result did not arrive, synthesis delayed
  • E42 — Subagent finished, HTML file not sent to chat
  • E44 — Subagent timeout, no auto-restart
  • E45 — Subagent finished, path shown but file not sent (9th E7-pattern occurrence)
  • E61 — Subagent claimed file created, file does not exist
  • E79 — File not sent after subagent completion
  • E110 — Subagent hung again + no status notification for 8 minutes

Related issues

Metadata

Metadata

Assignees

No one assigned

    Labels

    P1High-priority user-facing bug, regression, or broken workflow.clawsweeper:linked-pr-openClawSweeper found an open linked pull request for this issue.clawsweeper:needs-maintainer-reviewClawSweeper marked this issue as needing maintainer review before automation.clawsweeper:needs-product-decisionClawSweeper marked this issue as needing a product or behavior decision.clawsweeper:no-new-fix-prClawSweeper does not recommend queueing a new automated fix PR for this issue.clawsweeper:source-reproClawSweeper found a high-confidence source-level issue reproduction.impact:message-lossChannel message delivery can be lost, duplicated, or misrouted.impact:session-stateSession, memory, transcript, context, or agent state can drift or corrupt.issue-rating: 🦞 diamond lobsterVery strong issue quality with high-confidence source-level or clear reproduction.

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions