fix: CLI and Gateway processes can hang forever at exit on Node 24 and 26 by steipete · Pull Request #163788 · openclaw/openclaw

steipete · 2026-10-02T21:08:10Z

Related: nodejs/node#64274 (upstream fix pending in nodejs/node#66171)

What Problem This Solves

Fixes: openclaw commands (and the hook relay or a stopping Gateway) can hang forever at 0% CPU after they have already printed their output, on Node 24 and 26.

User Impact

User impact: CLI commands, Codex hook relays, and Gateway stop/restart exits finish instead of occasionally wedging forever. No configuration changes; operators who want V8's Maglev or concurrent Sparkplug back can pass --maglev / --concurrent-sparkplug to node.

Release note: CLI, hook relay, and Gateway processes no longer hang at exit on Node 24/26 (V8 compile-job deadlock in process.exit()).

Why This Change Was Made

process.exit() joins V8's background workers without disposing the isolate. If a concurrent Maglev or Sparkplug compile job is waiting for a main-thread garbage collection at that moment, the join never returns. Node 24 turned both compilers on by default; Node 22 had neither, and our Vitest workers already run with both off for this reason.

Every OpenClaw executable entrypoint (entry, the Gateway/service index, the native hook relay, the macOS node worker) now switches both off at startup when it runs as the main module. Library imports keep stock V8, Bun is skipped, and explicit operator flags win. V8 reads both flags per decision, so setting them at entry takes effect (--trace-opt: Maglev compiles in models list --json drop from 689 to 50, all before the entry runs).

Alternatives checked: letting the loop drain via process.exitCode works for probed commands (0–139 ms) but macOS system-CA natural exit hangs in Keychain shutdown (#113112) and ~246 direct runtime.exit call sites remain, including the gateway health route that hung. Adding flags to the respawn plan would miss the hook relay, gateway run/the service, gateway status, and the TUI.

Evidence

Blacksmith Testbox, 4 vCPU, Node 24.19, replay of the test/cli-json-stdout Gateway CLI children (gateway health --json, fresh state per run, 8-way). Every hang's gdb stack: main thread in WorkerThreadsTaskRunner::Shutdown → uv_thread_join, a V8 worker parked in CollectionBarrier::AwaitCollectionBackground.

Configuration	Hung	Parked job
Both compilers on (current behavior)	6/1019	5 Sparkplug, 1 Maglev
Only concurrent Sparkplug on	12/944	Sparkplug
Only Maglev on	2/1019	Maglev
This change (both off)	0/1019	—

Cost (same boxes): concurrent Sparkplug off is free; Maglev off costs about +5% CLI wall time sequentially and +10–20% at 8-way on 4 vCPUs. Gateway start→ready +1–5% sequential and faster under 6-way load (p50 15.0/16.0 s vs 20.7/17.4 s); SIGTERM→exit unchanged; the log-redaction hot-path benchmark is unchanged (steady state still reaches TurboFan). Gateway start/SIGTERM cycles: 0/240 exit hangs either way.

Also: Codex autoreview scoped-clean, tsgo:core, oxlint/oxfmt on touched files, runtime build including the bootstrap-import check. Warm-home commands without the e2e environment did not reproduce the hang on Linux (0/4000 both arms); the hang was also observed on macOS CLI children and in the CI e2e shard.

No unit test: the contract is process-level (a mocked setFlagsFromString test would only mirror the call); the A/B above exercises the real entrypoint. Follow-up: Gateway-spawned workers with their own entry files are not covered; retire this policy once supported Node floors ship nodejs/node#66171.

On Node 24 and 26, process.exit() joins V8's platform workers without disposing the isolate. A concurrent Maglev or Sparkplug compile job parked waiting for a main-thread GC then never finishes, so CLI commands, the hook relay, or a stopping Gateway can sit at 0% CPU forever after printing their output (nodejs/node#64274; fix pending in nodejs/node#66171). Every OpenClaw executable entrypoint (CLI entry, Gateway/service index, native hook relay, macOS node worker) now turns off Maglev and concurrent Sparkplug at startup when it runs as main, matching Node 22's tiering and the existing Vitest policy. Library imports keep stock V8; explicit --maglev or --concurrent-sparkplug flags still win. Testbox A/B (4 vCPU, Node 24.19, e2e CLI child replay, 8-way): 6/1019 hung with defaults, 0/1019 with the policy; re-enabling only Maglev hung 2/1019 and only concurrent Sparkplug 12/944, so both flags are needed.

clawsweeper · 2026-10-02T21:08:14Z

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

ClawSweeper review complete

ClawSweeper finished reviewing this revision. The review result is being finalized.

View the workflow run.

clawsweeper · 2026-10-02T21:16:17Z

Codex review: needs maintainer review before merge. Reviewed October 2, 2026, 5:15 PM ET / 21:15 UTC.

ClawSweeper review

What this changes

The PR disables two Node background compilers at OpenClaw executable startup to prevent exit hangs, preserves explicit operator flags, and documents the behavior.

Merge readiness

✅ Ready for maintainer review

This PR remains useful: current main lacks the mitigation, and the supplied production-path replay supports it. No blocking patch defect was found.

Likely related people: steipete and SunnyShu0925 (medium-confidence routing candidates).

Priority: P1
Reviewed head: 4d81d08ec1677f518a0b1d5b3892337c120f8fca

Review scores

Measure	Result	What it means
Overall readiness	🐚 platinum hermit (4/6)	A focused implementation with strong production-path A/B evidence and no identified blocking defect.
Proof confidence	🦞 diamond lobster (5/6)	Sufficient (live_output): The captured Testbox A/B exercises real gateway health --json children through the changed CLI bootstrap and exit boundary on Node 24.19, improving observed hangs from 6/1019 to 0/1019. Separate compiler arms and debugger stacks support causality; Gateway SIGTERM cycles are supplemental. No stored-data contract changes.
Patch quality	🐚 platinum hermit (4/6)	No actionable review findings were identified.

Verification

Check	Result	Evidence
Real behavior	Verified	Sufficient (live_output): The captured Testbox A/B exercises real gateway health --json children through the changed CLI bootstrap and exit boundary on Node 24.19, improving observed hangs from 6/1019 to 0/1019. Separate compiler arms and debugger stacks support causality; Gateway SIGTERM cycles are supplemental. No stored-data contract changes.
Evidence reviewed	10 items	Pinned introduced change: Read the complete merge-base-to-head diff. One shared helper supplies the policy to the CLI entry, executable index, native hook relay, and macOS node worker; the remaining change documents that policy. Runtime and settings boundary: The helper skips Bun and preserves explicit --maglev and --concurrent-sparkplug arguments, including underscore spellings. Entry, index, and macOS worker calls are guarded by main-module detection. No stored configuration, schema, authorization, CA policy, or exit-finalization contract is changed. Current-main necessity: Current main uses these compiler flags in the Vitest process environment, but has no equivalent production bootstrap policy. This PR extends the mitigation to executable processes rather than duplicating an existing production owner.
Findings	None	None.
Security	None	None.

How this fits together

OpenClaw's executable bootstrap prepares Node before CLI commands, Gateway services, hook relays, and the macOS node worker run. Its compiler policy affects whether those processes can finish shutdown after their work and output complete.

flowchart TD
  A[Executable invocation] --> B[OpenClaw bootstrap]
  B --> C[Node runtime and explicit flags]
  C --> D[Disable unsafe background compilers]
  D --> E[Command or service runs]
  E --> F[Output and resource cleanup]
  F --> G[Process exits]

Before merge

None.

Agent review details

Security

None.

PR surface

Source +36, Docs +4. Total +40 across 6 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	5	42	6	+36
Tests	0	0	0	0
Docs	1	4	0	+4
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	6	46	6	+40

Review metrics

None.

Technical review

Best possible solution:

Use one executable-bootstrap compiler policy while retaining operator opt-ins, then retire it when supported Node versions contain the upstream shutdown fix.

Do we have a high-confidence way to reproduce the issue?

Yes: the supplied Node 24.19 Testbox replay identifies a concrete, intermittently failing CLI-child exit path with debugger stacks. This read-only review did not execute a current-main reproduction.

Is this the best way to solve the issue?

Yes: a shared policy at executable startup addresses the evidenced compiler deadlock without replacing lifecycle cleanup or macOS CA handling.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against 6347c0c51250.

Labels

Label changes:

add P1: Observed exit deadlocks can stall real CLI workflows and Gateway stop or restart operations on supported Node runtimes.
add proof: sufficient: Contributor real behavior proof is sufficient. The captured Testbox A/B exercises real gateway health --json children through the changed CLI bootstrap and exit boundary on Node 24.19, improving observed hangs from 6/1019 to 0/1019. Separate compiler arms and debugger stacks support causality; Gateway SIGTERM cycles are supplemental. No stored-data contract changes.
add rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
add status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (live_output): The captured Testbox A/B exercises real gateway health --json children through the changed CLI bootstrap and exit boundary on Node 24.19, improving observed hangs from 6/1019 to 0/1019. Separate compiler arms and debugger stacks support causality; Gateway SIGTERM cycles are supplemental. No stored-data contract changes.

Label justifications:

P1: Observed exit deadlocks can stall real CLI workflows and Gateway stop or restart operations on supported Node runtimes.
rating: 🐚 platinum hermit: Overall readiness is 🐚 platinum hermit; proof is 🦞 diamond lobster and patch quality is 🐚 platinum hermit.
status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR. Sufficient (live_output): The captured Testbox A/B exercises real gateway health --json children through the changed CLI bootstrap and exit boundary on Node 24.19, improving observed hangs from 6/1019 to 0/1019. Separate compiler arms and debugger stacks support causality; Gateway SIGTERM cycles are supplemental. No stored-data contract changes.
proof: sufficient: Contributor real behavior proof is sufficient. The captured Testbox A/B exercises real gateway health --json children through the changed CLI bootstrap and exit boundary on Node 24.19, improving observed hangs from 6/1019 to 0/1019. Separate compiler arms and debugger stacks support causality; Gateway SIGTERM cycles are supplemental. No stored-data contract changes.

Evidence

What I checked:

Pinned introduced change: Read the complete merge-base-to-head diff. One shared helper supplies the policy to the CLI entry, executable index, native hook relay, and macOS node worker; the remaining change documents that policy. (src/bootstrap/node-exit-safe-compilers.ts:15, 4d81d08ec167)
Runtime and settings boundary: The helper skips Bun and preserves explicit --maglev and --concurrent-sparkplug arguments, including underscore spellings. Entry, index, and macOS worker calls are guarded by main-module detection. No stored configuration, schema, authorization, CA policy, or exit-finalization contract is changed. (src/bootstrap/node-exit-safe-compilers.ts:16, 4d81d08ec167)
Current-main necessity: Current main uses these compiler flags in the Vitest process environment, but has no equivalent production bootstrap policy. This PR extends the mitigation to executable processes rather than duplicating an existing production owner. (scripts/lib/vitest-process-env.mts:168, 6347c0c51250)
Latest-release check: The new bootstrap helper is absent from the supplied latest-release commit. No merged same-repository fixing PR was established, so implemented-on-main closure is unsupported. (src/bootstrap/node-exit-safe-compilers.ts, c074824a27c9)
Real after-fix exit evidence: The captured PR body reports real gateway health --json child processes on a four-vCPU Blacksmith Testbox with Node 24.19 and fresh state at eight-way concurrency: defaults hung 6/1019 times, versus 0/1019 with both compilers disabled. Separate compiler arms hung 12/944 and 2/1019 times; debugger observations place hangs in worker shutdown waiting on background GC. This exercises the changed CLI bootstrap and process-exit boundary, rather than a mocked flag setter. Gateway SIGTERM cycles had no hangs in either arm and therefore provide supplemental coverage only. (4d81d08ec167)
Documented performance and scope: The captured body reports about 5% additional sequential CLI wall time and 10–20% under eight-way load, with unchanged log-redaction steady-state performance. It explicitly identifies separately spawned worker entrypoints as follow-up scope and proposes retiring the policy after supported Node versions include the upstream fix. (4d81d08ec167)

Likely related people:

steipete: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)
SunnyShu0925: Suggested for follow-up; no historical authorship or introduction is verified. (role: unverified routing candidate; confidence: low)

Rating scale

Score	Internal tier	Crab rank	Meaning
6/6	S	🦀 challenger crab	Exceptional readiness
5/6	A	🦞 diamond lobster	Very strong readiness
4/6	B	🐚 platinum hermit	Good normal PR; ordinary maintainer review
3/6	C	🦐 gold shrimp	Useful, but confidence is limited
2/6	D	🦪 silver shellfish	Proof or implementation needs work
1/6	F	🧂 unranked krab	Not merge-ready
N/A	NA	🌊 off-meta tidepool	Rating does not apply

Overall follows the weaker of proof and patch quality.
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

Workflow

ClawSweeper keeps one durable marker-backed review comment per issue or PR.
Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
Maintainers can also comment @clawsweeper review to request a fresh review only.
Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

steipete · 2026-10-02T21:34:18Z

Merged via squash.

Prepared head SHA: 4d81d08ec1677f518a0b1d5b3892337c120f8fca
Landed commit: 1aa0aeb54959b391ca3cb4e283345ee83e2f782b

…enclaw#163788) On Node 24 and 26, process.exit() joins V8's platform workers without disposing the isolate. A concurrent Maglev or Sparkplug compile job parked waiting for a main-thread GC then never finishes, so CLI commands, the hook relay, or a stopping Gateway can sit at 0% CPU forever after printing their output (nodejs/node#64274; fix pending in nodejs/node#66171). Every OpenClaw executable entrypoint (CLI entry, Gateway/service index, native hook relay, macOS node worker) now turns off Maglev and concurrent Sparkplug at startup when it runs as main, matching Node 22's tiering and the existing Vitest policy. Library imports keep stock V8; explicit --maglev or --concurrent-sparkplug flags still win. Testbox A/B (4 vCPU, Node 24.19, e2e CLI child replay, 8-way): 6/1019 hung with defaults, 0/1019 with the policy; re-enabling only Maglev hung 2/1019 and only concurrent Sparkplug 12/944, so both flags are needed.

…enclaw#163788) On Node 24 and 26, process.exit() joins V8's platform workers without disposing the isolate. A concurrent Maglev or Sparkplug compile job parked waiting for a main-thread GC then never finishes, so CLI commands, the hook relay, or a stopping Gateway can sit at 0% CPU forever after printing their output (nodejs/node#64274; fix pending in nodejs/node#66171). Every OpenClaw executable entrypoint (CLI entry, Gateway/service index, native hook relay, macOS node worker) now turns off Maglev and concurrent Sparkplug at startup when it runs as main, matching Node 22's tiering and the existing Vitest policy. Library imports keep stock V8; explicit --maglev or --concurrent-sparkplug flags still win. Testbox A/B (4 vCPU, Node 24.19, e2e CLI child replay, 8-way): 6/1019 hung with defaults, 0/1019 with the policy; re-enabling only Maglev hung 2/1019 and only concurrent Sparkplug 12/944, so both flags are needed. (cherry picked from commit 1aa0aeb) Co-authored-by: RomneyDa <[email protected]>

@steipete

Related: #163788 ## What Problem This Solves Fixes commands hanging after output during forced Node shutdown without retaining the blanket Maglev and concurrent-Sparkplug workaround. ## User Impact CLI commands and the Gateway finish owned work and cleanup before exiting naturally, with Node's default compiler settings restored. Exit status, Unix signal termination, output, explicit compiler overrides, and system-CA trust are preserved. The Gateway still joins native skills-watcher retirement. This can retain synchronous FSEvents shutdown latency on macOS; the separate upstream certificate-loader stall is not fixed. Native macOS/Windows qualification is not claimed. ## Why This Change Was Made Existing lifecycle owners retain admitted work, native workers, child transports, updater recovery, and watchdogs through actual completion. The boundary guard prevents new forced normal exits. Cleanup dependencies are retained before a command can replace its package, without eagerly importing them into CLI bootstrap. No new configuration, dependency, or schema is introduced. ## Evidence - Earlier full pnpm build passed on composed tree faaafbe6b1f47e707f8392243898525990f3a452 (the terminal repair plus pinned CI base 97e81ce), including the canonical CLI-bootstrap import gate, in 4m35.7s. Explicit/requested exits restore terminal state after cleanup; ordinary completion stays reset-free. The build retains a Control UI startup-JS size warning; no budget was raised or waived. - The built packaged CLI completed a real-PTY model round trip and exited naturally with status 0. Eight dual-TTY failure probes and twelve piped/dual-TTY telemetry probes passed. These used the built executable and existing helpers, not a claimed canonical Vitest smoke run. - The composed memory helper parses with one run/execute binding and passes 20 publication tests. The final runtime, one-shot, and cleanup suites passed 40 tests, including red proof of premature watcher loading. Broker regressions: 84 passed; real-process negative controls cover IPC disconnect and Unix signal propagation after output settlement. - The SIGTERM capture fixture retains its 20-second parent deadline and persistence assertions: red was 99 instead of 143; repaired case passed in 708ms (11.19s wrapper). The latest message/update fixture matrix passed 25/25 (179.52s): synthetic pending resources release only after observed stall diagnostics and join completion; ordered exit listeners remain checked. These later repairs change test/support files only, not production code. - Latest pinned-main integration preserved QA error reporting without forced exit: 204 QA tests, 10 reconciliation tests, and 6 real service-child relay tests passed. The reconciliation case failed before removing an obsolete admission close. Scoped types, lint, and three-pass review are complete; this integration did not repeat the full application build. - CI repair qualification: 96 tests across seven files passed (196.15s); final QA integration: 34 tests across three files passed (43.28s). Scoped types, type-aware lint, formatting, and mock-export checking passed. Fresh independent P0–P2 reviews found no actionable issues. - Browser startup recovery: all three saved/chat/toast cases passed (35.743s). Output and local-chat shutdown regressions were exercised with the rebuilt CLI. Fixture repairs preserve native completion, privacy, output, and fallback-model assertions. - Historical pre-integration proof: 115 affected test files; 137 built-executable cases with two platform skips; package-integrity checks; Linux Node 24.19.0 compiler activity and real Gateway SIGTERM-to-natural-exit in 175ms; published 2026.9.9 driver update with readiness and durable/incognito state checks. [Historical verification](https://github.com/openclaw/openclaw/actions/runs/38073218249). Compiler traces are functional witnesses, not throughput benchmarks or universal deadlock proof. Current-head CI run 38106144930 is red: 12 failing test jobs are attributed to the tested baseline (294b2e3), with complete job and source-bound evidence for native admission. They cover browser catalog/focus and a stale Details selector, channel mocks, Codex import/steering, and Claude discovery fixtures. None is claimed fixed or green by this PR. The original Gateway read-admission and node-wire timeout observations remain unreproduced, not declared fixed. Earlier local smoke preparation was refused by managed-artifact separation; no guard was bypassed. Requested and directed by @steipete. ## Work sessions - [Original discussion with @steipete](https://team.openclaw.ai/chat/roboclaw/dashboard/fcef23f8-a69b-4b60-9869-3bd40ac48588) - [Implementation and verification](https://team.openclaw.ai/chat/roboclaw/subagent/c0adc270-c23a-4139-b3b7-00852b7db66f) ## Worked on by - @steipete  --- [View the OpenClaw team session](https://team.openclaw.ai/chat/roboclaw/subagent/c0adc270-c23a-4139-b3b7-00852b7db66f) Co-authored-by: steipete <[email protected]>

@steipete

Related: openclaw#163788 ## What Problem This Solves Fixes commands hanging after output during forced Node shutdown without retaining the blanket Maglev and concurrent-Sparkplug workaround. ## User Impact CLI commands and the Gateway finish owned work and cleanup before exiting naturally, with Node's default compiler settings restored. Exit status, Unix signal termination, output, explicit compiler overrides, and system-CA trust are preserved. The Gateway still joins native skills-watcher retirement. This can retain synchronous FSEvents shutdown latency on macOS; the separate upstream certificate-loader stall is not fixed. Native macOS/Windows qualification is not claimed. ## Why This Change Was Made Existing lifecycle owners retain admitted work, native workers, child transports, updater recovery, and watchdogs through actual completion. The boundary guard prevents new forced normal exits. Cleanup dependencies are retained before a command can replace its package, without eagerly importing them into CLI bootstrap. No new configuration, dependency, or schema is introduced. ## Evidence - Earlier full pnpm build passed on composed tree faaafbe6b1f47e707f8392243898525990f3a452 (the terminal repair plus pinned CI base 5706528), including the canonical CLI-bootstrap import gate, in 4m35.7s. Explicit/requested exits restore terminal state after cleanup; ordinary completion stays reset-free. The build retains a Control UI startup-JS size warning; no budget was raised or waived. - The built packaged CLI completed a real-PTY model round trip and exited naturally with status 0. Eight dual-TTY failure probes and twelve piped/dual-TTY telemetry probes passed. These used the built executable and existing helpers, not a claimed canonical Vitest smoke run. - The composed memory helper parses with one run/execute binding and passes 20 publication tests. The final runtime, one-shot, and cleanup suites passed 40 tests, including red proof of premature watcher loading. Broker regressions: 84 passed; real-process negative controls cover IPC disconnect and Unix signal propagation after output settlement. - The SIGTERM capture fixture retains its 20-second parent deadline and persistence assertions: red was 99 instead of 143; repaired case passed in 708ms (11.19s wrapper). The latest message/update fixture matrix passed 25/25 (179.52s): synthetic pending resources release only after observed stall diagnostics and join completion; ordered exit listeners remain checked. These later repairs change test/support files only, not production code. - Latest pinned-main integration preserved QA error reporting without forced exit: 204 QA tests, 10 reconciliation tests, and 6 real service-child relay tests passed. The reconciliation case failed before removing an obsolete admission close. Scoped types, lint, and three-pass review are complete; this integration did not repeat the full application build. - CI repair qualification: 96 tests across seven files passed (196.15s); final QA integration: 34 tests across three files passed (43.28s). Scoped types, type-aware lint, formatting, and mock-export checking passed. Fresh independent P0–P2 reviews found no actionable issues. - Browser startup recovery: all three saved/chat/toast cases passed (35.743s). Output and local-chat shutdown regressions were exercised with the rebuilt CLI. Fixture repairs preserve native completion, privacy, output, and fallback-model assertions. - Historical pre-integration proof: 115 affected test files; 137 built-executable cases with two platform skips; package-integrity checks; Linux Node 24.19.0 compiler activity and real Gateway SIGTERM-to-natural-exit in 175ms; published 2026.9.9 driver update with readiness and durable/incognito state checks. [Historical verification](https://github.com/openclaw/openclaw/actions/runs/38073218249). Compiler traces are functional witnesses, not throughput benchmarks or universal deadlock proof. Current-head CI run 38106144930 is red: 12 failing test jobs are attributed to the tested baseline (960ccb8), with complete job and source-bound evidence for native admission. They cover browser catalog/focus and a stale Details selector, channel mocks, Codex import/steering, and Claude discovery fixtures. None is claimed fixed or green by this PR. The original Gateway read-admission and node-wire timeout observations remain unreproduced, not declared fixed. Earlier local smoke preparation was refused by managed-artifact separation; no guard was bypassed. Requested and directed by @steipete. ## Work sessions - [Original discussion with @steipete](https://team.openclaw.ai/chat/roboclaw/dashboard/fcef23f8-a69b-4b60-9869-3bd40ac48588) - [Implementation and verification](https://team.openclaw.ai/chat/roboclaw/subagent/c0adc270-c23a-4139-b3b7-00852b7db66f) ## Worked on by - @steipete  --- [View the OpenClaw team session](https://team.openclaw.ai/chat/roboclaw/subagent/c0adc270-c23a-4139-b3b7-00852b7db66f) Co-authored-by: steipete <[email protected]>

openclaw-barnacle Bot added docs Improvements or additions to documentation cli CLI command changes size: XS labels Oct 2, 2026

steipete marked this pull request as ready for review October 2, 2026 21:09

steipete merged commit 1aa0aeb into main Oct 2, 2026
192 of 193 checks passed

steipete deleted the steipete/node24-exit-deadlock-cli branch October 2, 2026 21:33

github-actions Bot mentioned this pull request Oct 2, 2026

📡 Upstream Digest — 2026-10-02 21:38 UTC curtismercier/openclaw-mods#1939

Open

meinharrd mentioned this pull request Oct 6, 2026

fix(ci): remove process.exit() from the macOS Ant download step, add a hang watchdog (#544) solardev-xyz/freedom-browser#574

Merged

clawsweeper Bot mentioned this pull request Oct 8, 2026

[Bug]: intermittent Linux CLI hang after malformed --container diagnostic #141083

Open

roboclaw-bot mentioned this pull request Oct 10, 2026

fix: finish shutdown without disabling V8 compilers #168587

Merged

Navigation Menu

Repository navigation

fix: CLI and Gateway processes can hang forever at exit on Node 24 and 26 - #163788

fix: CLI and Gateway processes can hang forever at exit on Node 24 and 26#163788
steipete merged 1 commit into
mainfrom
steipete/node24-exit-deadlock-cli

steipete commented Oct 2, 2026

Uh oh!

clawsweeper Bot commented Oct 2, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented Oct 2, 2026

Security

PR surface

Review metrics

Technical review

Labels

Evidence

Rating scale

Workflow

Uh oh!

Uh oh!

steipete commented Oct 2, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

steipete commented Oct 2, 2026

What Problem This Solves

User Impact

Why This Change Was Made

Evidence

Uh oh!

clawsweeper Bot commented Oct 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

ClawSweeper review complete

Uh oh!

clawsweeper Bot commented Oct 2, 2026

ClawSweeper review

What this changes

Merge readiness

Review scores

Verification

How this fits together

Before merge

Security

PR surface

Review metrics

Technical review

Labels

Evidence

Rating scale

Workflow

Uh oh!

Uh oh!

steipete commented Oct 2, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

clawsweeper Bot commented Oct 2, 2026 •

edited

Loading