Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug]: API timeout on 2026.4.26 / event loop saturation on 2026.4.29 — VPS deployment [2 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
openclaw/openclaw#76210•Fetched 2026-05-03 04:40:43
View on GitHub
Comments
2
Participants
2
Timeline
5
Reactions
2
Author
lerahc-uilerahc-ui
Participants
clawsweeper[bot]clawsweeper[bot]lerahc-uilerahc-ui
Timeline (top)
commented ×2labeled ×2closed ×1

Environment:

Vultr VPS, 2 vCPU, 3.8GB RAM, Ubuntu 22.04 Docker deployment via docker-compose Model: anthropic/claude-sonnet-4-6

Issue 1 — Version 2026.4.26: Every API call results in model fallback decision: candidate_failed reason=timeout detail=This operation was aborted. Stuck session diagnostic shows state=processing age=240s+ on every message regardless of context size. Bootstrap context is under 8,000 characters. Issue 2 — Version 2026.4.29: acpx and runway plugins download and install 44 packages at startup (143 seconds first run). After startup, eventLoopUtilization=1 and CPU pegged at 160%+ continuously at idle with no messages being processed. eventLoopDelayMaxMs=197031 observed. Result: Neither version is able to process messages reliably on this hardware.

Root Cause

Environment:

Vultr VPS, 2 vCPU, 3.8GB RAM, Ubuntu 22.04 Docker deployment via docker-compose Model: anthropic/claude-sonnet-4-6

Issue 1 — Version 2026.4.26: Every API call results in model fallback decision: candidate_failed reason=timeout detail=This operation was aborted. Stuck session diagnostic shows state=processing age=240s+ on every message regardless of context size. Bootstrap context is under 8,000 characters. Issue 2 — Version 2026.4.29: acpx and runway plugins download and install 44 packages at startup (143 seconds first run). After startup, eventLoopUtilization=1 and CPU pegged at 160%+ continuously at idle with no messages being processed. eventLoopDelayMaxMs=197031 observed. Result: Neither version is able to process messages reliably on this hardware.

RAW_BUFFERClick to expand / collapse

Bug type

Regression (worked before, now fails)

Beta release blocker

No

Summary

Environment:

Vultr VPS, 2 vCPU, 3.8GB RAM, Ubuntu 22.04 Docker deployment via docker-compose Model: anthropic/claude-sonnet-4-6

Issue 1 — Version 2026.4.26: Every API call results in model fallback decision: candidate_failed reason=timeout detail=This operation was aborted. Stuck session diagnostic shows state=processing age=240s+ on every message regardless of context size. Bootstrap context is under 8,000 characters. Issue 2 — Version 2026.4.29: acpx and runway plugins download and install 44 packages at startup (143 seconds first run). After startup, eventLoopUtilization=1 and CPU pegged at 160%+ continuously at idle with no messages being processed. eventLoopDelayMaxMs=197031 observed. Result: Neither version is able to process messages reliably on this hardware.

Steps to reproduce

Steps to reproduce:

Deploy OpenClaw via Docker on Vultr VPS (2 vCPU, 3.8GB RAM, Ubuntu 22.04) Configure with Anthropic API key and WhatsApp channel Start container and wait for gateway ready status Send any message via browser webchat or WhatsApp

Actual behavior:

On 2026.4.26: Session enters stuck state after ~240 seconds with no response. Log shows model fallback decision: candidate_failed reason=timeout On 2026.4.29: CPU pegged at 160%+ at idle before any message is sent. eventLoopUtilization=1 continuously. acpx and runway plugins install 44 packages at startup blocking the event loop for 143 seconds

Additional context: Direct Anthropic API call from the same VPS responds in 0.27 seconds, confirming the network and API key are not the issue. Problem is internal to OpenClaw's message processing pipeline.Sonnet 4.6

Expected behavior

Expected behavior: Agent receives message and returns a response within a reasonable time

Actual behavior

Actual behavior:

On 2026.4.26: Session enters stuck state after ~240 seconds with no response. Log shows model fallback decision: candidate_failed reason=timeout On 2026.4.29: CPU pegged at 160%+ at idle before any message is sent. eventLoopUtilization=1 continuously. acpx and runway plugins install 44 packages at startup blocking the event loop for 143 seconds

OpenClaw version

2026.4.26 and 2026.4.29 (latest as of May 2, 2026)

Operating system

Ubuntu 22.04.5 LTS

Install method

Docker via docker-compose. Image pulled from ghcr.io/openclaw/openclaw:latest and tagged as openclaw:local

Model

anthropic/claude-sonnet-4-6

Provider / routing chain

Direct Anthropic API — no proxy, no OpenRouter, no LiteLLM. API key configured in .env and auth-profiles.json

Additional provider/model setup details

Direct API call from the same VPS to Anthropic responds in 0.27 seconds confirming connectivity is not the issue. Timeout occurs within OpenClaw's internal processing pipeline before or during the API request construction.

Logs, screenshots, and evidence

Impact and severity

No response

Additional information

No response

extent analysis

TL;DR

The most likely fix involves optimizing OpenClaw's internal message processing pipeline to prevent timeouts and excessive CPU usage, potentially by addressing the event loop utilization issue and the installation of 44 packages at startup.

Guidance

  • Investigate the cause of the high eventLoopUtilization and eventLoopDelayMaxMs values, which may indicate a performance bottleneck in the message processing pipeline.
  • Consider optimizing or disabling the installation of 44 packages at startup, which blocks the event loop for 143 seconds and may contribute to the CPU pegging issue.
  • Review the configuration of the acpx and runway plugins to determine if there are any settings that can be adjusted to reduce their resource usage.
  • Test the system with a smaller message payload or a different model to see if the issue is specific to the anthropic/claude-sonnet-4-6 model or the current message size.

Example

No specific code snippet can be provided without further information on the internal workings of OpenClaw, but optimizing the startup process and event loop utilization could involve reviewing and refining the Docker deployment scripts or the OpenClaw application code.

Notes

The issue seems to be internal to OpenClaw's message processing pipeline, and resolving it may require modifications to the application code or its configuration. The fact that direct API calls to Anthropic respond quickly suggests that the network and API key are not the issue.

Recommendation

Apply a workaround by optimizing the startup process and event loop utilization, as the root cause of the issue appears to be related to the internal processing pipeline and not a version-specific bug that can be fixed by upgrading.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

Expected behavior: Agent receives message and returns a response within a reasonable time

#api#container setup#orchestration issue#cache issue#memory leak

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

openclaw - 💡(How to fix) Fix [Bug]: API timeout on 2026.4.26 / event loop saturation on 2026.4.29 — VPS deployment [2 comments, 2 participants]