openclaw - 💡(How to fix) Fix [Bug]: model-resolution and auth stages take 11-14s and 8-10s respectively on every agent dispatch (even with 0 plugins, 1 model) [3 comments, 4 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#77649Fetched 2026-05-06 06:23:24
View on GitHub
Comments
3
Participants
4
Timeline
6
Reactions
2
Author
Timeline (top)
commented ×3labeled ×2closed ×1

Node: 24.13.1 Windows 10 x64 OpenClaw: 2026.5.3-1

Even with 0 plugins, 1 model (deepseek-v4-flash), no feishu, no moonshot:

  • model-resolution: 11852-14091ms every agent run
  • auth: 7779-10882ms every agent run
  • Total dispatch: 20-28s
  • openclaw health: 10.5s (should be <100ms)
  • openclaw status: 24s (should be <1s)
  • Single thread consuming 97% CPU (153s out of 157s)
  • eventLoopDelayMaxMs up to 45s
  • RSS memory 700MB+ with 4 models defined (270MB with 1 model)

DeepSeek API responds in 203ms (network not the issue). DNS is 5ms (fast).

Root Cause

Node: 24.13.1 Windows 10 x64 OpenClaw: 2026.5.3-1

Even with 0 plugins, 1 model (deepseek-v4-flash), no feishu, no moonshot:

  • model-resolution: 11852-14091ms every agent run
  • auth: 7779-10882ms every agent run
  • Total dispatch: 20-28s
  • openclaw health: 10.5s (should be <100ms)
  • openclaw status: 24s (should be <1s)
  • Single thread consuming 97% CPU (153s out of 157s)
  • eventLoopDelayMaxMs up to 45s
  • RSS memory 700MB+ with 4 models defined (270MB with 1 model)

DeepSeek API responds in 203ms (network not the issue). DNS is 5ms (fast).

Fix Action

Fix / Workaround

Even with 0 plugins, 1 model (deepseek-v4-flash), no feishu, no moonshot:

  • model-resolution: 11852-14091ms every agent run
  • auth: 7779-10882ms every agent run
  • Total dispatch: 20-28s
  • openclaw health: 10.5s (should be <100ms)
  • openclaw status: 24s (should be <1s)
  • Single thread consuming 97% CPU (153s out of 157s)
  • eventLoopDelayMaxMs up to 45s
  • RSS memory 700MB+ with 4 models defined (270MB with 1 model)
RAW_BUFFERClick to expand / collapse

Bug type

Regression (worked before, now fails)

Beta release blocker

No

Summary

Node: 24.13.1 Windows 10 x64 OpenClaw: 2026.5.3-1

Even with 0 plugins, 1 model (deepseek-v4-flash), no feishu, no moonshot:

  • model-resolution: 11852-14091ms every agent run
  • auth: 7779-10882ms every agent run
  • Total dispatch: 20-28s
  • openclaw health: 10.5s (should be <100ms)
  • openclaw status: 24s (should be <1s)
  • Single thread consuming 97% CPU (153s out of 157s)
  • eventLoopDelayMaxMs up to 45s
  • RSS memory 700MB+ with 4 models defined (270MB with 1 model)

DeepSeek API responds in 203ms (network not the issue). DNS is 5ms (fast).

Steps to reproduce

openclaw gateway restart

Expected behavior

openclaw health spend 10s (normal < 100ms)

Actual behavior

openclaw health spend 10s (normal < 100ms)

OpenClaw version

OpenClaw 2026.5.3-1 (2eae30e)

Operating system

Window 10

Install method

No response

Model

deepseek v4

Provider / routing chain

Non

Additional provider/model setup details

No response

Logs, screenshots, and evidence

Impact and severity

No response

Additional information

No response

extent analysis

TL;DR

The issue can be mitigated by investigating and optimizing the OpenClaw gateway's performance, focusing on the high CPU usage and event loop delays.

Guidance

  • Investigate the single thread consuming 97% CPU and optimize its performance to reduce the overall dispatch time.
  • Check the event loop delays and optimize the code to reduce the eventLoopDelayMaxMs to an acceptable range.
  • Verify the OpenClaw gateway's configuration and ensure it is properly set up to handle the defined models, as the RSS memory increases significantly with more models.
  • Consider monitoring the system resources and OpenClaw gateway's performance after restarting to identify any patterns or bottlenecks.

Example

No code snippet is provided as the issue lacks specific code details.

Notes

The issue seems to be related to the OpenClaw gateway's performance, but the exact cause is unclear. Further investigation is needed to determine the root cause of the high CPU usage and event loop delays.

Recommendation

Apply workaround: Optimize the OpenClaw gateway's performance and configuration to reduce CPU usage and event loop delays, as the issue is likely related to the gateway's performance rather than a specific version or upgrade.

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

openclaw health spend 10s (normal < 100ms)

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]: model-resolution and auth stages take 11-14s and 8-10s respectively on every agent dispatch (even with 0 plugins, 1 model) [3 comments, 4 participants]