Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix v2026.4.5: gateway process 100% CPU after upgrade from v2026.4.2 [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#61701•Fetched 2026-04-08 02:55:44
View on GitHub
Comments
3
Participants
4
Timeline
7
Reactions
2
Author
ENAwarenessENAwareness
Participants
beezdevopsbeezdevopscoracoocoracooENAwarenessENAwarenessnickcj931nickcj931
Timeline (top)
commented ×3cross-referenced ×2subscribed ×2

After upgrading from v2026.4.2 to v2026.4.5, the gateway process immediately consumes 100%+ CPU and never settles. Rolling back to v2026.4.2 resolves the issue — CPU drops to ~1% after startup.

Root Cause

After upgrading from v2026.4.2 to v2026.4.5, the gateway process immediately consumes 100%+ CPU and never settles. Rolling back to v2026.4.2 resolves the issue — CPU drops to ~1% after startup.

Fix Action

Workaround

Rolled back to v2026.4.2.

Code Example

channels.telegram: channels.telegram.streamMode is legacy; use channels.telegram.streaming instead. Run "openclaw doctor --fix".
tools.web.search: tools.web.search provider-owned config moved to plugins.entries.<plugin>.config.webSearch. Run "openclaw doctor --fix".

---

PID USER   PR NI  VIRT     RES    SHR S %CPU %MEM  COMMAND
xxx leo    20  0  10.0g  766104  59716 R 136  4.7   openclaw-gateway
RAW_BUFFERClick to expand / collapse

Description

After upgrading from v2026.4.2 to v2026.4.5, the gateway process immediately consumes 100%+ CPU and never settles. Rolling back to v2026.4.2 resolves the issue — CPU drops to ~1% after startup.

Environment

  • Host: Ubuntu 24.04, x86_64, HP Pavilion laptop (NAS)
  • Docker: containerized deployment (self-hosted, openclaw:local image built from source)
  • Config: 7 Telegram bot accounts, sandbox.mode: "off", tools.exec: {host: "gateway", security: "full", ask: "off"}
  • Previous version: v2026.4.2 (working fine)

Steps to Reproduce

  1. Running v2026.4.2 normally (CPU ~1% idle)
  2. git checkout v2026.4.5 && docker build -t openclaw:local . && docker compose up -d
  3. Gateway starts, CPU immediately hits 100%+ and stays there indefinitely
  4. docker logs shows repeated legacy config warnings in a loop:
channels.telegram: channels.telegram.streamMode is legacy; use channels.telegram.streaming instead. Run "openclaw doctor --fix".
tools.web.search: tools.web.search provider-owned config moved to plugins.entries.<plugin>.config.webSearch. Run "openclaw doctor --fix".
  1. Manually fixing the config keys (removing streamMode, migrating tools.web.search to plugins.entries.brave.config.webSearch) did NOT resolve the CPU issue
  2. openclaw doctor --fix was killed by OOM (exit 137) — unable to run under high CPU load
  3. Rolling back to v2026.4.2 and restarting: CPU returns to normal (~1%)

Top output during v2026.4.5

PID USER   PR NI  VIRT     RES    SHR S %CPU %MEM  COMMAND
xxx leo    20  0  10.0g  766104  59716 R 136  4.7   openclaw-gateway

Expected Behavior

Gateway should start normally with legacy config keys (warning is fine, but should not cause infinite loop / 100% CPU).

Workaround

Rolled back to v2026.4.2.

Additional Context

  • The config has streamMode: "block" (legacy) and tools.web.search (legacy) from pre-v2026.4.5
  • Even after manually migrating these keys, the CPU spike persisted, suggesting the issue may not be solely config-related
  • The v2026.4.2 gateway handles the same config file without issues

extent analysis

TL;DR

The most likely fix for the high CPU usage issue in v2026.4.5 is to identify and address the root cause of the repeated legacy config warnings, potentially by fixing the configuration or adjusting the code to handle legacy configurations without entering an infinite loop.

Guidance

  • Investigate the openclaw doctor --fix command to understand why it's unable to run under high CPU load and consider running it with increased resources or in a controlled environment to fix the legacy config issues.
  • Review the changes between v2026.4.2 and v2026.4.5 to identify any new features or bug fixes that could be causing the infinite loop or high CPU usage, focusing on how legacy configurations are handled.
  • Consider temporarily disabling the channels.telegram and tools.web.search features to isolate if the issue is related to these specific components or if it's a more general problem with the gateway process.
  • Monitor the docker logs for any other repeating patterns or errors that could indicate the source of the high CPU usage, beyond just the legacy config warnings.

Notes

The issue seems to be related to how the gateway process in v2026.4.5 handles legacy configurations, but the exact cause is not clear from the provided information. The fact that manually fixing the config keys did not resolve the issue suggests there might be other factors at play.

Recommendation

Apply a workaround by rolling back to v2026.4.2 until the root cause of the high CPU usage in v2026.4.5 can be identified and addressed, as this version is known to work with the current configuration without issues.

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…
#runtime error#dependency conflict#environment setup#docker error#permission error

Still need to ship something?

×6

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

Back to top recommendations

TRENDING