openclaw - 💡(How to fix) Fix bug(gateway): repeated SIGTERM-driven restarts make the gateway look crash-looped [1 comments, 2 participants]

From the outside this feels like "gateway keeps dying", but the underlying cause is likely config/UI/reload churn rather than an internal fatal exception. That makes debugging much harder and can also break channel login flows while a user is trying to relink WhatsApp.

TL;DR

The gateway restart loop is likely caused by external SIGTERM signals triggered by config reloads or Control UI actions, and auditing these paths may help identify the root cause.

Guidance

• Review the code paths related to Control UI channel actions, config apply/reload, and channel health/relink actions to identify potential sources of SIGTERM signals.
• Investigate the watchdog/doctor style flows to determine if they are contributing to the restart loop.
• Check the logs for patterns or correlations between the SIGTERM signals and specific user actions or system events.
• Consider adding logging or debugging statements to track the origin of the SIGTERM signals and the restart requests.

Example

No code snippet is provided as the issue does not contain specific code references.

Notes

The issue appears to be related to external factors rather than an internal crash, making it harder to debug. The repeated config reloads and channel reconnect churn may be indicative of a larger issue with the system's configuration or communication flows.

Recommendation

Apply workaround: Audit and refine the config reload and Control UI action paths to prevent unnecessary SIGTERM signals and restarts, as the root cause is likely related to these areas.