Cron jobs severely delayed or failing with 60000ms `cron.run` WebSocket timeout (v2026.2.6)
Environment:
* OpenClaw v2026.2.6
* Hosted via Coolify (VPS)
* Channel: Telegram
The Problem:
When I ask the bot to set a scheduled reminder (cron job) for our main chat, it almost never executes on time. The jobs suffer from one of two symptoms:
1. Massive Delays: A reminder scheduled for 1:30 PM might randomly appear in the chat hours later (e.g., at 7:16 AM UTC).
2. Complete Failure (Timeout): The job fails entirely, and the bot eventually sends a system message saying: "I encountered an error when trying to run the cron job. It seems like the gateway timed out."
The Logs:
We dug into the gateway logs during one of the complete failures, and we caught the exact moment it dies.
When the scheduled time arrives, the internal loopback request for
Our Suspected Cause:
It appears the gateway is suffering from lock contention or a heartbeat coalescing bug when dealing with
Our Questions:
1. Is the 60,000ms WebSocket timeout for local tool execution hardcoded, or can we increase it in
2. Is this massive execution delay a known issue with
3. Until this is patched, what is the most reliable way to schedule a simple text reminder so it arrives on time? Will forcing the bot to use
Thanks for the help!
* OpenClaw v2026.2.6
* Hosted via Coolify (VPS)
* Channel: Telegram
The Problem:
When I ask the bot to set a scheduled reminder (cron job) for our main chat, it almost never executes on time. The jobs suffer from one of two symptoms:
1. Massive Delays: A reminder scheduled for 1:30 PM might randomly appear in the chat hours later (e.g., at 7:16 AM UTC).
2. Complete Failure (Timeout): The job fails entirely, and the bot eventually sends a system message saying: "I encountered an error when trying to run the cron job. It seems like the gateway timed out."
The Logs:
We dug into the gateway logs during one of the complete failures, and we caught the exact moment it dies.
When the scheduled time arrives, the internal loopback request for
cron.runOur Suspected Cause:
It appears the gateway is suffering from lock contention or a heartbeat coalescing bug when dealing with
sessionTarget: "main"Our Questions:
1. Is the 60,000ms WebSocket timeout for local tool execution hardcoded, or can we increase it in
openclaw.json2. Is this massive execution delay a known issue with
sessionTarget: "main"3. Until this is patched, what is the most reliable way to schedule a simple text reminder so it arrives on time? Will forcing the bot to use
sessionTarget: "isolated"Thanks for the help!
The AI that actually does things. Emails, calendar, home automation - all from your favorite chat app. New shell, ready to help. The lobster way.
134,827Members
Resources
Was this page helpful?
