v2026.2.6-3 | WSL2 | Isolated cron jobs skipping scheduled slots
✅Solved
Hourly cron job (0 * * * *, isolated, next-heartbeat) skips slots — nextRunAtMs jumps forward without executing. Same setup, same config, morning-briefing (0 9 * * *) fires perfectly at 9:00 AM, but hourly-pulse due at 10:00 AM gets skipped and jumps to 11:00 AM. Overnight, zero jobs fire despite gateway staying alive. Jobs seem to only run when there's user message activity. Heartbeat is 15m, maxConcurrentRuns default (1). Is there a known issue with frequently-scheduled cron jobs being skipped, or should I use wakeMode: "now" instead of next-heartbeat?
Solution
3) Add a timeout to the long job(s) so they can’t block the lane forever (isolated jobs support
4) If you’re on WSL2 and expect overnight reliability: ensure the host doesn’t sleep (or move the Gateway to an always-on Linux/VPS).
-
-
- a log slice around a “skipped hour” (
If it’s reproducible with
payload.timeoutSeconds4) If you’re on WSL2 and expect overnight reliability: ensure the host doesn’t sleep (or move the Gateway to an always-on Linux/VPS).
Debug info that will make this actionable
Can you paste:-
openclaw --version-
openclaw cron runs --id <jobId> --limit 50- a log slice around a “skipped hour” (
openclaw logs --limit 300 | rg "cron:"If it’s reproducible with
maxConcurrentRuns > 1
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
Recent Announcements
Similar Threads
Was this page helpful?
