0% 5h quota used · resets in 2h live · 08/09/2026 12:04 CEST

The daemon watching itself

Heartbeats, alert state, locks and boots, as the health tick last recorded them.

Daemon live
pid 1 · up 23m
Last tick 35s ago
08/09/2026 12:04
Alerts held 0
per-key failure state
Spent today $53.34
$53.34ceiling $540

Pollers

a loop silent for three times its interval counts as stalled
LoopEveryLast runSilent forLast error
approvals300s08/09/2026 12:004mnoneticking
triggers300s08/09/2026 12:004mnoneticking
queue30s08/09/2026 12:043.1snoneticking
chat60s08/09/2026 12:031mnoneticking
cli-version86400s08/09/2026 11:4023mnoneticking
balances3600s08/09/2026 12:004mnoneticking

No alert state on file — nothing has failed often enough to be tracked.

Locks

one per task; a lock older than the task budget plus fifteen minutes is stale and gets stolen

No locks held — nothing is running right now.

Boots

four in ten minutes is a crash loop
  • 08/09/2026 11:40 24m ago

What this page cannot see: the container or the host being gone. Every check above runs inside the process it is checking, so a dead daemon shows up here only as a health file that stopped moving — and if this console is reading a stale copy of the directory, not even that. Closing that gap needs a watcher outside the box.