johno/gastown
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

fix(patrol): Make Deacon health checks conditional (gt-cp9bt)

Browse Source 
Previously, the Deacon patrol formula sent HEALTH_CHECK nudges to
Witnesses and Refineries on every patrol cycle (~1-2 minutes). This
disturbed idle agents constantly and wasted their context.

Now health check nudges are only sent when concerning signals are
detected:
- Agent not running or no heartbeat
- Status check hangs or fails
- Queue stuck or merge failures

The status commands (gt witness status, gt refinery status) provide
sufficient liveness info for healthy agents. Reserve nudges for
actual problems.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
toast
2026-01-01 18:14:22 -08:00
committed by Steve Yegge
parent 03eeb38129
commit e267b9baa8
1 changed files with 19 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
.beads/formulas/mol-deacon-patrol.formula.toml
View File

@@ -257,26 +257,33 @@ For each rig, run:
```bash
gt witness status <rig>
gt refinery status <rig>
```
# Health ping (clears backoff as side effect)
**IMPORTANT: Do NOT send health pings on every patrol cycle.**
Health check nudges disturb idle agents and waste their context. Only send
a HEALTH_CHECK nudge when you have **concerning signals** (see table below).
**When to send health pings** (only if concerned):
```bash
# Only if agent appears unresponsive or status check fails:
gt nudge <rig>/witness 'HEALTH_CHECK from deacon'
gt nudge <rig>/refinery 'HEALTH_CHECK from deacon'
```
**Health Ping Benefit**: The nudge commands serve dual purposes:
1. **Liveness verification** - Agent responds to prove it's alive
2. **Backoff reset** - Any nudge resets agent's backoff to base interval
This ensures patrol agents remain responsive even during quiet periods when the
feed has no mutations. Deacon patrols every ~1-2 minutes, so maximum backoff
is bounded by the ping interval.
**Signals to assess:**
| Component | Healthy Signals | Concerning Signals |
|-----------|-----------------|-------------------|
| Witness | State: running, recent activity | State: not running, no heartbeat |
| Refinery | State: running, queue processing | Queue stuck, merge failures |
| Witness | State: running, recent activity | State: not running, no heartbeat, or status check hangs |
| Refinery | State: running, queue processing | Queue stuck, merge failures, or not responding |
**When NOT to ping:**
- Agent reports state=running with recent activity
- No pending work in queues
- Status check returns promptly with healthy output
The status commands (`gt witness status`, `gt refinery status`) provide
sufficient liveness information. Reserve nudges for actual problems.
**Tracking unresponsive cycles:**
@@ -296,7 +303,7 @@ health_state:
| Cycles Unresponsive | Suggested Action |
|---------------------|------------------|
| 1-2 | Note it, check again next cycle |
| 1-2 | Note it, send HEALTH_CHECK nudge |
| 3-4 | Attempt restart: gt witness restart <rig> |
| 5+ | Escalate to Mayor with context |