Monitoring

Monitor server health, formation status, request latency, and resource usage. MUXI exposes health endpoints and Prometheus metrics for integration with your monitoring stack.

Health Endpoints

Server Health

curl http://localhost:7890/health

{
  "status": "healthy",
  "formations": 3,
  "uptime": 86400
}

Formation Health

curl http://localhost:8001/health

{
  "status": "healthy",
  "agents": ["assistant", "researcher"],
  "uptime": 3600
}

Logs

Server Logs

# If running in foreground
# Logs output to stdout

# If running as service
journalctl -u muxi-server -f

Formation Logs

# Via CLI
muxi logs my-assistant --follow

# Direct file access
tail -f ~/.muxi/server/logs/my-assistant.log

Log Levels

# Server config
logging:
  level: debug    # debug, info, warn, error

Metrics

Server Status

muxi-server status

MUXI Server Status
==================
Status:     running
Port:       7890
Uptime:     24h 15m
Formations: 3 running, 0 stopped
Memory:     256MB

Formation Status

muxi server list

ID              STATUS    PORT   UPTIME    MEMORY
my-assistant    running   8001   2h 15m    128MB
support-bot     running   8002   1d 3h     256MB
research-bot    stopped   -      -         -

Log Format

JSON Format (Production)

logging:
  format: json

{"level":"info","time":"2025-01-08T10:00:00Z","message":"Formation started","formation":"my-assistant","port":8001}

Text Format (Development)

logging:
  format: text

2025-01-08 10:00:00 INFO Formation started formation=my-assistant port=8001

Observability Events

MUXI emits 349 typed events for observability.

Event Categories

Streaming Events

curl http://localhost:8001/v1/events \
  -H "X-MUXI-Admin-Key: ..." \
  -H "Accept: text/event-stream"

Event Integration

Forward events to observability platforms:

• Datadog
• Splunk
• Elastic
• Custom webhook

Alerting

Health Check Alerts

Monitor health endpoint:

# Simple check
curl -f http://localhost:7890/health || alert

# With timeout
curl -f --max-time 5 http://localhost:7890/health || alert

Log-Based Alerts

Watch for error patterns:

tail -f /var/log/muxi/server.log | grep -i error

Dashboard Integration

Prometheus Metrics

Export metrics for Prometheus (if enabled):

curl http://localhost:7890/metrics

Grafana

Import MUXI dashboard for visualization.

Troubleshooting

Formation Not Responding

1. Check health endpoint
2. View logs for errors
3. Check resource usage
4. Restart if needed

muxi server restart my-assistant

High Memory Usage

Check formation memory:

muxi server get my-assistant

Restart to clear memory:

muxi server restart my-assistant

Slow Responses

Check logs for slow operations:

muxi logs my-assistant | grep -i slow

Next Steps

• Production - Production deployment
• Deep Dives: Observability - Event details