Monitoring

Logging

[logging]
level = "info"                  # trace, debug, info, warn, error
format = "json"                 # "json" or "pretty"
output = "stdout"               # "stdout", "stderr", or "none"
per-request-spans = false       # Per-request tracing spans (~13% CPU overhead)
# file-path = "/var/log/frogdb/frogdb.log"

# [logging.rotation]
# max-size-mb = 100
# frequency = "daily"           # "daily", "hourly", or "never"
# max-files = 5

Runtime change: CONFIG SET loglevel debug

Prometheus Metrics

Endpoint: GET http://<host>:9090/metrics

[http]
enabled = true
bind = "127.0.0.1"
port = 9090

See Metrics Reference for the complete list of all exported metrics.

Key Metrics to Monitor

Metric	Alert When
frogdb_memory_used_bytes / frogdb_memory_maxmemory_bytes	> 80% for 5 min
histogram_quantile(0.99, frogdb_commands_duration_seconds)	> 100ms for 5 min
frogdb_persistence_errors_total	Any increase
frogdb_connections_rejected_total	Any increase
max(frogdb_shard_keys) / avg(frogdb_shard_keys)	> 2.0 (hot shard)

Distributed Tracing

FrogDB supports OpenTelemetry (OTLP):

[tracing]
enabled = true
otlp-endpoint = "http://localhost:4317"
sampling-rate = 0.1
service-name = "frogdb"

Health Checks

Endpoint	Response
GET /health/live	200 OK if process is running
GET /health/ready	200 OK if accepting commands
PING (Redis protocol)	PONG if healthy

Kubernetes probes:

livenessProbe:
  httpGet:
    path: /health/live
    port: 9090
  initialDelaySeconds: 5
readinessProbe:
  httpGet:
    path: /health/ready
    port: 9090
  initialDelaySeconds: 5

FrogDB-Specific Introspection

STATUS

Comprehensive server and shard status:

STATUS JSON
GET /status/json    # HTTP endpoint

DEBUG HOTSHARDS

Per-shard traffic analysis to detect imbalanced key distribution:

DEBUG HOTSHARDS [PERIOD <seconds>]

INFO Sections

Standard Redis INFO sections plus FrogDB extensions:

INFO server|clients|memory|persistence|stats|replication|cpu|keyspace|hotshards|frogdb