Monitoring Redis with INFO, MONITOR, SLOWLOG, and latency commands

Observing Redis in production

Redis exposes rich internal metrics. Learn to read them before you need them in an incident.

INFO — snapshot of everything

redis-cli INFO memory    # used_memory, mem_fragmentation_ratio
redis-cli INFO stats     # total_commands_processed, keyspace_hits/misses
redis-cli INFO replication  # role, connected_slaves, repl_offset

# Cache hit ratio
# keyspace_hits / (keyspace_hits + keyspace_misses)

SLOWLOG — find slow commands

# redis.conf: log commands slower than 10ms
slowlog-log-slower-than 10000

SLOWLOG GET 10    # last 10 slow commands
# Shows: id, timestamp, duration_micros, command, args

MONITOR — live command stream

redis-cli MONITOR
# Streams every command in real time — WARNING: halves throughput
# Use only for debugging, never in production permanently

Latency tracking

CONFIG SET latency-monitor-threshold 10  # ms
LATENCY LATEST
LATENCY HISTORY event_name

Memory analysis

redis-cli --bigkeys          # find largest keys by type
redis-cli --memkeys          # memory usage per key
MEMORY USAGE mykey           # bytes for one key
redis-cli --stat             # live rolling stats dashboard