Redis — alert on a high mem-fragmentation ratio
Redis is up but `mem_fragmentation_ratio > 1.5` — real RAM use is much bigger than what `used_memory_human` reports. OOM will hit suddenly.
Recipe
#!/usr/bin/env bash
# /etc/cron.d/redis-frag
# */5 * * * * root /opt/redis-frag.sh
HOST=${REDIS_HOST:-127.0.0.1}
PORT=${REDIS_PORT:-6379}
THRESH=${THRESH:-1.5} # alert above 1.5
INFO=$(redis-cli -h "$HOST" -p "$PORT" INFO memory)
RATIO=$(echo "$INFO" | awk -F: '/mem_fragmentation_ratio/{gsub(/\r/,"",$2); print $2}')
USED=$(echo "$INFO" | awk -F: '/used_memory_human/{gsub(/\r/,"",$2); print $2}')
# Bash floats: use bc
if (( $(echo "$RATIO > $THRESH" | bc -l) )); then
echo "redis-frag: ratio=$RATIO used=$USED (threshold $THRESH)"
curl -fsS "$HEARTBEAT_URL" --data "ratio=$RATIO,used=$USED"
exit 2
fi
echo "OK (ratio=$RATIO used=$USED)"Same thing in Enterno.io
Wrap in an Enterno heartbeat — 30-day fragmentation trend lets you plan MEMORY PURGE or a restart a week before OOM.
Related recipes
Redis slave is behind master — read-after-write returns stale data. No native alert, you need an external one.
long_query_time = 1, slow_query_log enabled. You need to know when the slow-query rate per minute suddenly jumps (a deploy broke an index, ORM went N+1).
Memcached fills up and starts evicting keys under load; the app cache-misses and hammers the DB. Want an evictions/min threshold.