managing-keydb

name	managing-keydb
description	Use when working with Keydb — keyDB multi-threaded Redis-compatible datastore management, active replication monitoring, FLASH storage tier analysis, and sub-key expiration inspection. Covers thread configuration, multi-master replication, MVCC diagnostics, and memory-to-flash ratio tuning. Read this skill before any KeyDB operations.
connection_type	keydb
preload	false

KeyDB Management Skill

Monitor, analyze, and optimize KeyDB instances safely.

MANDATORY: Discovery-First Pattern

Always run INFO ALL before any operations. Never assume threading model or active replication topology.

Phase 1: Discovery

#!/bin/bash

keydb_cmd() {
    keydb-cli -h "${KEYDB_HOST:-localhost}" -p "${KEYDB_PORT:-6379}" \
              ${KEYDB_PASSWORD:+-a "$KEYDB_PASSWORD"} \
              --no-auth-warning "$@" 2>/dev/null || \
    redis-cli -h "${KEYDB_HOST:-localhost}" -p "${KEYDB_PORT:-6379}" \
              ${KEYDB_PASSWORD:+-a "$KEYDB_PASSWORD"} \
              --no-auth-warning "$@"
}

echo "=== Server Info ==="
keydb_cmd INFO server | grep -E 'keydb_version|redis_version|uptime_in_days|server_threads|os'

echo ""
echo "=== Threading ==="
keydb_cmd CONFIG GET server-threads 2>/dev/null
keydb_cmd CONFIG GET server-thread-affinity 2>/dev/null

echo ""
echo "=== Memory Overview ==="
keydb_cmd INFO memory | grep -E 'used_memory_human|used_memory_peak_human|maxmemory_human|mem_fragmentation_ratio|maxmemory_policy'

echo ""
echo "=== Keyspace ==="
keydb_cmd INFO keyspace

echo ""
echo "=== Active Replication ==="
keydb_cmd INFO replication | grep -E 'role|connected_slaves|master_link_status|master_host|active_replica'

echo ""
echo "=== FLASH Tier ==="
keydb_cmd CONFIG GET storage-provider 2>/dev/null || echo "FLASH storage not configured"

Phase 1 outputs: KeyDB version, thread count, memory usage, keyspace, replication topology, FLASH status.

Phase 2: Analysis

#!/bin/bash

keydb_cmd() {
    keydb-cli -h "${KEYDB_HOST:-localhost}" -p "${KEYDB_PORT:-6379}" \
              ${KEYDB_PASSWORD:+-a "$KEYDB_PASSWORD"} \
              --no-auth-warning "$@" 2>/dev/null || \
    redis-cli -h "${KEYDB_HOST:-localhost}" -p "${KEYDB_PORT:-6379}" \
              ${KEYDB_PASSWORD:+-a "$KEYDB_PASSWORD"} \
              --no-auth-warning "$@"
}

echo "=== Performance Stats ==="
keydb_cmd INFO stats | grep -E 'instantaneous_ops_per_sec|total_commands_processed|keyspace_hits|keyspace_misses|evicted_keys'

echo ""
echo "=== Multi-Master Status ==="
keydb_cmd INFO replication | grep -E 'active-replica|multi_master'

echo ""
echo "=== Slow Log ==="
keydb_cmd SLOWLOG GET 10

echo ""
echo "=== Client Connections ==="
keydb_cmd CLIENT LIST | awk -F'[ =]' '{for(i=1;i<=NF;i++) if($i=="addr") print $(i+1)}' | \
    cut -d: -f1 | sort | uniq -c | sort -rn | head -10

echo ""
echo "=== Sub-Key Expires ==="
keydb_cmd INFO stats | grep -E 'subkey|expired' | head -5

 
 
keydb_cmd INFO persistence | grep -E 

Output Format

KEYDB ANALYSIS
==============
Version: [version] | Threads: [count] | Mode: [standalone/active-replica]
Memory: [used]/[max] | Keys: [count] | FLASH: [enabled/disabled]

ISSUES FOUND:
- [issue with affected component]

RECOMMENDATIONS:
- [actionable recommendation]

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative