AWS ElastiCache Deep Skill

Deep analysis of AWS ElastiCache Redis and Memcached clusters with parallel execution and anti-hallucination guardrails.

Relationship to other AWS skills:

• aws-elasticache-deep/ → ElastiCache-specific deep analysis (replication, failover, engine metrics)
• aws/ → "How to execute" (parallel patterns, throttling, output format)

CRITICAL: Parallel Execution Requirement

ALL independent operations MUST run in parallel using background jobs (&) and wait.

#!/bin/bash
export AWS_PAGER=""

for cluster in $clusters; do
  get_cluster_metrics "$cluster" &
done
wait

Helper Functions

#!/bin/bash
export AWS_PAGER=""

# List replication groups (Redis)
list_replication_groups() {
  aws elasticache describe-replication-groups \
    --output text \
    --query 'ReplicationGroups[].[ReplicationGroupId,Status,ClusterEnabled,AutomaticFailover,MultiAZ,NodeGroups[0].NodeGroupMembers[0].CacheNodeId]'
}

# List cache clusters
list_cache_clusters() {
  aws elasticache describe-cache-clusters --show-cache-node-info \
    --output text \
    --query 'CacheClusters[].[CacheClusterId,Engine,EngineVersion,CacheNodeType,NumCacheNodes,CacheClusterStatus,ReplicationGroupId]'
}

# Get cluster metrics
get_cluster_metrics() {
  local cluster_id=$1 days=${2:-1}
  local end_time start_time
  end_time=$(date -u +"%Y-%m-%dT%H:%M:%S")
  start_time=$(date -u -d "$days days ago" +"%Y-%m-%dT%H:%M:%S" 2>/dev/null || date -u -v-${days}d +"%Y-%m-%dT%H:%M:%S")

  aws cloudwatch get-metric-statistics \
    --namespace AWS/ElastiCache --metric-name CPUUtilization \
    --dimensions Name=CacheClusterId,Value="$cluster_id" \
    --start-time "$start_time" --end-time "$end_time" \
    --period $((days * 86400)) --statistics Average Maximum \
    --output text --query "Datapoints[0].[\"$cluster_id\",\"CPU\",Average,Maximum]" &

  aws cloudwatch get-metric-statistics \
    --namespace AWS/ElastiCache --metric-name DatabaseMemoryUsagePercentage \
    --dimensions Name=CacheClusterId,Value="$cluster_id" \
    --start-time "$start_time" --end-time "$end_time" \
    --period $((days * 86400)) --statistics Average Maximum \
    --output text --query "Datapoints[0].[\"$cluster_id\",\"Memory\",Average,Maximum]" &

  aws cloudwatch get-metric-statistics \
    --namespace AWS/ElastiCache --metric-name CacheHitRate \
    --dimensions Name=CacheClusterId,Value="$cluster_id" \
    --start-time "$start_time" --end-time "$end_time" \
    --period $((days * 86400)) --statistics Average \
    --output text --query "Datapoints[0].[\"$cluster_id\",\"HitRate\",Average]" &

  aws cloudwatch get-metric-statistics \
    --namespace AWS/ElastiCache --metric-name Evictions \
    --dimensions Name=CacheClusterId,Value="$cluster_id" \
    --start-time "$start_time" --end-time "$end_time" \
    --period $((days * 86400)) --statistics Sum \
    --output text --query "Datapoints[0].[\"$cluster_id\",\"Evictions\",Sum]" &
  wait
}

# Get replication lag (Redis)
get_replication_lag() {
  local cluster_id=$1
  local end_time start_time
  end_time=$(date -u +"%Y-%m-%dT%H:%M:%S")
  start_time=$(date -u -d "1 hour ago" +"%Y-%m-%dT%H:%M:%S" 2>/dev/null || date -u -v-1H +"%Y-%m-%dT%H:%M:%S")
  aws cloudwatch get-metric-statistics \
    --namespace AWS/ElastiCache --metric-name ReplicationLag \
    --dimensions Name=CacheClusterId,Value="$cluster_id" \
    --start-time "$start_time" --end-time "$end_time" \
    --period 300 --statistics Average Maximum \
    --output text --query 'Datapoints[*].[Timestamp,Average,Maximum]' | sort -k1 | tail -5
}

Common Operations

1. Cluster Inventory with Engine Details

#!/bin/bash
export AWS_PAGER=""
aws elasticache describe-cache-clusters --show-cache-node-info \
  --output text \
  --query 'CacheClusters[].[CacheClusterId,Engine,EngineVersion,CacheNodeType,NumCacheNodes,CacheClusterStatus,PreferredMaintenanceWindow]' | sort -k2

2. Replication Health (Redis)

#!/bin/bash
export AWS_PAGER=""
aws elasticache describe-replication-groups \
  --output text \
  --query 'ReplicationGroups[].[ReplicationGroupId,Status,ClusterEnabled,AutomaticFailover,MultiAZ,MemberClusters[]]'

# Check replication lag for all replica nodes
REPLICAS=$(aws elasticache describe-cache-clusters --output text \
  --query 'CacheClusters[?Engine==`redis`].CacheClusterId')
END=$(date -u +"%Y-%m-%dT%H:%M:%S")
START=$(date -u -d "1 hour ago" +"%Y-%m-%dT%H:%M:%S" 2>/dev/null || date -u -v-1H +"%Y-%m-%dT%H:%M:%S")
for replica in $REPLICAS; do
  aws cloudwatch get-metric-statistics \
    --namespace AWS/ElastiCache --metric-name ReplicationLag \
    --dimensions Name=CacheClusterId,Value="$replica" \
    --start-time "$START" --end-time "$END" \
    --period 300 --statistics Average Maximum \
    --output text --query "Datapoints[-1].[\"$replica\",Average,Maximum]" &
done
wait

3. Memory and Eviction Analysis

#!/bin/bash
export AWS_PAGER=""
END=$(date -u +"%Y-%m-%dT%H:%M:%S")
START=$(date -u -d "1 day ago" +"%Y-%m-%dT%H:%M:%S" 2>/dev/null || date -u -v-1d +"%Y-%m-%dT%H:%M:%S")
CLUSTERS=$(aws elasticache describe-cache-clusters --output text --query 'CacheClusters[].CacheClusterId')
for cluster in $CLUSTERS; do
  {
    mem=$(aws cloudwatch get-metric-statistics \
      --namespace AWS/ElastiCache --metric-name DatabaseMemoryUsagePercentage \
      --dimensions Name=CacheClusterId,Value="$cluster" \
      --start-time "$START" --end-time "$END" \
      --period 86400 --statistics Average Maximum \
      --output text --query 'Datapoints[0].[Average,Maximum]')
    evictions=$(aws cloudwatch get-metric-statistics \
      --namespace AWS/ElastiCache --metric-name Evictions \
      --dimensions Name=CacheClusterId,Value="$cluster" \
      --start-time "$START" --end-time "$END" \
      --period 86400 --statistics Sum \
      --output text --query 'Datapoints[0].Sum')
    printf "%s\tMem:%s\tEvictions:%s\n" "$cluster" "$mem" "${evictions:-0}"
  } &
done
wait

4. Cache Hit Rate Analysis

#!/bin/bash
export AWS_PAGER=""
END=$(date -u +"%Y-%m-%dT%H:%M:%S")
START=$(date -u -d "7 days ago" +"%Y-%m-%dT%H:%M:%S" 2>/dev/null || date -u -v-7d +"%Y-%m-%dT%H:%M:%S")
CLUSTERS=$(aws elasticache describe-cache-clusters --output text --query 'CacheClusters[].CacheClusterId')
for cluster in $CLUSTERS; do
  {
    hits=$(aws cloudwatch get-metric-statistics \
      --namespace AWS/ElastiCache --metric-name CacheHits \
      --dimensions Name=CacheClusterId,Value="$cluster" \
      --start-time "$START" --end-time "$END" \
      --period 604800 --statistics Sum \
      --output text --query 'Datapoints[0].Sum')
    misses=$(aws cloudwatch get-metric-statistics \
      --namespace AWS/ElastiCache --metric-name CacheMisses \
      --dimensions Name=CacheClusterId,Value="$cluster" \
      --start-time "$START" --end-time "$END" \
      --period 604800 --statistics Sum \
      --output text --query 'Datapoints[0].Sum')
    printf "%s\tHits:%s\tMisses:%s\n" "$cluster" "${hits:-0}" "${misses:-0}"
  } &
done
wait

5. Connection Count Monitoring

#!/bin/bash
export AWS_PAGER=""
END=$(date -u +"%Y-%m-%dT%H:%M:%S")
START=$(date -u -d "1 day ago" +"%Y-%m-%dT%H:%M:%S" 2>/dev/null || date -u -v-1d +"%Y-%m-%dT%H:%M:%S")
CLUSTERS=$(aws elasticache describe-cache-clusters --output text --query 'CacheClusters[].CacheClusterId')
for cluster in $CLUSTERS; do
  aws cloudwatch get-metric-statistics \
    --namespace AWS/ElastiCache --metric-name CurrConnections \
    --dimensions Name=CacheClusterId,Value="$cluster" \
    --start-time "$START" --end-time "$END" \
    --period 3600 --statistics Average Maximum \
    --output text --query "Datapoints[-1].[\"$cluster\",Average,Maximum]" &
done
wait

Anti-Hallucination Rules

1. Redis vs Memcached metrics - ReplicationLag only applies to Redis. CacheHitRate calculation differs between engines. Always check the engine type first.
2. EngineCPUUtilization vs CPUUtilization - For Redis, use EngineCPUUtilization for the Redis process CPU. CPUUtilization includes OS overhead. For Memcached with multiple cores, CPUUtilization may underreport per-core usage.
3. DatabaseMemoryUsagePercentage - This is Redis-only. For Memcached, use BytesUsedForCacheItems divided by maxmemory.
4. Cluster mode enabled vs disabled - Redis cluster mode (sharded) uses NodeGroups with multiple shards. Non-cluster mode has one shard with primary/replicas.
5. Failover requires Multi-AZ - Automatic failover only works with Multi-AZ enabled on the replication group. Check AutomaticFailover and MultiAZ fields.

Output Format

Present results as a structured report:

Aws Elasticache Deep Report
═══════════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]

Target ≤50 lines of output. Use tables for multi-resource comparisons.

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

Common Pitfalls

• Node type naming: ElastiCache uses cache. prefix (e.g., cache.r6g.large), not the EC2 naming convention.
• Reserved node pricing: ElastiCache RIs are separate from EC2 RIs. Check with describe-reserved-cache-nodes.
• Maintenance windows: During maintenance, nodes may be temporarily unavailable. Check PreferredMaintenanceWindow.
• CloudWatch statistics syntax: Use spaces not commas: --statistics Average Maximum.
• Snapshot retention: Automatic backups (Redis only) retain snapshots for a configurable period. Check SnapshotRetentionLimit.