AWS AmazonElastiCache: Note automatic failover requirement for distributed KV caching clusters
Summary
Adds a note that cache.i8ge.48xlarge distributed KV caching clusters have no replicas or Multi-AZ, so automatic failover must still be enabled for cluster mode, and failed nodes are replaced with loss of in-memory objects.
Security assessment
Documents availability/durability behavior (node replacement and data loss on a single-node configuration) rather than any security control; no vulnerability, credential, or encryption content.
Evidence
+Distributed KV caching clusters (`cache.i8ge.48xlarge`) don't support replicas or Multi-AZ. You must still enable automatic failover, because cluster mode requires it. With 0 replicas, there is no replica to promote. When a node fails, ElastiCache replaces it and the objects on the node are lost.
Diff
diff --git a/AmazonElastiCache/latest/dg/AutoFailover.md b/AmazonElastiCache/latest/dg/AutoFailover.md index d1b55a1..d0e3b1b 100644 --- a//AmazonElastiCache/latest/dg/AutoFailover.md +++ b//AmazonElastiCache/latest/dg/AutoFailover.md @@ -33,0 +34,4 @@ Note +Note + +Distributed KV caching clusters (`cache.i8ge.48xlarge`) don't support replicas or Multi-AZ. You must still enable automatic failover, because cluster mode requires it. With 0 replicas, there is no replica to promote. When a node fails, ElastiCache replaces it and the objects on the node are lost. For more information, see [Requirements and limits for distributed KV caching](./distributed-kv-caching-limits.html) and [Resilience and data persistence](./distributed-kv-caching-how-it-works.html#distributed-kv-caching-how-it-works.resilience). +