Amazon Aurora PostgreSQL
Aurora is not RDS, and your monitoring should know
CloudWatch measures instances. Your clients talk to endpoints. DBExplore probes the endpoints, tracks reader rotation, and treats replica lag in the units Aurora actually reports.
What goes wrong
Healthy instances, broken writer
During failover the writer CNAME can still resolve to an instance that no longer accepts writes. Every instance metric looks fine while clients time out.
Lag skew across the reader endpoint
A synchronous standby reports lag but is not in the reader rotation. Naive skew alerts compare nodes nobody reads from.
Serverless capacity surprises
ACU scaling lags a traffic spike, and the max capacity ceiling is the outage nobody set an alert on.
What DBExplore does about it
Endpoint-level probes
Reader-endpoint lag skew and writer reachability during failover are measured from the endpoints your clients use, not from instance metrics.
Replica lag in consistent units
Cluster-level lag reported the way Aurora actually means it, next to replay lag and sync state from the topology.
Volume and capacity headroom
Storage against the cluster ceiling, serverless capacity against its maximum, and volume I/O on one runway chart.
Wait events across writer and readers
Performance data correlated with active session history for every node in the cluster.
Amazon Aurora PostgreSQL: questions we get first
Aurora Serverless v2?
Yes. ACU utilization and max-capacity headroom are first-class metrics with forecasting.
Global Database?
Cross-region replication lag and failover state are tracked per secondary cluster.
How does the agent connect?
A container, Kubernetes deployment or sidecar, or native package next to your databases, using a read-only monitoring role over a small connection pool with short statement timeouts. No superuser.
Can DBExplore change anything without approval?
Not by default. Every tenant starts at approve. Disruptive actions always need two approvers, and every mutating action passes a fail-closed policy gate first.
Run DBExplore on your Amazon Aurora PostgreSQL.
We onboard teams in small batches. Tell us about your fleet and we will reach out when a seat opens. One email, no drip campaign.