All blog posts
Verified the Remote Query Cache Plugin in AWS Advanced JDBC Wrapper 3.3.0 with Aurora PostgreSQL (1M rows) and node-based ElastiCache for Valkey. A 749ms window function query dropped to 2ms. TTL-based consistency control worked as expected.
Hands-on verification of AgentCore Runtime's new managed session storage. Tested file restoration after stop/resume, pip/git workspace persistence, session isolation, and graceful shutdown behavior with real measurement data.
Hands-on verification of EKS Pod Identity session policies. Restrict IAM role permissions per association without creating additional roles. Covers session tag exclusivity, policy size limits, and trade-offs to evaluate before adoption.
Verified centralized Network Firewall Proxy architecture across VPCs. SourceVpc enables per-VPC policy control via PrivateLink endpoints, while SourceIp is NATed and unusable in cross-VPC setups.
Enabled TLS intercept on Network Firewall Proxy using ACM Private CA. Verified PreRequest phase HTTP method blocking, PostResponse Content-Type filtering, and documented PCA configuration gotchas and Preview-specific limitations.
Hands-on verification of AWS Network Firewall Proxy (Public Preview). Tested NAT Gateway integration, PreDNS domain filtering, source-based access control, and DNS spoofing resistance.
Upgraded AWS JDBC Wrapper from 2.6.4 to 3.2.0 and re-tested Blue/Green Switchover on both PostgreSQL and MySQL. MySQL's connection failure turned out to be timing-dependent (0–1 failures). Summarizes 3.x migration guidance.
Tested Blue/Green Switchover on Aurora MySQL and compared with PostgreSQL. IN_PROGRESS phase was only 3s (vs 12s for PG), but the BG plugin may have 0–1 connection errors (timing-dependent). HikariCP's stale writer problem is engine-agnostic. Plain JDBC stayed on old host for 31s after recovery.
Compared downtime during Blue/Green Switchover across Plain JDBC, HikariCP with retry, and AWS JDBC Driver BG plugin. The plugin recorded 0 failures in 400 queries during testing. HikariCP kept connecting to the old writer — a critical pitfall.
Measured real downtime during Aurora PostgreSQL failover vs Blue/Green Switchover. Failover had 12 connection failures in two separate windows, Blue/Green Switchover ~26s with 6 failures. DNS TTL of 60s is the key bottleneck.