# Daily Cloud Security Brief: Enhanced AWS Resilience, AI Collaboration, and Analytics with Terraform Insights

- Cloud: AWS
- Date: 2026-09-28
- Canonical page: https://jonathanglass.cloud/whats-new/aws/2026-09-28.html
- Dataset: https://jonathanglass.cloud/data/aws/daily_summaries.json

Greetings Cloud Security Architects, DevSecOps Leads, and Terraform Specialists,

Today's AWS updates bring significant enhancements across database resilience, AI-driven contact center operations, and big data analytics, with varying levels of immediate Infrastructure as Code (IaC) support via Terraform. Let's dive into the details and critical security considerations.

### Database Highlights

*   **Amazon RDS for SQL Server now supports Multi-AZ deployments for Developer Edition.** This update allows organizations to build and test high availability (HA) configurations using SQL Server Developer Edition (2019, 2022, and 2025 Enterprise Developer Edition) at no additional license cost. By maintaining a synchronous standby replica in a different Availability Zone and providing automatic failover, this feature is crucial for validating disaster recovery strategies and ensuring application resilience in non-production environments. From a security and architectural standpoint, this enables early integration of HA principles into development workflows, reducing the risk of misconfigurations in production and aligning with compliance requirements for system availability. *Terraform Support*: Supported in [Terraform AWS Provider: aws_db_instance](https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/db_instance) (arguments `multi_az = true` and `engine = "sqlserver-developer"`) in **v5.68.0+**.

### Contact Center Innovation

*   **Amazon Connect Customer now supports agent-to-agent (A2A) collaboration with specialized AI agents.** This innovative feature allows human agents to bring in AI agents during live customer interactions, enabling specialized AI assistance for complex requests. Connect Customer manages context passing, unified observability, guardrails, and escalation controls across all participating AI agents, supporting both text and bidirectional voice. The A2A protocol has been expanded for voice streaming and interaction-level session continuity. Security insights include the critical need for robust access controls for AI agents, stringent data privacy measures (especially for PII shared during interactions), and continuous monitoring of AI agent behavior via unified observability for audit and compliance. Architectural considerations involve securely integrating and managing the lifecycle of specialized AI agents within the contact center ecosystem. *Terraform Support*: Pending provider release for this specific feature (use AWS CLI/SDK or custom provisioner in the interim). The core `aws_connect_instance` resource is supported in [Terraform AWS Provider](https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/connect_instance).

### Analytics & Big Data Advancements

*   **Amazon EMR 7.14 is now available with new features across EMR on EC2, EMR on EKS, and EMR Serverless.** This release upgrades Apache Spark to 3.5.8 and Apache Iceberg to 1.10.1, introducing faster materialized view refreshes for Iceberg tables using change data capture. For Amazon EMR on EKS, clusters now support Spark Connect endpoints with token-based authentication for interactive Spark sessions and can run workloads on IPv6 Amazon EKS clusters. Amazon EMR Serverless sees an increased storage limit for Spark jobs from 200 GiB to 1 TiB. From a security perspective, the introduction of Spark Connect endpoints with token-based authentication for EMR on EKS is a significant enhancement, promoting more secure interactive sessions and aligning with least privilege principles. IPv6 support on EKS improves network scalability and security posture. The increased storage limit for EMR Serverless requires continued vigilance on data encryption at rest and in transit, as well as robust access controls for larger datasets. *Terraform Support*: The ability to specify EMR 7.14 is available via the `release_label` argument in `aws_emr_cluster` (for EMR on EC2/EKS) and `aws_emrserverless_application` (for EMR Serverless). New features like Spark Connect endpoints with token-based authentication on EMR on EKS would typically be configured via `configurations_json` within `aws_emr_cluster` or `aws_emr_container_virtual_cluster` resources, or potentially new dedicated arguments in future provider releases. IPv6 support for EKS is configured at the `aws_eks_cluster` level. Core EMR resources are supported in [Terraform AWS Provider: aws_emr_cluster](https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/emr_cluster) and [Terraform AWS Provider: aws_emrserverless_application](https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/emrserverless_application) in **v2.0.0+**.

## Announcements in this summary

- [Amazon RDS for SQL Server](https://aws.amazon.com/about-aws/whats-new/2026/09/amazon-rds-sql-server-multi-az-developer-edition/) (database)
- [Amazon Connect](https://aws.amazon.com/about-aws/whats-new/2026/09/Amazon-Connect-Customer-A2A) (Contact Center)
- [Amazon EMR](https://aws.amazon.com/about-aws/whats-new/2026/09/amazon-emr-7-14-available/) (Analytics)
