1. Release Notes
  2. Feature Updates
  3. Features Released in Software Version 3.0.2-b852928 (May 2023)

Features Released in Software Version 3.0.2-b852928 (May 2023)

Load Balancer as a Service (Layer 7)

The Load Balancing service provides automated traffic distribution from one public or private entry point where incoming requests are received, to a set of backend servers in the virtual cloud network (VCN) where the requests are processed.

To efficiently manage compute resources associated with a load balancer, you can attach a compute instance pool to a load balancer backend set. Doing this adds each instance in the pool as a backend server in the backend set.

A load balancer manages TCP or HTTP traffic based on typical distribution policies like round-robin, least connections, or IP hash.

For optimal utilization of the backend resources, request routing on the listener side can be further refined by using multiple virtual host names and path route rules. Load balancers also provide configuration options for SSL traffic handling and session persistence.

For information about load balancing, see the Load Balancing Overview chapter of the Oracle Private Cloud Appliance Concepts Guide. Instructions to configure load balancers can be found in the chapter Load Balancer as a Service of the Oracle Private Cloud Appliance User Guide

Instance Pool Autoscaling

An instance pool defines a set of compute instances that is managed as a group. Autoscaling a pool enables you to use resources more effectively by stopping or removing instances when demand is lower and starting or adding instances when demand is higher.

For more information, see "Using Schedule-Based Autoscaling" in the section titled "Working with Instance Pools" in the Oracle Private Cloud Appliance User Guide.

You can use instance pool autoscaling along with load balancing by attaching an instance pool that has an autoscaling configuration to a load balancer backend set. See "Managing Instance Pool Load Balancer Attachments" in the section titled "Working with Instance Pools" in the Oracle Private Cloud Appliance User Guide.

High Availability Configuration and Fault Domain Enforcement Default

The Service Enclave has new commands that give administrators more control over how to implement instance high availability.

  • Instance high availability: When enabled, instances are automatically reboot migrated off of an unreachable compute node. The default is enabled.

  • Instance fault domain resolution: When enabled, instances that are running in a fault domain that is not the fault domain that is specified in their instance configuration (their selected fault domain) are automatically migrated back to their selected fault domain when resources become available in that fault domain. The default is enabled.

    Instances can become displaced (running in a fault domain that is not their selected fault domain) during compute node evacuation or failure. You can list all currently displaced instances.

  • Instance restart: When enabled, instances that were stopped by the Compute service (not by an administrator) are automatically restarted in their selected fault domain when resources become available in that fault domain. The default is enabled.

    Instances can be stopped by the Compute service during compute node evacuation or compute node failure when no fault domain has resources to accommodate the instances, or when strict fault domain enforcement is enabled and no other compute node in the selected fault domain can accommodate the instances. You can list all instances that are currently stopped by the Compute service.

  • Strict fault domain enforcement: When enabled, instances that cannot be accommodated in the current fault domain during compute node evacuation or failure will be stopped by the Compute service. If the force option is not used for compute node evacuation, instances will be still running in the current compute node and the compute node evacuation will fail.

    The default is disabled: Instances will be migrated to a different fault domain if possible.

For more information, see "Migrating Instances from a Compute Node" and "Configuring the Compute Service for High Availability" in the Hardware Administration chapter of the Oracle Private Cloud Appliance Administrator Guide.

Routing Options in Virtual Networking

The virtual networking configuration in the Compute Enclave provides additional routing options:

  • setting a private IP address as a route rule target

  • associating a route table with a dynamic routing gateway (DRG) attachment.

With this enhancement, existing data center infrastructure and applications can be integrated into the network communication between Private Cloud Appliance compute instances, or between instances and other network services external to the appliance.