What's new in this release

New features and enhancements are available in IBM® Cloud Pak for Network Automation 2.5.0.

Lifecycle management operations

Retrying and rolling back intent processes
You can now retry or roll back the most recent lifecycle management intent process that failed or was canceled. When you retry an intent process, it tries again to change the assembly and its components to their intended states. When you roll back an intent process, it attempts to change the assembly and its components to the states that they were in before the intent process.
You can retry and roll back lifecycle management intent processes by using the network automation UI or by using the Intents API.
The intents metric counts the number of intents to reach the retry accepted, rollback accepted, and the rollback successful stages. Also, the intents_failed metric counts the number of retry and rollback intent requests that are not accepted.
The following screen capture shows a failed intent process in the network automation UI and the menu that you can use to retry or roll back the process:
Execution page that shows a failed intent process and the open New Intent menu
For more information, see Intents overview and Retrying or rolling back intent processes.
Specifying sequence of upgrade transitions on resources
You can now define the sequence in which Reconfigure or Upgrade lifecycle transitions are completed on different resource components in an assembly. For example, if a database component needs a configuration update before a similar change is made to a related application, you can control that sequence by defining upgrade dependencies for the application.
When you define the upgrade dependencies for a resource, you specify the resources whose transitions must occur first. That is, you list the other resources in the assembly whose planned Reconfigure or Upgrade lifecycle transitions are to be run before the transition is run on this resource.
You can define upgrade dependencies in the Designer tool or in assembly descriptors. For more information about upgrade dependencies, see composition section in Assembly descriptors.

Customizing relationship operations on multiple component instances

The functionality and flexibility of relationship operations are enhanced in the following ways:
  • An assembly-level operation might be called by a relationship in another assembly that contains or references the assembly. If that operation is published from an assembly component that has multiple instances, the operation is now, by default, called for every component instance.
  • You can customize this default behavior by setting the multiple-instances attribute for the operation in the assembly descriptor. You can specify that operations are called for every component instance, any one instance, or the first instance.

For more information, see Customizing operations on multiple component instances.

Installing on a self-managed OpenShift cluster on AWS

You can now deploy a self-managed Red Hat® OpenShift® Container Platform cluster on Amazon Web Services (AWS). You can then install IBM Cloud Pak for Network Automation on the cluster.

In this way, you can save the cost of setting up and managing on-premises hardware if you already use AWS.

For more information, see Installing IBM Cloud Pak for Network Automation on a self-managed cluster on AWS.

Removed features

The following services are removed in this release of IBM Cloud Pak for Network Automation:
  • IRP Manager
  • Logical data unit (LDU) counting
For more information about removed features, see Deprecated and removed features.

IBM Cloud Pak for Network Automation social media

Read blogs and discussions from network automation experts to get the latest information about how to use IBM Cloud Pak for Network Automation to automate the operational lifecycle of your cloud network services. Join the Cloud Pak for Network Automation group.