ONAP ‘Jakarta’ Release Notes

This page provides the release notes for the ONAP ‘Jakarta’ release. This includes details of software versions, known limitations, and outstanding trouble reports.

Release notes are cumulative for the release, meaning this release note will have an entry for each Major, Minor, and Maintenance release, if applicable.

Each component within the ONAP solution maintains their own component level release notes and links to those release notes are provided below. Details on the specific items delivered in each release by each component is maintained in the component specific release notes.

Jakarta Major Release 10.0.0

Project

Open Network Automation Platform (ONAP)

Release name

Jakarta

Release version

10.0.0

Release date

2022, June 30th

Features

ONAP Jakarta focusses on:

  • Security enhancements in the A&AI, CCSDK, MSB, and MultiCloud projects, reducing log4j vulnerability and removing most GPLv3 dependencies

  • Deepened O-RAN integration in the OOF SON and CCSDK projects with O-RAN O1 models and the O-RAN AI Policy interface (consumed downstream by the O-RAN Software community)

  • Enabling a richer set of day-2 configuration for Cloud-Native Network Functions (CNF) through CDS API extensions

  • Intent based networking (IBN) for closed loop for E2E Network Slicing

  • New functionality in the Configuration Persistence Service (CPS) that allows more granular control of configuration-heavy network services like RAN

  • Simplification of control loop automation architecture, enabling easy deployment of new control modules

  • New Network Function lifecycle management features based on real-life use cases

  • Modeling: Solidified the data model for CNFs using the novel Application Service Descriptor (ASD) approach, while continuing alignment with data models produced by SDOs such as ETSI

  • An overhaul of the policy framework allowing easy composition of control loop policies and better observability

  • Continued 5G Super Blueprint integrations, including EMCO, Magma 1.6, Anuket and KubeRef RI2

Functional Requirements

Increased Cloud Native Functionality

  • Improved synchronization of k8s resources after the creation of the CNF. When some change occurs for the CNF in the k8s cluster, k8splugin sends a notification to the cnf-adapter which performs an update of the changes into A&AI.

  • CDS integration with k8splugin

    • The creation of the profile allows the specification of labels and additional k8sresource types to be returned by the status API

    • Better Configuration API support including rollback, improved deletion of the configuration with or without removal of the configuration resources in the cluster

    • Creation of the configuration template without a dedicated helm chart which allows for easy update of the override values by configuration API without a need to duplicate configuration template helm chart from the main Helm chart. The configuration template (Helm chart) is taken from the main definition.

  • New model for CNF modeling: Application Service Descriptor (ASD) model

E2E Network Slicing

  • CPS Integration with SDN-R for RAN Slice allocate and reconfigure scenarios

  • E2E network Slicing with CPS is completed for allocation and re-use scenarios

  • E2E Closed loop with CPS is functional

  • IBN based closed loop with ML MS (POC) and Config DB is functional

  • Optimization of cm-handle registration with CPS-DMI Plugin to upload yang model

  • CPS Integration Stabilization for RAN Slice activate/deactivate scenarios

  • Addition of call to OOF for allocateNSSI to enable TN NSSI reuse in TN NSSMF

  • Addition of call to OOF for terminateNxi API to deallocate NSSI (without terminating TN NSSI even when NSI is terminated) in TN NSSMF

  • Closed-loop enhancement in CCVPN to support Transport Slicing’s closed-loop

Control Loop evolutions

  • Control Loop in TOSCA LCM Improvement: abstract Automation Composition Management (ACM) logic with a generic Automation Composition definition, isolating Composition logic from ONAP component logic. It elaborates APIs that allow integrate with other design systems as well as 3PP component integration.

  • The current PMSH and TCS control loops are migrated to use an Automation Composition approach. Support for Automation Compositions in SDC is also introduced.

  • A Metadata set allows a global set of metadata containing rules or global parameters that all instances of a certain policy type can use. Metadata sets are introduced in the Policy Framework in the Jakarta release. This means that different rule set implementations can be associated with a policy type, which can be used in appropriate situations.

  • Introduction of Prometheus for monitoring Policy components so that necessary alerts can be easily triggered and possible outages can be avoided in production systems.

    • Expose application level metrics in policy components. An end user can plug in a Prometheus instance and start listening to the metrics exposed by Policy components and either raise alerts or show them on a Grafana dashboard for operations team to keep monitoring the health of the system.

    • Improve the policy/api and policy/pap readiness probes to handle database failures so that the policy/api and policy/pap kubernetes pods are marked ready only if the Policy database pod is ready.

    • Provide sample Grafana dashboards for policy metrics

  • Migration of Policy Framework components to Springboot to support easier handling, configuration and maintenance.

  • Policy Framework Database Configurability. The Policy Framework can be configured to use any JDBC-compliant RDBMS and configuration files are supplied for the Postgres RDBMS. MariaDB remains the default RDBMS for the Policy Framework in ONAP

  • System Attribute Improvements

    • Transaction boundaries on REST calls are implemented per REST call

    • JDBC backend uses Spring and Hibernate rather than Eclipselink

    • All GUIs are now included in the policy/gui microservice

    • Documentation is rationalized and cleaned up, testing documentation is now complete

    • Scripts are added to make release of the Policy Framework easier

Fault management

  • DCAE Helm Transformation finalized

  • Topic alignment for DCAE microservices: use standard topics for PM-Mapper, Slice-Analysis and KPI-MS

Extended O-RAN Integration

  • The O-RAN A1 interface (from the CCSDK project) provides a flexible way for RAN operators to manage wide area RAN network optimization

  • Enhanced A1 interface controller and A1 Policy capabilities are now usable by any service provider deploying and using ONAP. This functionality is used downstream in the O-RAN-SC Non-RealTime RIC project

  • The OOF SON project has updated the SDN-R to use O-RAN aligned O1 YANG models and the RAN-Sim to use O-RAN aligned O1 YANG models

  • Convergence on VES message formats for Performance Management, Fault Management, Configuration Management

Controllers

  • SDN-C is upgraded to OpenDaylight Phosphorus release

  • Enhancements to CCVPN, Network Slicing, and ONAP A1 Interface

Service Design

  • Improved support for TOSCA features

  • Automation Composition Management model

  • Support for large CSAR via S3 storage

Inventory

  • Schema updated for CCVPN use case mainly enhancing and bug fixes of the Cloud Leased Line (CLL) service

ONAP Operations Manager

  • Introduction of Strimzi Kafka Operator

  • Migration of all Kafka native clients to use Strimzi Apache Kafka.

  • Disable VID, Portal

Non-Functional Requirements

The following ‘non-functional’ requirements are followed in the Jakarta Release:

Best Practice

  • ONAP shall use STDOUT for logs collection

  • IPv4/IPv6 dual stack support in ONAP

  • Containers must crash properly when a failure occurs

  • Containers must have no more than one main process

  • Application config should be fully prepared before starting the application container

  • No root (superuser) access to database from application container

Code Quality

  • Each ONAP project shall improve its CII Badging score by improving input validation and documenting it in their CII Badging site

  • Each ONAP project shall define code coverage improvements and achieve at least 55% code coverage

Security

  • Python language 3.8

  • Java language v11

  • All containers must run as non-root user

  • Continue hardcoded passwords removal

  • Flow management must be activated for ONAP

  • Each project updates the vulnerable direct dependencies in their code base

  • Pilot for automating the creation of a Software Bill of Materials (SBOM). Tools for automated SBOM creation are now rolled into the CI chain of ONAP

Documentation

  • Documentation cleaned up. Chapters which include unmaintained projects were removed to avoid misunderstandings at the readership

  • Projects (repositories) which do not create a stable release branch are no longer included in the release documentation. This should help to improve the release management process

  • Beginning with this release we are providing example configuration files for setting up a proper process of documentation creation

  • The Interactive Architecture Overview is updated

  • The guide to set up a development system for documentation is updated

Tests & Integration

  • Create Java and Python base images

  • Adapt robot tests to DCAE project changes: Cloudify to Helm migration

  • New test: basic_cnf_macro

  • Release ONAP data provider tool

  • Automate repositories INFO.yaml updates

Important

Some non-functional requirements are not fully finalized. Please, check details on the Integration

Documentation Sources

The formal ONAP ‘Jakarta’ Release Documentation is available in ReadTheDocs.

The Developer Wiki remains a good source of information on meeting plans and notes from committees, project teams and community events.

CII Best Practice

ONAP has adopted the CII Best Practice Badge Program.

In the Jakarta release,

  • 100% projects passed 90% of the CII badge

  • 85% projects passed the CII badge

  • 11% projects passed the CII Silver badge

Project specific details are in the release notes for each component.

ONAP Maturity Testing Notes

For the Jakarta release, ONAP continues to improve in multiple areas of Scalability, Security, Stability and Performance (S3P) metrics.

In Jakarta the Integration team focussed in

  • Automating ONAP Testing to improve the overall quality

  • Adding security and E2E tests

More details in ONAP Integration Project

Known Issues and Limitations

Known Issues and limitations are documented in each project Release Notes.