Internal Release Notes – 23.50.0 OpenTV Video Platform - DRAFT

Release summary 

This release comes with a new functionality, User Profiles, which will allow the operators to create and manage different profiles for their subscribers. It also has several features that allow OPF3 integration with the Velocix platform for main use cases, as well as several enhancements around recommendations and improvements around RAILS to enrich user experience. 

Playbook changes to change in Auth Facade egress rules

Internal, Claro Brazil

This feature supports the essential networking infrastructure requirements e.g. VPN, around Claro Brasil migration, including ingress and egress to the OPF platform and the various Claro components that will remain on premise post migration.

Device Entitlements API (devicePlayoutClass)

Product

RMG provides a new API to get the entitlements based on the calling device and account without requiring specific filtering.

This is applicable when "devicePlayoutClass" functionality is used to describe products.

Mi Claro Application Deeplinking to ClaroTV 

Internal, Claro Colombia

This feature is to allow a user who has already signed into their MiClaro App to launch ION without having to enter their user credentials a second time.

This would only typically happen when ION has never been signed into previously (first time installation, no valid refresh token), or the client refresh token has expired.

Additionally, Claro Colombia is requesting add to the deeplink a service where if the app is not installed yet the user should be re-directed to the App Store (similar as is working now on the Nagra Sports App). 

Enhance RAILS Hierarchy views to return properties

Product

This feature enhances RAILS hierarchy views to return the template properties (and those of it's children).

Enhance RAILS skeleton response to include extended metadata

Product

This feature to enhance the RAILS skeleton response to include all extended metadata. This should be similar to a true limit=0 response (i.e, the RAIL and all it's metadata, but no attempt to gather or resolve any content).

Support a NAGRA PRI/NPVR route to Velocix origin server recorder interface

Product

This feature is to support an implementation of the NAGRA PRI interface to Velocix origin CMAPI interface.

Migrate Velocix client recordings to NAGRA User Recordings

Internal / 4iG

This feature supports the extraction of Velocix asset recording metadata and their construction into Nagra UserRecording objects for specified accounts.

Support STCU encoder type for Velocix origin

Product

This feature is to support new Velocix origin recorder.

As part of the STCU integration, a specific encoder type is required in CPM to support the live format in use by Velocix.

Billing per device types

Internal, Altice US

This feature is to extract data for billing purposes for SFR and AUSA,

• Number of devices per device type, as billing price may be different per device type

• Absolute number when the Operator wants to check consistency with his CRM DB or when Nagra wants to monitor a migration

• Increase per month for Operator billing

EPG content delivery enhancement – multiple channels

Product

This feature is to allow multiple channels to be specified in the /epg timeslice API. Currently it supports either all channels or one channel:

• /metadata/v1/epg?from=-8h&to=8h

• /metadata/v1/epg?from=-8h&to=8&channel=bbc_one

This will provide an efficiency gain in the client, preventing the need to call the individual channel API multiple times.

SSO Integration (AGS)

Internal, 4iG

This feature is to integrate with 4iG's Single Sign-On solution.

Title resolution for template views

Product

This feature is to fix an issue where the title for each rail is not the same as in the API that was previously used on the other releases.

Account Enhancements – User Profiles

Product

This feature provides additional functionality to the initial User Profiles feature, supporting CRM Gateway API access and additional client-facing APIs to manage and remove User Profiles for an account.

CWM-GW required to support SMSDomain V2.2

Internal - Altice US

Altice-US have changed their CRM platform (AmDocs) to use SMSDomain V2.2 instead of V2.0

Allow Signon "session" token to be generated from any Operator

Internal

The current "session" token creation for application deeplinking has been created with AGS and has been made "Claro Columbia" specific, when it could be used by any operator.
This feature is to make the functionality available for any operator.
This functionality would then be used by 4iG for there "Simplified Mobile Device Signon flow".

Client general API facade

Internal, Claro Brazil

This feature provides performance validation of the API endpoints supported by the Claro auth facade that are related to general Client support.

STB Upgrade management

Internal, Claro Brazil

This feature provides performance validation of the API endpoints supported by the Claro auth facade that are related STB upgrade manager process.

Recommendations enhancement – content exclusions

Product

This is a feature that covers the current gap in recommendation functionality, whereby content can be excluded from the recommendation process.

Recommendations enhancement – additional client control flags

Product

This feature intends to push additional client control flags into Ncanto. This feature covers the possibility of converting all the metadata fields into core fields.

Regionalized DVB info

Product

This feature is to add regional support for DVB info.

Client MDRM and licence API facade

Internal, Claro Brazil

This feature provides performance validation of the API endpoints supported by the Claro auth facade that are related to content authorization and licensing.

Ingress re-architecture - continuation

Internal

The purpose of the feature is to rearchitect the ingress solution of the OPF3 platform, to be able to remove the "SNI-ROUTER" EC2s and the "SERVICE-ADVERTISER" ad-hoc solution.

D2Go - EKT Support for LIVE use cases

Internal, Euskaltel

This feature supports OPF3 Download to Go headend requirements for Live and Catch-up content management.

OPF3 Client facade - additional service APIs - Ph2 - Q4

Internal, HOT

This feature identifies and resolves any gaps in migration facades for Medialive 2.x Client API requests used by HOT devices. This will allow use of existing Client devices in a compatibility mode where all the original Client application requests can be serviced by OPF 3 backend services without any change to Client applications.

OpCon channel logo  

Product

This feature is for all operators. When using OpCon, they will now be able to manage channel images directly from the OpCon channel management screens. This includes:

• Uploading new images

• Deleting existing images

• Setting preferred images 

Traceability framework is available in OPF3 and ISTIO traces can be visualized on Grafana Tempo - Continuation

Internal

The Goal of this feature is to introduce Open Telemetry Collector and Grafana Tempo onto the OPF3 platform.  OTEL Collector is in charge of gathering the traces generated from the different applications that are processing a given requests.

Grafana Tempo would be the UI to navigate through the information provided by Applications' traces.

CRM-GW configuration for SOAP API to OPF & SSP

Internal, Claro Colombia

CRM-GW currently provisions accounts into OPF, SSP and Medialive SDP.

A new configuration for CRM-GW is provided to provision accounts & devices into OPF and SSP only, to allow the decommissioning of the On-Prem MediaLive platform.

HTTP-ROUTER nginx improvements

Internal

The goal of the feature is to perform the following improvements in the HTTP-ROUTER component:

• Upgrade nginx version from 1.9 (very very old one) to 1.20

• All HTTP-ROUTER ConfigMaps are identified and documented in Confluence and understand all configs in those ConfigMaps

• No manual changes are needed for any customer/platform. If we need to customize any parameter/value, we should handle it accordingly in the config repo

• Identify all LUA scripts and owners

• Move HTTP-ROUTER code to GitLab

• Use Helm to deploy http-router

• Have http-router as csit-ready

• Investigate how to get rid of the http-router restarts when app configs change

[BCP DRP] E2E readiness: restore environment (simple customer customizations) + nanoSAT (Phase 1.2)

Internal

Today, we are able to restore/deploy an environment (infra), we do partial restore of PROD dumps for customers.  There is no e2e test demoing that we are able to fully restore an environment from scratch and validate.

BCP/DRP : We are capable of restoring any customer environment in a different region and account from scratch (without time constraint)

[BCP DRP] E2E readiness: restore environment (without customer customizations) + nanoSAT (Phase 1.1) 

Internal

We need to test and document restoration procedure to start increasing our confidence on our capability to restore prod systems.

Multitenant API (Shared OPF/Umbrella)

Internal

To further develop our ability to support multi-operator deployments of OPF we require that operators are able to manage and configure the platform as per a one-operator OPF deployment without the need for a lead operator or OPF representative to take responsibility for making the changes. 

This feature will allow multiple operators to use the Operator tier endpoints as per current implementations with the addition of a 'check' to ensure that tenants only modify the data that relates to their tenant id.

Remove HTTPRouter from internal service to service calls

Internal

Today in OPF3 all internal requests from one service to another are routed through HTTPRouter. This is largely for legacy reasons allowing some level of caching, logging aggregation, manipulation of requests via Lua scripts, token management, Keycloak.

This feature is to move services to make direct calls to each other via K8s. 

OPF3's MongoDB community replacement by MongoDB Atlas and critical services segregated

Internal

We use an ancient version of Mongo BD community which is not supported and have limited tooling to analyze what is going on with the database.

OPF3 product is composed of several modules/services, out of which at least 15 services are using MongoDB.
Today's deployment architecture is to have one MongoDB cluster serving the 15 services.

• Used MongoDB Atlas as a replacement of MongoDB community.

• Using MongoDB as a service will ease the work of DBAs: Will be up to date, Backups are self-managed, administration is more efficient

Gracenote ON API updates

Product

This feature is to take care of some upgrades to the Gracenote ON API Schema. CIM-GO has been enhanced to comply with the latest schema changes.   

[Logging] Clean-up OPF app logs ingested in Elasticsearch  

Internal

As per today, logs generated by application are massive and not easily usable for troubleshooting, alerting and monitoring.  Some applications are generating too much garbage logs while others application not generating logs with the useful information.

Purpose of that feature to align OPF applications on a common guidance and log format in order to:

• Improve information available for troubleshooting & OPF observability

• Keep logging cost controlled

• Enable log queries/aggregations in Kibana

• Enable business alerting and dashboarding based on application logs

Event <> Editorial <> Technical Data Model Optimization

Internal

Feature to support content aggregation in the platform (not necessarily the act of aggregation, but support for aggregated behaviour). This covers single event <> editorial relationships <> technical relationships,  search, recommendations, content / episode / season / series recordings etc.

RabbitMQ Services Standardisation  

Internal

The use of events, notifications, publishers and consumers makes up a core part of the OPF platform, key functionality fails when the Rabbit infrastructure is not maintained or configured correctly and in situations where messages are lost or fail to process, we see issues downstream across the platform and reporting. 

This feature is to introduce a standard approach to make the usage more consistent with a minimum level of  

• Introduce standards for message headers

• Introduce a standard minimum approach to dealing with producer errors and retries

• Introduce a standard minimum approach to dealing with consumer errors and retries 

• Introduce a standard approach to the use of DLQ's;  no value in filling up a queue with no way to re-run the message 

• Documentation to be produced for every consumer to detail the message using a standard tooling

Keycloak resources deployment automation (incl. Realms, Clients, Roles and IDPs with AD federation)

Internal

The goal of that feature is to fully automate our Keycloak deployment, including:

• Keycloak module deployment (already automated with Ansible Playbook)

• Keycloak resources deployment (Realms, Clients, Roles and IDPs with Nagra Azure AD configuration)

• Keycloak integration with OpCon/UIC and monitoring UIs

Support from WFM/Elemental adapter to encode different types using a single encoding template of AWS MediaConvert

Product, Sports

VOD encoding support from OPF3 for a single encoding template for different outputs as opposed to a different template for each output group in AWS Elemental MediaConvert.

Umbrella OpCon for Chopin

Internal, Chopin

The solution delivered for Umbrella as part of the Q1 release does not meet the requirements that Chopin has as a lead tenant. They require the ability to manage accounts and entitlements via OpCon and not just via the API's.

This feature will add that functionality to Opcon.

[BCM] MongoDB credentials

Internal

SRE have identified some issues around Automated Key Rotation so the decision has been made that for the time being keys will manually rotated when a system is upgraded.

As such there is no longer a need to integrate the scripts for Automatic key rotation provided by SRE.   This feature is to migrate away from Fabric8 to JKube, this will put us in a good position when Automated Key rotation is re-introduced in the future.

Altice-US Migration. Phase 1. ADM SSP Password Support

Internal, Altice US

In an OPF3 deployment it is possible to change the password hashing configuration within K8.  However, this config isn't exposed in Puppet/Foreman for SSP so manual file changes are required.

In relation to migration, SDP has a hash iteration of 1000, and hashLength of 256.

If ADM doesn't have the same config then it's not possible to sign-on with migrated passwords.  The default ADM config is:  hashLengthBits: 128  &  hashIterationCount: 

This feature is exposure of the config so it can be managed within Foreman/Puppet.

Altice-US Support for _Id in SSP token generation

Internal, Altice US

In a SSP deployment the device_id field should be used as the device identifier within the Signon-token, and NOT the device Public ID.  This change already exists in some APIs, I believe specifically for TEF.

Some change is required to 'detect' if the deployment is SSP and if so, to use the _id field of the device in the token.

Proxy VPC Egress targets load balancing

Internal

Currently egress targets in the Proxy VPC must be defined per IP and port : we cannot use DNS resolution to define targets.  When we are targetting a service behind a DNS type of load balancing where different hosts are exposing the same service on the same port, we must create a target per IP address and port  and hence we lose the load balancing capability; if any of the target host fails we are not able to  recover unless the application embeds the logic for the failover internally.

Goal of this feature is to adapt Proxy VPC egress deployment to handle all targets of the same service (port) as a list of IPs and put them under the same target group instead of creating a different target group per IP.

EKT DVB-C purge content

Product

DVB-C content is provided pre-encoded and OPF3 simply copies this content to the relevant destination in the Edgeware NAS as part of the content preparation. The purpose of this feature is to provide a new OPF3 workflow adapter than can be used to purge content from the Edgeware NAS.

Enable ADM & RMG with Reverse Auth Facade in deployment

Internal

Ability to enable ADM & RMG in a deployment where Reverse Auth Facade currently exists for authenticating user recording request with ML-SDP.

CIM-GO: Channel ID Mappings for 

Internal

At QXC, somehow the wrong channel/station ids are configured and are being used as part of cim-goc ingestion, ex: HBO UK is used instead of HBO US.

But with cim-go and new API key, those channels (for ex: HBO UK) are not available in US region. 

So based on the discussions with stakeholders, multiple options were proposed and one of the solution option is to define a mapping between those channels and ingest HBO US channel metadata into HBO UK channel created in OPF. 

Upgrade Neo4j Database to 4.4.LTS and respective Drivers/other dependencies

Internal

We need to do database upgrade based on various factors like bugfixes, taking advantage of new features and most important is to be in line with the official database release support.

The Neo4j 4.3.19 (currently in OPF3 ) is already out of support. So we need to upgrade  to the 4.4.LTS. 4.4.LTS is long term support till June 30, 2025.   

AWS Elemental MediaPackage integration with third-party CDN 

Product

The live-to-VOD assets or LTCU assets extracted/generated at AWS MediaPackage are directly served from origin/S3 bucket as of today. From this release, it will push/publish that asset to Lumen CDN and serve it from the CDN.  

OPF Auto Scaling finalization 

Internal

Auto Scale of OPF3 front end components.  All modules highlighted in Green, in the confluence page OPF Auto-Scaling - UEX-PU Platform Architecture - Confluence (atlassian.net)

AutoScaling. - OTV Performances - Confluence (atlassian.net)

[SLA] Simplified Validity Dates - ProductLinks  

Internal

This feature is to align on the simplification of validity dates on ProductLinks.

[SLA] RabbitMQ cluster HA and upgrade Rabbitmq cluster to 3.12-management 

Internal

In current OPF deployment, RabbitMQ cluster is deployed on 3 nodes, but application only connect to the first node. And when that node is done applications are no longer sent or consume messages. We do not benefit of service HA of this cluster.

Also We need to do rabbitmq  database upgrade to 3.12  based on various factors like bugfixes, taking advantage of new features and most important is to be in line with the official release support.

In order to improve global OPF resiliency and SLA, we need application being able to continue using RabbitMQ even if one node is down (due to failure and maintenance).

CPM Purge of expired and soflty deleted entities  

Internal

A Set of functionality has been provided  by CPM to purge naturally expired and soft deleted entities once the retention period of the entities has been reached. The Live and VOD content validity differs as Live content is valid for shorter duration and VOD content stays in the system for longer duration. That's why the retention period of Live and VOD contents are different.

These set of API's are being invoked through a background Job based on  cron expressions. This Job also maintain a sequence through which purge of entities will be followed.

[SLA] Removal of remaining RMG <> CPM dependencies

Internal

This feature covers: Directing the remaining RMG Operator level Entitlements requests to MDS rather than CPM,  Directing all CPM releated calls in AuthFacade to MDS.

[CLB][PH4] - PPV Products & Purchase (Continuation)

Internal, Claro Brazil

Ability for clients to purchase PPV products.  Entitlements need adding to RMG on SSP (Broadcast CONNECT devices) and OPF.

Adding on products to events with "tags" to indicate relationship to technical channel.

[HOT] OPF3 Client facade - additional service APIs - Ph2

Internal, HOT

This feature identifies and resolves any gaps in migration facades for Medialive 2.x Client API requests used by HOT devices. This will allow use of existing Client devices in a compatibility mode where all the original Client application requests can be serviced by OPF 3 backend services without any change to Client applications.

[HOT] Phase 2 Client Compatibility

Internal, HOT

Any Hot-specific XML transforms (E-32729) need to still be present on the deprecation of CEX.  This compatibility needs to be provided within CDF.

Media card aggregation API improvements

Internal

Two Improvements: (1) Media card aggregation API follow up using Search Facade for show more seasons with latest content as desc. (2) Year as parameter for grouping items when seasons are not available.

Automate SAS re-sync when upgraded

Internal

Currently each new version of SAS requires a manual resync.  this feature is to automate all possible cases to remove one more manual task from the upgrade process. 

Foxtel Feedback on OpCon

Internal, Foxtel

Foxtel feedback based on discussion around OpCon

• Some fields to add and remove from the content advanced search. Foxtel to supply current list of search fields. 

• Ability to select channel tags from a list, instead of a free text entry. Foxtel to supply a list of channel  name to channel tag mappings. 

• Change node status indicator on the node list to be based on Foxtel custom fields (CUST_PublishStartTime) and (CUST_PublishEndTime) instead of Node start and end which Foxtel don't use. Status is slightly different as no dates counts as "active' in the Foxtel system        • Expired

• Future

• Active no dates

• Active with dates

• Support both portrait and landscape images in Manage Content. Foxtel have landscape images for none Movie content.

• Fully test the new manage content screen with Foxtel data. Foxtel to supply  data.

• Include channel tag in the information popup on manage content

• Filter the node link grid by node link status (SAS impact)

[TEF] Manage Job Priority 

Internal, Telefonica

Telefónica would like to be able to edit the status and priority of the jobs in OpConsole (SCP-258) but there is no option for that. Both in an individual form and for several jobs at the same time.  Presently the job priority can be viewed within OpCon, and changed via API.

The requirement is to set the priority at the content, and then associated jobs will derive their priority based on this. It must also be possible to change the job, and content, priority within OpCon.

[TEF] Search by Source Asset URL in OpCon

Internal, Telefonica

This feature is to:

• support the search of content based on the source asset URI

• include display of the source asset on the content screen.

Optimisation of CRM-GW for bocPurchase flow Spill over of E-36149

Internal, Foxtel

Recent performance issues at Foxtel have identified issues with CRM-GW's ability to service a high level call rate to bocPurchase. The current workaround is for Foxtel to deploy an increased number of CRM_GW and CAG instances, this has cost implications.

Engineering have identified a number of changes which will significantly improve the call rate that the bocPurchase API can support. This feature is to cover those changes.

OpCon PIN reset

Product

This feature adds the ability to reset PINs on a subscriber account via Operator Console.

Profile Age Rating Cache Impact

Internal

Caching for template results, CB needs to take into consideration the profile's age settings.

The age rating from the profile will be supplied by the client as part of the request this needs to be added to the cache key, 

When requesting Rails/Template/Individual Items the client will have a profile and that profile will be for a specific max age rating, the profile age rating will be present in the token. 

only return the contents with an age rating < the specified age rating and make sure the cache works.  

D2G Full OPF3 service

Product

Download to Go supports operator requirements for marking some Content to be downloadable to Client for offline playout. This is managed at the CMS Product and Content level and allows Content to be streamed and/or downloadable where applicable, with distinct settable ranges of validity for each case. Settings are managed by operator via a combination of

• CMS Ingest of Product and Content

• CMS API from a CRM

• Visually via Nagra OpCon.

 A simple default fallback duration is available where operators want all Content to have a common download duration without having to manage individual Product download settings.

[OPF] [ProxyVPC] Dynamic egress targets reconfiguration

Internal

Today, when egress targets are deployed in the Proxy vpc they are targeting a list of IP address and ports. Recent experience during integration with 4iG shows that our infrastructure implementation on this is not reliable enough to cope with a dynamic change in an IP address.

[4iG] Parental Ratings - Series / Season

Product

Feature to enhance the series / season structure and support within CMS and MDS.  

Support for parental ratings provides the ability to correctly include / exclude Series & Seasons

User Recordings - Parental Ratings

Product

User Recordings is currently unable to support the multiple parental ratings format introduced in 2018/19

which manifests into two key issues:

• Only a single parental rating value is provided, populated in both the recording ratings field and the editorial Rating field, and only when MDS is configured to present the value as a Rating field

• Only a single rating precedence value is considered, populated in the ratingPrecedence.predecessor and ratingPrecedence.successor fields, again only when MDS is configured to present the value as a Rating field

This feature is for User Recordings to:

• Handle MDS presenting parental ratings in a Rating field (backwards compatible mode)

• Handle MDS presenting parental ratings in a Ratings field, and populate those ratings (as the value is an array)

User Recordings - Locale & Device Filter Enhancements

Product

Feature to optimise the response size of the User Recordings APIs by allowing clients to filter by locale and device

Enhance RMS metadata coverage

Internal

This feature is to address the gaps noted during migration of existing npvr recordings (Velocix cVRM) to OPF User Recording in terms of metadata coverage. 

This requires changes to RMG to support additional fields, and on deployment has a pre-requisite that 4iG keep at least 90 days of metadata available online to allow retrieval of existing metadata into new OPF user recordings.

Compatibility matrix 

OPF	Q4 : 23.50.STD0
SSP 24Q1GA (24.10STD1)
DWH (24Q1 build)

List of new issues

The following issues are newly opened in this OPF3_23.50STD0 release.

List of known issues

The following issues are still open (not resolved) in this OPF3_23.50STD0 release.

List of resolved issues

The following issues are resolved as part of the OPF3 (23.50 STD0) release.

Deprecated APIs