Aura – Developers Workspace

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Configure your releases

This section group guidelines with the specific configuration required by an OB for installing a particular Aura version

Index

Dependencies update in Prince 10.0.0 release

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Work with Kernel

Guidelines that include different tasks required for working with Kernel

Index

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

General operational guidelines

This section includes general-purpose guidelines for the operation of Aura that are transversal to different Aura components

Index

Aura operational flows: Flowcharts related to Direct Line, use cases and key channels in Aura
Description of Aura make-up processes
Hot swapping processes in Aura
Procedure for managing locales in Aura
Guidelines for the elimination of expired Aura users
Information regarding Aura Continuous Integration

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Install and configure in Aura Virtual Assistant

Install components in Aura Virtual Assistant, customize your system once deployed through the activation/deactivation of features and adapt the configuration to your requirements

Introduction

Once Aura Virtual Assistant is deployed, OBs can adapt their system to their specific requirements and users need in terms of components, types of clients, demanded experiences, etc. Tasks in this section can be organized in four groups:

Installation of components in Aura Virtual Assistant

Get started with Aura Bot: install and configure aura-bot
Install global use cases: guidelines for the installation of global use cases and global dialogs in charge of their logic.

Activation of features

After deployment, learn how to activate or deactivate features in Aura such as the aura-file-manager

Configure Aura

Use the Aura configuration updater to easily change channels configuration and POEditor resources.

Keda configuration

Guidelines for the configuration of Keda, used for the autoscaling of pods

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Develop features over Aura Platform

Guidelines for developers willing to develop features over Aura Platform

Introduction

Aura is designed as a configurable Platform in which developers can build specific features both for customizing their Aura system or in order to contribute to Aura global code.

For this purpose, Aura Platform put at the OBs’ disposal tools, guidelines and rules that should be followed by the developers:

Use Aura utilities: specific procedures based on Microsoft Bot Framework 4 functionalities that ease the execution of different tasks in Aura.
Develop an Aura bridge plugin: guidelines for the generation of a new plugin and its activation/deactivation in aura-bridge
Follow common rules in order to contribute to Aura Platform global repositories
Know how Aura Platform releases are deployed, following the “release train” software development methodology, through the Aura Release Train Manager

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Manage channels in Aura

Guidelines for the management of channels in Aura: how to activate of different channels; connect a channel to Auraline and more

Introduction

An Aura channel is any communication means a Telefónica client may use to interact with Aura, typically to gather information about, as well as to manage, the client’s Telefónica products and services.

Firstly, we recommend reading the descriptive documentation regarding channels in Aura:

Introduction to Aura channels
Aura communication protocols
Channels authentication
Aura channel model: updated model for the communication between Aura and a channel.
Descriptive information regarding key channels available in Aura: WhatsApp channel and Aura Webclient

The current section includes practical guidelines for the management of channels in Aura:

General guidelines

Guidelines for specific OBs/scenarios

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Manage skills

The following guidelines include the detailed processes required for working with skills in Aura new distributed architecture (Phase 1)

ℹ️ Before facing these documents, The Aura Platform Team highly recommends reading the basic information regarding Aura distributed architecture to have a clear overview of the functional architecture, its benefits and foundations

Introduction

Aura new architecture is conceived as a multi-bot system, based on Microsoft skills architecture, that works with an orchestrator at the top (Aura Root). Currently, skills are independent domain bots able to manage specific experiences (use cases) of the same type.

In this framework, constructors can develop and integrate into Aura different type of skills, as independent modules:

Skills developed over Aura cluster (Microsoft Bot Framework Direct Line protocol).
Skills developed outside Aura cluster with a technology compatible with Microsoft Bot Framework (Composer, etc.).
Third-party bots, developed with other technologies available on the market. These skills will require an adapter for its communication with Aura Root.

In the phase 1 of the architecture implementation, these are the main tasks that must be executed by constructors in order to make the most of Aura:

Develop a skill
Connect a skill to Aura Root
Connect a skill to Aura Minigroot for testing purposes
Message exchange between channels and skills
Hot swapping processes to add or delete a skill

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Test Aura

Test the proper operation of Aura and its components

Introduction

Discover the different types of tests that can be executed to check the performance of Aura system or of an specific component:

QA global test set (QA tool): tool for OB developers to check the quality of Aura system, with their own configuration and installed use cases.
Tests for Aura Bot components: different types of tests that can be executed over Aura Bot.
Available Postman collections for APIs testing.

Docs:

Mon, 01 Jan 0001 00:00:00 +0000

Monitor Aura system

The operation of your Aura system can be monitored in a continuous basis. Discover Aura monitoring tools, the different logs and metrics that are generated and how to implement an alert system.

Introduction

Aura monitoring system is crucial to control how Aura system works internally, in order to keep the service functional and, moreover, to understand the behavior of our clients, leading to evolve Aura accordingly.

Monitoring is based in the generation by different Aura components of logs and metrics, that are stored for their subsequent analysis and decision-making based on the obtained results. Both logs and metrics together create a complementary observability foundation to have an essential knowledge of the system performance in predictable and unpredictable ways.

Within this process, key external monitoring tools are used, such as ElasticSearch, Kibana and Grafana.

Stages in Aura monitoring process

Sections below show, at a glance, the steps and recommended tools both for logs and metrics management and include links to the corresponding documents for more details.

Aura logs management

Logs are files generated by different Aura components that record specific single events, warnings and errors as they occur.

Logs generation
Different Aura components generate logs every time a relevant event happens.
Logs storage
Aura logs are stored in Elasticsearch.
Logs visualization and analysis
Logs can be managed through different tools:
- Kibana: basic operational logging analyzer of Aura.
- Moreover, two additional tools, fluentd and Grafana can be used for specific aspect related to logs management.

📃 Read here detailed information regarding Aura logs management.

Aura metrics management

Metrics offer an aggregated view of Aura performance based on meaningful aggregated logs. They are typically generated at fixed-time intervals and represent a specific aspect of the monitored system.

Metrics generation
Different Aura components generate metrics periodically based on aggregated logs.
Metrics storage
Once generated, Aura metrics are pooled by Prometheus, which is in charge of gathering and exposing them.
Metrics analysis
Aura metrics are analyzed in order to have a meaningful interpretation of data and to obtain an overall evaluation of Aura’s performance.

For the management of metrics, we recommend using Aura dashboards, which are generated in Grafana. These dashboards can be retrieved by making queries to the system.

📃 Read here detailed information regarding Aura metrics management.

Aura alerts

Prometheus has a list of alert rules that are part of the platform configuration and can be editable.

📃 Discover the alerts currently set for Aura system in Aura alerts document.