Tags:

• aura-kpis
• entity
• urm
• dataset

Categories:

• components

Aura entities

Introduction

Aura entities are files stored by different Aura components that contain relevant information related to key system processes or actions: user messages, Aura components interactions, message handling, applications, etc.

They are useful to measure and evaluate the performance of the system against defined objectives through the generation of KPIs, processes tracking, identification of issues and decision-making.

Each Aura component generates a series of entities which are uploaded into Azure Storage for different purposes.

aura-kpis-uploader is the component in charge of the management of entities for every Aura component. On an hourly basis, it uploads all the generated files to Kernel datasets. Once there, specific algorithms are executed to calculate the KPIs of each instance or for other purposes.

Detailed information regarding Aura entities is found in the following documents:

• Types of logs in Aura: Operational logs and KPIs entities logs.
• Description of Aura KPIs blob container.
• Aura Databricks Jobs: Component that imports Avro-formatted files into Kernel datasets.
• Aura KPIs uploaderComponent in charge of handling entities and dimensions.
• KPI entity handler: aura-bot module in charge of managing Aura entities.
• Status codes stored in KPIs entities: List of response codes stored in KPIs by aura-bot.
• Aura entities definition: Database that includes the different entities currently used in Aura.

Types of logs in Aura

aura-bot writes two different types of logs:

• Operational logs

  Operational logs are written using AuraLogger that writes each row, by default, in JSON format in the standard output of the POD running each instance of aura-bot.

  These logs are used to monitor or debug aura-bot. The standard output of each POD is aggregated to be stored in an ElasticSearch cluster, to make it available in Kibana.

• Entities logs

An entity is a specific definition of one of the actors involved in the processing of each activity in aura-bot. So, the bot is in charge of writing the corresponding rows for each activity on each entity.

This section aims to describe how the entities are handled by aura-bot and how the rows are written and included in the entity files.

Logs are written in a blob file: a blob container in Azure Storage using an internal library aura-kpi-handler, that provides classes and utilities to decouple Aura components: how the rows are written from how the needed information is gathered.

The database Aura entities definition includes the different entities currently used in Aura. Entities are generated by different Aura’s modules, each of them in charge of performing a different task: aura-bot, aura-groot, aura-services, etc.

Currently, these logs are used to generate KPIs for the measurement of Aura performance, tracking processes or identifying issues.

Aura KPIs blob container

The following figures show the Aura KPIs blob container and an example of content of the blob container for aura-bot.