Introduction

Juniper Apstra is a powerful automation solution that manages the full life cycle of data center switching fabrics. Apstra’s Intent-Based Networking (IBN) approach to automation helps you design, build, deploy, operate and validate your network.

Apstra validates that:

• The user-supplied inputs are valid.

• The user inputs are consistent and compatible with the constraints of the network.

• The expected telemetry outputs are correct when the network is stable.

• There are no gaps between the expected and actual telemetry.

Once you deploy your network, Apstra collects various telemetry data from its managed devices. This data is automatically aggregated and validated against the intended state of each telemetry type, such as interfaces, LLDP, BGP, and so on. This capability in Apstra is called Intent-Based Analytics, or IBA. IBA is an invaluable tool for obtaining accurate and relevant data for robust operations and informed decision-making.

Starting with Release 4.2.0, Apstra introduces its Custom Telemetry Collection. This collection enables you to easily configure Apstra to collect new telemetry data from managed devices. Apstra then uses that data in IBA probes to visualize and analyze your data.

In this document, you will learn:

• The fundamentals of IBA.

• How to define a custom telemetry service.

• How to create a new IBA probe to visualize and analyze data from your telemetry service.

We'll also walk you through an example use case that shows you how to:

• Define a custom telemetry service that gathers the BFD session state from managed devices.

• Create an IBA probe that ingests and visualizes the BFD session state data.

• Customize your IBA probe to raise anomalies for BFD sessions that are down.

• Store the history of anomalies in a time-series database.

Let’s dive in!