Slide 3 of 14

Configuration pipelines

Configuration pipelines

A configuration pipeline is the unit of remote configuration in Fleet Management. It’s configuration to perform one operation, such as monitoring MySQL. You can match multiple pipelines to a single collector and multiple collectors to a single pipeline.

Each pipeline has a name, configuration content, and matching attributes.

Anatomy of a configuration pipeline

Key terms

TermWhat it means
Remote configurationAll the configuration pipelines matched to a collector
Configuration pipelineThe pieces of config you create and assign, usually representing one monitoring operation

Pipeline components

ComponentDescription
NameUnique identifier for the pipeline
ContentStandard configuration for one operation
Matching attributesRules that assign the pipeline to collectors

Pipelines are reusable. Create once, assign to many collectors using attribute matching.

Script

In the previous slide, we talked about remote configuration as the solution to the scaling challenge. Now let’s get specific about what remote configuration actually is. In Fleet Management, remote configuration is delivered through something called a configuration pipeline.

A configuration pipeline is the unit of remote configuration in Fleet Management. When you “configure remotely,” you’re creating at least one pipeline. When collectors “pull configuration,” they’re pulling a set of one or more pipelines. The terms are closely related—pipelines are how remote configuration works in practice.

Each pipeline has three parts: a unique name, the configuration content, and matching attributes that determine which collectors receive it.

The content is standard configuration—components like prometheus.scrape, loki.source.file, or prometheus.remote_write. You can write custom configs from scratch, or you can use integration templates that generate pre-configured pipelines for common use cases.