Overview

The Scouter server is a rust-based server that is designed to be run independent of the python client. The server is responsible for handling the event system (via Kafka, RabbitMQ or the default queue system), database CRUD operations, and alerting.

Features:

• Event System - Scouter supports various third-party event systems such as Kafka and RabbitMQ. The event system is used to send events to the Scouter server for processing
• Database Storage - The Scouter server leverages Postgres (sqlx) for short-term data storage and DataFusion for long-term data storage
• Alerting - Integrates with OpsGenie and Slack for alerting
• Data Retention and Partitioning - Built-in data retention and partitioning strategies to keep your database clean and performant
• Authentication - Built-in authentication system for users

Getting Started

There are a few different ways to get up an running with the Scouter server.

Prerequisites

• Scouter relies on a Postgres database for storing and retrieving data. You will need to have a Postgres database up and running before you can use Scouter. Scouter currently supports Postgres 16.3 and above.

Once you have a Postgres database up and running, set the following environment variable before starting the Scouter server:

export DATABASE_URI={your_postgres_uri}

The default database URI is postgresql://postgres:postgres@localhost:5432/postgres. You can change this to point to your own Postgres database.

Docker

It is recommended to use one of our pre-built Docker images to get up and running quickly. New docker images are built and tagged on every version release and currently build for the following platforms:

Amd64 (x86_64-unknown-linux-gnu)

Image	Tag Suffix	Features
ubuntu	ubuntu	(kafka, RabbitMQ)
alpine	alpine	(kafka, RabbitMQ)
scratch	scratch	(kafka, RabbitMQ)
debian	debian	(kafka, RabbitMQ)
distroless	distroless	(kafka, RabbitMQ)

Arm64 (aarch64-unknown-linux-gnu)

Image	Tag Suffix	Features
ubuntu	ubuntu	(kafka, RabbitMQ)
alpine	alpine	(kafka, RabbitMQ)
scratch	scratch	(kafka, RabbitMQ)
debian	debian	(kafka, RabbitMQ)
distroless	distroless	(kafka, RabbitMQ)

Pull your image

docker pull demml/scouter:ubuntu-amd64-kafka-latest

Run your image

docker run -d --name scouter -p 8080:8080 demml/scouter:ubuntu-amd64-kafka-latest

Execute prebuilt binary

Binaries for various architectures are published on every release. You can find them on the github release page, download and execute the binary.

Binaries can be found here

Build from source

To build the Scouter server from source, you will need to have the following dependencies installed:

• Rust (with Cargo) link

Then you can build the server using the following command:

cargo build scouter-server --release

Feature Flags

Scouter server is built with a few feature flags that can be enabled or disabled at build time. The following feature flags are available:

• kafka - Installs the necessary dependencies to use Kafka as the event system (rdkafka crate)

cargo build scouter-server --release --features "kafka"

• rabbitmq - Installs the necessary dependencies to use RabbitMQ as the event system (lapin crate)

cargo build scouter-server --release --features "rabbitmq"

Environment Variables

You can set a variety of environment variables to configure the Scouter server to your needs.

Database Variables

Variable	Description	Feature	Default
DATABASE_URI	Database connection URL	All	postgresql://postgres:postgres@localhost:5432/postgres
MAX_POOL_SIZE	Maximum number of database connections in the pool	All	30
DATA_RETENTION_PERIOD	Number of days to retain data in the database. After this time, data will be pushed to long-term storage via DataFusion	All	30

Polling Variables (For background drift detection)

Variable	Description	Feature	Default
POLLING_WORKER_COUNT	Number of polling workers for processing scheduled tasks and alerting	All	4

Kafka Variables (if enabled, for record consumption)

Variable	Description	Feature	Default
KAFKA_BROKERS	Comma-separated list of Kafka broker addresses	Kafka	localhost:9092
KAFKA_TOPIC	Kafka topic for sending data	Kafka	scouter_monitoring
KAFKA_GROUP	Kafka consumer group ID	Kafka	scouter
KAFKA_OFFSET_RESET	Kafka offset reset policy	Kafka	earliest
KAFKA_USERNAME	Kafka username for authentication	Kafka	None
KAFKA_PASSWORD	Kafka password for authentication	Kafka	None
KAFKA_SECURITY_PROTOCOL	Kafka security protocol (e.g., PLAINTEXT, SSL, SASL_PLAINTEXT, SASL_SSL)	Kafka	SASL_SSL
KAFKA_SASL_MECHANISM	Kafka SASL mechanism	Kafka	PLAIN
KAFKA_CERT_LOCATION	Path to the Kafka CA certificate file	Kafka	None

RabbitMQ Variables (if enabled, for record consumption)

Variable	Description	Feature	Default
RABBITMQ_CONSUMER_COUNT	Number of RabbitMQ consumers	RabbitMQ	3
RABBITMQ_PREFETCH_COUNT	Number of messages to prefetch per consumer	RabbitMQ	10
RABBITMQ_ADDR	RabbitMQ server address	RabbitMQ	amqp://guest:[email protected]:5672/%2f
RABBITMQ_QUEUE	RabbitMQ queue name	RabbitMQ	scouter_monitoring
RABBITMQ_CONSUMER_TAG	RabbitMQ consumer tag	RabbitMQ	scouter

Authentication

Scouter has built-in authentication support for JWT tokens. You can set the following environment variables to enable authentication:

• SCOUTER_ENCRYPT_SECRET: Secret key used to sign JWT tokens. If not set, scouter will use a default deterministic key. This is not recommended for production use cases. Scouter requires a pbdkdf2::HmacSha256 key with a length of 32 bytes.
• SCOUTER_REFRESH_SECRET: Secret key used to sign refresh tokens. If not set, scouter will use a default deterministic key. This is not recommended for production use cases. Scouter requires a pbdkdf2::HmacSha256 key with a length of 32 bytes.
• SCOUTER_BOOTSTRAP_KEY: Secret key used to bootstrap the server. This is used to create the initial admin user and should be set to a strong random value. If not set, scouter will use a default. This is also a pbdkdf2::HmacSha256 key with a length of 32 bytes. This can be used as a shared key for integration work as well. For example, when setting up an opsml server, you can user this key to sync user accounts between the two servers.

ObjectStore Variables (For long-term storage)

Variable	Description	Feature	Default
SCOUTER_STORAGE_URI	The URI of the object store to use for long-term storage. The default is ./scouter_storage. Currently, gcs, s3, azure and local storage are supported. If using cloud storage, provide the appropriate prefix and bucket (e.g. gs://scouter, s3://scouter, az://scouter)	ObjectStore	./scouter_storage
AWS_REGION	The AWS region to use for S3 storage. The default is us-east-1	ObjectStore	us-east-1
GOOGLE_ACCOUNT_JSON_BASE64	The base64 encoded JSON key file to use for GCS storage. This is an optional environment variable that can be used in addition to how the object-store GoogleCloudStorageBuilder retrieves credentials	ObjectStore	None

Object Store Providers

Scouter leverage's the object-store crate for writing to object stores. The following object stores are supported. Please refer to the object-store crate for more information on how to configure each object store.

Providers:

• google_cloud_storage - link
• aws_s3 - link
• azure - link