Fjall Docs — Deploy Production-Ready AWS Infrastructure in Minutes

Prerequisites

Complete these first:

Deploy Application
A deployed Fjall application

Overview

As your application grows, add more resources to its infrastructure.ts. The fjall add command edits the file through the codemod engine, then fjall deploy provisions the change on AWS. Every add follows one shape:

fjall add <type> --app <app> --name <PascalCaseName> [--<property> <value> ...]

Part	Rule
`<type>`	One of: `database`, `storage`, `compute`, `messaging`, `cdn`, `network`, `pattern`, `vpc-peer`, `vpc-peer-accepter`, `cross-plan-connection`
`--app`	The application that owns the `infrastructure.ts` file
`--name`	Required, PascalCase (for example `Analytics`, `WebCdn`)
`--<property>`	Optional resource properties, each with a trailing value

There is no colon syntax. fjall add database:Aurora is invalid. The database variant is a property: fjall add database --type Aurora.

Add a database

Add a database and pick the variant with --type:

fjall add database --app api --name Analytics --type Aurora

The variant determines the database engine and topology.

`--type` value	Use case
`Aurora`	Aurora Serverless clusters
`Instance`	Standard RDS instances
`GlobalAurora`	Multi-region Aurora clusters
`ClickHouse`	Analytics database for high-volume telemetry
`DynamoDB`	Serverless key-value and document store

Multiple databases

An application can hold several databases. Give each a distinct PascalCase name:

# Primary transactional database
fjall add database --app api --name Primary --type Aurora

# Analytics database
fjall add database --app api --name Analytics --type Aurora

# Secondary instance
fjall add database --app api --name Secondary --type Instance

Interactive connection prompt

When you run fjall add interactively and existing compute resources are present, the wizard offers to wire the database to them:

? Found existing compute resources. Connect this database to them? Yes

? Select services to connect to the database:
  ◉ ApiCompute
  ◯ ApiWorker

If exactly one compute target exists, the wizard selects it automatically and skips the multi-select.

Add storage (S3)

Add an S3 bucket:

fjall add storage --app api --name Assets

Add compute

The compute type covers Lambda functions and ECS services. Pick the variant interactively, or pass it as a property in non-interactive mode.

fjall add compute --app api --name ImageProcessor

The wizard prompts for the variant and its settings:

? Compute type:
  ❯ Lambda (event-driven function)
    ECS (long-running service)
? Memory (MB): 1024
? Timeout (seconds): 300
? Connect to storage? Yes

Variant	Use case
Lambda	Event processing, APIs
ECS	Long-running services

Add a CDN

Add a CloudFront distribution:

fjall add cdn --app api --name WebCdn

What happens on add

When you add a resource, Fjall configures the supporting wiring automatically:

Security groups open the required ports
Environment variables carry connection strings into connected compute
Secrets hold database passwords in Secrets Manager
IAM roles receive the permissions the resource needs

Connection environment variables

Connected compute receives connection details as environment variables. The first database uses unsuffixed names, and additional databases get numbered suffixes:

# First database
DATABASE_HOST=api-db.cluster-xxx.amazonaws.com
DATABASE_PORT=5432
DATABASE_NAME=ApiDatabase
DATABASE_PASSWORD=<from-secrets-manager>

# Additional databases get numbered suffixes
DATABASE_HOST_1=analytics-db.cluster-xxx.amazonaws.com
DATABASE_PORT_1=5432
DATABASE_NAME_1=AnalyticsDB
DATABASE_PASSWORD_1=<from-secrets-manager>

Worked examples

Microservice setup

# API function
fjall add compute --app api --name ApiGateway

# Authentication service
fjall add compute --app api --name AuthService

# User database
fjall add database --app api --name UserDb --type Aurora

Data pipeline

# Data ingestion function
fjall add compute --app api --name DataIngestion

# Processing service
fjall add compute --app api --name DataProcessor

# Data warehouse
fjall add database --app api --name DataWarehouse --type Aurora

Non-interactive mode

Pass --non-interactive to add a resource without prompts. Supply every required property as a flag:

# Aurora database
fjall add database \
  --app api \
  --name Analytics \
  --type Aurora \
  --non-interactive

# Lambda function
fjall add compute \
  --app api \
  --name Processor \
  --non-interactive

Deploy after adding

Adding a resource only edits infrastructure.ts. Provision it with a deploy:

fjall deploy api

Fjall reports what it will change:

Resources to create:
  + AWS::RDS::DBCluster (Analytics)
  + AWS::Lambda::Function (Processor)
  + AWS::EC2::SecurityGroup (2)

Resources to update:
  ~ AWS::ECS::TaskDefinition (environment variables)

List resources

fjall list reads the resources defined in an application’s infrastructure.ts. The application is selected with -a, --app, and an optional [type] positional filters the output:

# All resources in the api application
fjall list --app api

# Databases only
fjall list database --app api

api:
  Compute:
    - ApiCompute (ECS/Fargate)
    - ImageProcessor (Lambda)
  Database:
    - Primary (Aurora)
    - Analytics (Aurora)
  Storage:
    - Assets (S3)

Best practices

Practice	Why
Plan connections	Decide which resources need to communicate before adding them
Use descriptive PascalCase names	Resources stay easy to identify in `infrastructure.ts` and the AWS console
Right-size resources	Match instance sizes to expected load to control cost
Add and deploy one resource at a time	Smaller changes are easier to review and roll back

All resource types

Type	Command	Description
Database	`fjall add database --type <variant>`	RDS, Aurora, ClickHouse, or DynamoDB
Compute	`fjall add compute`	Lambda function or ECS service
Storage	`fjall add storage`	S3 bucket
CDN	`fjall add cdn`	CloudFront distribution
Messaging	`fjall add messaging`	EventBridge bus or queue
Network	`fjall add network`	Additional VPC network resources
Pattern	`fjall add pattern`	Composed application pattern

Add resources at any time. Fjall configures networking and security automatically on the next deploy.

Install & Setup

Application Deploy

Organisation Setup

Organisation Deploy

Add Resources

Overview

Add a database

Multiple databases

Interactive connection prompt

Add storage (S3)

Add compute

Add a CDN

What happens on add

Connection environment variables

Worked examples

Microservice setup

Data pipeline

Non-interactive mode

Deploy after adding

List resources

Best practices

All resource types

Next Steps

Deploy Application

Remove Resources

​Overview

​Add a database

​Multiple databases

​Interactive connection prompt

​Add storage (S3)

​Add compute

​Add a CDN

​What happens on add

​Connection environment variables

​Worked examples

​Microservice setup

​Data pipeline

​Non-interactive mode

​Deploy after adding

​List resources

​Best practices

​All resource types

​Next Steps

Deploy Application

Remove Resources

Overview

Add a database

Multiple databases

Interactive connection prompt

Add storage (S3)

Add compute

Add a CDN

What happens on add

Connection environment variables

Worked examples

Microservice setup

Data pipeline

Non-interactive mode

Deploy after adding

List resources

Best practices

All resource types

Next Steps