Developer Guide

Developer Guide

Introduction

This document is intended for developers who want to contribute to the project. It contains information about the project structure, how to build the project, and how to run the tests.

TLDR

Want to quickly get Armada running and test it? Install the Pre-requisites and then run:

mage localdev minimal testsuite

To get the UI running, run:

mage ui

A note for Devs on Arm / Windows

There is limited information on issues that appear on Arm / Windows Machines when running this setup.

Feel free to create a ticket if you encounter any issues, and link them to the relavent issue:

Design Docs

Please see these documents for more information about Armadas Design:

Other Useful Developer Docs

Pre-requisites

Using Mage

Mage is a build tool that we use to build Armada. It is similar to Make, but written in Go. It is used to build Armada, run tests, and run other useful commands. To see a list of available commands, run mage -l.

LocalDev Setup

LocalDev provides a reliable and extendable way to install Armada as a developer. It runs the following steps:

Note: If you edit a proto file, you will also need to run mage proto to regenerate the Go code.

It has the following options to customize further steps:

mage localdev minimal is what is used to test the CI pipeline, and is the recommended way to test changes to the core components of Armada.

Testing if LocalDev is working

Running mage testsuite will run the full test suite against the localdev cluster. This is the recommended way to test changes to the core components of Armada.

You can also run the same commands yourself:

go run cmd/armadactl/main.go create queue e2e-test-queue

# To allow Ingress tests to pass
export ARMADA_EXECUTOR_INGRESS_URL="http://localhost"
export ARMADA_EXECUTOR_INGRESS_PORT=5001

go run cmd/testsuite/main.go test --tests "testsuite/testcases/basic/*" --junit junit.xml

Running the UI

In LocalDev, the UI is built seperately with mage ui. To access it, open http://localhost:8089 in your browser.

For more information see the UI Developer Guide.

Choosing components to run

You can set the ARMADA_COMPONENTS environment variable to choose which components to run. It is a comma separated list of components to run. For example, to run only the server and executor, you can run:

export ARMADA_COMPONENTS="server,executor"

Running Pulsar backed scheduler with LocalDev

Ensure your local environment is completely torn down with

mage LocalDevStop

And then run

mage LocalDev minimal-pulsar

Ensure your local dev environment is completely torn down when switching between pulsar backed and legacy setups.

If the eventsingester or the scheduleringester don’t come up then just manually spin them up with docker-compose up.

Debugging

The mage target mage debug supports multiple methods for debugging, and runs the appropriate parts of localdev as required.

NOTE: We are actively accepting contributions for more debugging guides!

It supports the following commands:

VSCode Debugging

After running mage debug vscode, you can attach to the running processes using VSCode. The launch.json file can be found Here

For using VSCode debugging, see the VSCode Debugging Guide.

Delve Debugging

The delve target creates a new docker-compose file: ./docker-compose.dev.yaml with the correct volumes, commands and images for debugging.

If you would like to manually create the compose file and run it yourself, you can run the following commands:

mage createDelveCompose

# You can then start components manually
docker compose -f docker-compose.dev.yaml up -d server executor

After running mage debug delve, you can attach to the running processes using Delve.

$ docker compose exec -it server bash
root@3b5e4089edbb:/app# dlv connect :4000
Type 'help' for list of commands.
(dlv) b (*SubmitServer).CreateQueue
Breakpoint 3 set at 0x1fb3800 for github.com/armadaproject/armada/internal/armada/server.(*SubmitServer).CreateQueue() ./internal/armada/server/submit.go:137
(dlv) c
> github.com/armadaproject/armada/internal/armada/server.(*SubmitServer).CreateQueue() ./internal/armada/server/submit.go:140 (PC: 0x1fb38a0)
   135: }
   136:
=> 137: func (server *SubmitServer) CreateQueue(ctx context.Context, request *api.Queue) (*types.Empty, error) {
   138:         err := checkPermission(server.permissions, ctx, permissions.CreateQueue)
   139:         var ep *ErrUnauthorized
   140:         if errors.As(err, &ep) {
   141:                 return nil, status.Errorf(codes.PermissionDenied, "[CreateQueue] error creating queue %s: %s", request.Name, ep)
   142:         } else if err != nil {
   143:                 return nil, status.Errorf(codes.Unavailable, "[CreateQueue] error checking permissions: %s", err)
   144:         }
   145:
(dlv)

All outputs of delve can be found in the ./delve directory.

External Debug Port Mappings:

Armada service Debug host
server localhost:4000
executor localhost:4001
binoculars localhost:4002
eventingester localhost:4003
lookout localhost:4004
lookoutv2 localhost:4005
lookoutingester localhost:4006
lookoutingesterv2 localhost:4007
jobservice localhost:4008

Other Debugging Methods

Run mage debug local to only spin up the dependencies of Armada, and then run the individual components yourself.

For required enviromental variables, please see The Enviromental Variables Guide.

Finer-Grain Control

If you would like to run the individual mage targets yourself, you can do so. See the Manually Running LocalDev guide for more information.