e2e

README

End-to-End Tests

Spins up and tests Omni devnets in Docker Compose based on a testnet manifest. To run the CI testnet:

# In repo root
# Install the e2e app
go install github.com/omni-network/omni/e2e

# Build docker image of the code to test.
make build-docker

# Run one of the "manifests" in networks/ directory: e2e -f <manifest>
e2e -f e2e/networks/single.toml

This creates and runs a testnet named single under e2e/runs/single/.

Conceptual Overview

Please refer to the cometBFT E2E test framework for more details.

In order to perform any action on a network (deploy/test/show logs), the following process is followed to create a network Definition:

1. A network is initially declared in a manifest file, see manifests/ folder. It defines the desired network topology. See the e2e/types#Manifest type for details.
2. Then the infrastructure provider (only docker compose supported at the moment) subsequently generates the e2e/types#InfrastructureData from the manifest. This defines the instance IPs and ports of everything we will deploy.
3. Subsequently, we generate a Testnet struct which is basically contains all the configuration/keys/peers/images/files/folders required to deploy a network. See e2e/types#Testnet for details.
4. We then instantiate a netman.Manager which is responsible for deploying portals. It takes a Testnet struct as input.
5. Finally, we instantiate new InfrastructureProvider which can deploy the network. It takes a Testnet struct and InfrastructureData as input.

These objects are then wrapped in a e2e/app#Definition that can be used to perform any action on a network.

Test Stages

The e2e test has the following stages, which can also be executed explicitly by running e2e -f <manifest> <stage>:

• setup: generates configuration files.

• start: starts Docker containers.

• wait: waits for a few blocks to be produced, and for all nodes to catch up to it.

• stop: stops Docker containers.

• cleanup: removes configuration files and Docker containers/networks.

Auxiliary commands:

• logs: outputs all node logs.

• tail: tails (follows) node logs until canceled.