Trento Playground

A simple setup delivering a local trento playground that lets you control its behaviors.

It targets whoever wants to try or showcase trento features, with the ability to instrument the system to behave in a certain way.

PMs, POs, presales engineers, etc...

Dependencies

In order to leverage this playground make sure to have installed

• Docker
• Docker Compose
• make
• git

Windows users

Install Docker Desktop and/or leverage Windows Subsystem for Linux (WSL).

Mac users

Playground allows to override the docker compose configuration to allow the explicit specification of the image platform for usage on Mac.

Copy the docker-compose.override.yaml.dist file to docker-compose.override.yaml.

$ cp docker-compose.override.yaml.dist docker-compose.override.yaml

The default content of the compose override file already specifies linux/amd64 as the platform for the trento-web and wanda services

services:
  trento-web:
    platform: linux/amd64

  wanda:
    platform: linux/amd64

Usage

1. clone the current repository
2. enter in the directory
3. run commands

$ git clone [email protected]:trento-project/playground.git
$ cd playground
$ make

Commands

Run make or make help to see the list of available commands.

Usage:
 make [target]
 make [target] [args]

Available targets:
 help:                    Help
 clean:                   Cleans up local setup
 get-scenarios:           Prepares photofinish scenarios
 get-catalog:             Prepares Checks catalog
 get-facts-config:        Prepares fake facts configuration
 start:                   Starts containers
 load-default-scenario:   Loads default photofinish scenario healthy-27-node-SAP-cluster
 load-scenario:           Loads a specific photofinish scenario by name. Usage: make load-scenario scenario=<scenario-name>

Getting started

Run make start to initialize the local environment.

What this command does:

• get photofinish scenarios
• get a local copy of the checks catalog
• get a local copy of the fake facts configuration
• start containers

Note that start keeps the terminal busy, so that when hitting CTRL+C the containers are shut down.

Open other terminal to run further commands.

Hang tight and wait until everything is ready...

When ready browse to http://localhost:4000 and log in with the following credentials: user admin, password adminpassword

Environment variables

It is possible to customize the playground by setting environment variables.

Copy the .env.dist file to .env and edit it as per the specific needs.

$ cp .env.dist .env

See .env.dist for a list of available variables.

Changing the images tags

By default demo images are used for weba nd wanda containers. Edit the .env file to use a different tag.

Example

WEB_IMAGE_TAG=2.1.0 # a specific release tag
# WEB_IMAGE_TAG=1790-env # a PR branch tag ${PR_NUMBER}-env
WANDA_IMAGE_TAG=rolling

Loading photofinish scenarios

Scenarios are loaded with photofinish.

With containers up and ready, run make load-default-scenario to load the default healthy-27-node-SAP-cluster scenario.

Run make load-scenario scenario=<scenario-name> to run any of the available scenarios. ie: make load-scenario scenario=aws-landscape

Take a look at ./data/photofinish/.photofinish.toml for extra information about scenarios and scenarios names.

Catalog

After the setup, the whole community catalog is available for experimentation in ./data/catalog/.

Any changes to the existing checks, newly added checks or removed ones are reflected in the running instance.

Facts Gathering

After the setup, also a configuration file useful to instrument what target agents should return during facts gathering is available at ./data/facts-gathering/fake_facts.yaml.

Any changes to the file are reflected in the running instance.

Please see Wanda demo doc for extra information about the file format.

Alerting

Trento sends email when something worth to be notified happens. See Alerting.

A local mailserver is available at localhost:8025 to inspect sent emails.