How does it work?
On a high level, your
lando configuration should contain ALL the things you need to run, develop and test your project. The ideal we strive for is best expressed as follows:
A user should be able to
git clonea repository, run
lando startand get EVERYTHING they need to develop their site locally in a few minutes.
Implicit in the above is that your lando configuration should always be checked into version control at the root directory of your project.
We've found a useful pattern here is for the senior-est, most DevOps-y person on your team to craft the initial lando configuration and be responsible for its maintenance but to generally "set it and forget it".
Ok cool, I dig it. What do I need to get started?
In order for you to get started using the awesome power of Lando, you need to have the following:
- A Landofile, generally this is called
- Your application's codebase
If you are unclear on how to manually set this up, check out our lando init command which can help you initialize code from various sources for use with Lando.
So, what does this Landofile look like anyway?
As mentioned above, the Landofile is generally named
.lando.yml but is both configurable and overridable. It needs a
name, but beyond that, it can contain any combination things, although generally it only uses a small subset as follows:
Note that the values in each key above are not particularly relevant to what we are discussing which is the purpose of each. To that end, some information about each and links to their docs, if applicable, is shown below:
name - Nothing special here. This should be a unique way for you to identify your app.
compose - This is a list of
docker-compose files you can tell Lando to start. The paths are relative to the root of your project. Generally, this option is not used in favor of Lando recipes and/or services but is an option that makes sense under some circumstances.
Note that services located in these files will still get injected with Lando's default environment variables and also be networked like any other Lando service but will not get any of the extra Lando secret sauce.
If you are interested in using something Lando does not currently offer as a service or recipe, we recommend looking at the custom compose service.
env_file - A list of environment files relative to your project's root directory. Each file should contain a list of
KEY=VALUE pairs that will then get injected into the environment of every service.
events - Events allow the user to run arbitrary commands, or combinations of commands, on arbitrary services, or combinations of services before or after certain parts of the Lando runtime. A good example is clearing out an application's cache after a database is imported.
config - Config allows you to set some of the more important things your recipe provides. These settings are usually different depending on the recipe you select.
proxy - Proxy settings allow users to map arbitrary domain names to arbitrary ports inside of arbitrary services. Think: I go to the browser and type in
millard.filmore.for.lyfe and it loads my application.
services - Services are simplified but still highly configurable Docker containers. They are able to run build steps such as installing a
php-extension or running
yarn install and can also be overridden down to the
docker-compose level. They also get some automatic networking and security features.
tooling - Tooling allows users to run arbitrary commands, or combinations of commands, on arbitrary services, or combinations of services when invoked.
This is helpful so you can run
lando yarn install instead of
docker exec -it SERVICE yarn install or so
lando test can run
composer test and
yarn test in their respective services. You can also specify options, including interactive ones, to build out more complicated commands like
lando pull-my-database-from-aws --user=me --password=*** or to have a single command run on many services - e.g.
lando db-import dump.sql -h database2.
A LATE TL;DR - We HIGHLY recommend you read through the linked documentation above to get a sense of the things the Landofile can do for you and/or your team.
How do I make one?
There are various ways to craft your Landofile but we've found and observed a strategy to be generally a good approach as follows:
- Shop around for a recipe that can serve as a good starting point. lando init is helpful for this.
- Add in additional services when you need more JUICE.
- Define more complex tooling and events to simplify difficult command patterns and automate common tasks.
- Add in some build steps to further automate setting your services up or to mix in additional project dependencies.
- Define custom services as a catch all for anything else you need.
- Create custom recipes or services to lock down your new power tools.
- Rinse and repeat.
You have some examples?
You can check out our large repository of tested-every-build and working examples. A few example Landofiles to help give you a sense of how simple or complex they can be is shown below:
Cold as ice
Lighting the match
He's heating up!
He's on fire!!
- appserver: drush cc all -y
- apt update -y && apt-get install vim -y
- /helpers/my-script-to-install-php-extension.sh memcached
- composer install
command: yarn start
- appserver: composer test
- frontend: yarn test