Estimated Reading Time: 2 mins
Meteor’s default installer requires sudo on Linux. We use a script to change install location and make sudo unnecessary:
\curl -sSL https://raw.githubusercontent.com/codeship/scripts/master/packages/meteor.sh | bash -s
This setup script will always pull for the most recent meteor tool. You can also call meteor commands with a specified release:
meteor create test --release 0.6.1
You can use npm or yarn to install your dependencies.
We set the
$PATH to include the
node_modules/.bin folder so all executables installed through npm can be run.
We also automatically cache the
$REPO_ROOT/node_modules directory between builds to optimize build performance. You can read this article to learn more.
You can also use Yarn to install your dependencies as an alternative to npm. Yarn is pre-installed on the build VMs. We configure
yarn to write into
$HOME/cache/yarn, which is also cached.
meteor npm install -g yarn meteor yarn
Note that we do not officially support or integrate with this module and that it is possible for this to cause resource and build failure issues, as well.
meteor deploy METEOR_APP_URL
Add the following environment variables to your project configuration:
And include the following command to your deployment pipeline:
\curl -sSL https://raw.githubusercontent.com/codeship/scripts/master/deployments/meteor.sh | bash -s
Sometimes Meteor commands will run into Node’s default memory limit resulting in this error:
FATAL ERROR: CALL_AND_RETRY_LAST Allocation failed - process out of memory Aborted (core dumped)
If this happens in your build you can increase the memory limit by setting this environment variable for the project:
We also have a couple of code examples and sample projects available, that make it easier to get started with Codeship.