Skip to content

Latest commit

 

History

History
58 lines (38 loc) · 3.53 KB

DEPLOY.md

File metadata and controls

58 lines (38 loc) · 3.53 KB
Raw

We deploy this application on Cloud Foundry using continuous deployment. The following branches map to Cloud Foundry applications

git branch CF organization CF space CF application public URL notes
dev open-opportunities dev openopps https://openopps-test.18f.gov A public testing service for the latest development code
staging open-opportunities staging openopps https://openopps-staging.18f.gov A staging server for reviewing releases
master open-opportunities master openopps https://openopps.digitalgov.gov The production instance of Open Opportunities (for federal employee use only)

Deployment Process

All changes to the application should follow the GitHub flow process outlined in the CONTRIBUTING.md document, which consists of making branches for changes and submitting them as pull requests against the dev branch.

PRs that are merged into dev will go through Travis for continuous integration testing and deployment.

Staging

New releases destained for production should be tagged following semver and pushed submitted as a pull request against the staging branch.

PRs that are merged into staging will go through Travis for continuous integration testing and deployment.

Deploy to production

Once a release has been validated on staging, open a new PR for the release against master.

PRs that are merged into master will go through Travis for continuous integration testing and deployment.

Deploying to Cloud Foundry with a migration

If you have a migration that needs to be run, you'll need a few things:

  1. cf-ssh installed on your machine. Read the documentation here
  2. "Access a postgres database using cf-ssh" which can be found on the cloud.gov documentation
  3. Patience. Cloud Foundary will create new instances of the application within the specific space and these builds and connections may take a while to complete.

Using cf-ssh, access the production instance of Open Opportunities. Make sure you're on the proper branch on your local environment. This is because cf-ssh uses the local files to build the new instance. Please note that this might take a few minutes. Once you're into the production box follow the instructions on the cloud.gov documentation for installing psql. We need psql because many of the migration scripts we have in /tools/postgres/migrate are bash scripts that run queries through psql.

Now you can simply run ./tools/postgres/migrate/yourscript.sh and the migration should be in place.

If you have a data migration script (generally the .js scripts in /tools/postgres, you can run those from here too with node /path/to/migration-script. If your script doesnt disable HTTP, you need to set the port to something like PORT=9999 node path/to/mirgration-script, otherwise it may conflict with the port already running.

Cloud Foundry configuration

Cloud Foundry applications read their settings from env vars set in manifest.yml files and user-provided services. Each CF space has a unique env-openopps user-provided service for storing sensitive env vars. If you have access to deploy to CF, you should have access to this service too.

To change values stored in the env-openopps service, see the Cloud Foundry docs.