Terra Import Service. Tech doc here.
A walkthrough of the code in this repo is available at WALKTHROUGH.md.
Create and activate the Python virtualenvironment:
$ python3 -m venv venv
$ source venv/bin/activate
(venv) $ pip install --user -r requirements.txt
You should periodically run the pip install line within your venv to keep it up-to-date with changes in dependencies.
Activate and deactivate the venv:
$ source venv/bin/activate
<do all your work here>
(venv) $ deactivate
To run tests:
(venv) $ ./pytest.sh
To run the type linter, go to the repo root directory and run:
(venv) $ ./mypy.sh
You should make mypy happy before opening a PR. Note that errors in some modules will be listed twice. This is annoying, but the good news is that you only have to fix them once.
Deployments to non-production and production environments are performed in Jenkins. In order to access Jenkins, you will need to be on the Broad network or logged on to the Broad VPN.
A deployment to dev environment will be automatically triggered every time there is a commit or push to the
develop branch on Github. If you would like to deploy a different
branch or tag to the dev environment, you can do so by following the instructions below, but be aware that a new
deployment of the develop branch will be triggered if anyone commits or pushes to that branch.
- Log in to Jenkins
- Navigate to the import-service-manual-deploy job
- In the left menu, click Build with Parameters
and select the
BRANCH_OR_TAGthat you want to deploy, theTARGETenvironment to which you want to deploy, and enter theSLACK_CHANNELthat you would like to receive notifications of the deploy jobs success/failure - Click the
Buildbutton
When doing a production deployment, each step of the checklist must be performed.
- Create and push a new semver tag for the commit you want to deploy; typically this will be
the head of the develop branch. You should look at the existing tags
to ensure that the tag is incremented properly based on the last released version. Tags should be plain semver numbers
like
1.0.0and should not have any additional prefix likev1.0.0orreleases/1.0.0. Suffixes are permitted so long as they conform to the semver spec.
You must deploy to each tier one-by-one and manually test
in each tier after you deploy to it. Your deployment to a tier should not be considered complete until you have
successfully executed each step of the manual test
on that tier. To deploy the application code, navigate to the import-service-manual-deploy
job and click the "Build with Parameters" link. Select the TAG that you just created during the preparation steps and
the TIER to which you want to deploy:
-
devdeploy job succeeded and manual test passed - (Technically, this same commit is probably already running ondevcourtesy of the automaticdevdeployment job. However, deploying again is an important step because someone else may have triggered adevdeployment and we want to ensure that you understand the deployment process, the deployment tools are working properly, and that everything is working as intended.) -
alphadeploy job succeeded and manual test passed -
stagingdeploy job succeeded and manual test passed -
proddeploy job succeeded and manual test passed - In order to deploy toprod, you must be on the DSP Suitability Roster. You will need to log into the production Jenkins instance and use the "import-service-manual-deploy" job to release the same tag to production.
NOTE:
- It is important that you deploy to all tiers. Because Import Service is an "indie service", we should strive to make sure
that all tiers other than
devare kept in sync and are running the same versions of code. This is essential so that as other DSP services are tested during their release process, they can ensure that their code will work properly with the latest version of Bond running inprod.