Cloudharness allows to run workflows through Argo workflows providing the Argo installation and a Python library to run workflows programmatically.
The Argo-ui is available on the argo
subdomain in your deployment. Can use the Argo-ui to monitor and inspect workflows.
By default the ui is protected by a gatekeeper, so an admin user credentials are needed to use the application. To disable the gatekeepers use the parameter -u
when running harness-deployment
.
The workflow micro service application defines some common use cases tasks and a REST api to monitor workflows. It is accessible at the address workflows.MYDOMAIN when it is included in the deployment.
Tasks can be created below an application is the tasks
directory.
Every Dockerfile below the tasks
subdirectory is recognized as an image to build aside the containing application.
For instance, given this folder structure:
applications
myapp
tasks
t1
Dockerfile
t2
Dockerfile
The two artifacts myapp-t1
and myapp-t2
are built and available to be used as custom tasks inside workflow.
See the Python api to know how to create and run a workflow.