IBM Devops Deploy is a powerful deployment tool in the DevOps world. It can deploy to newer tech like containers to age old platforms like mainframe. This write-up will focus on a feature intended for mainframe deployments. The Devops-Deploy tool comes with a rich set of zOS features to version software packages and deploy them. One such feature is the Deploy-Dataset plugin step.
This plugin step is used to deploy/copy datasets and directories in version to target datasets and directories in respective zOS LPAR. A DEPLOY version may contain several datasets packaged in it. For the plugin to deploy each dataset to the right target datasets, it relies on a user input namely ‘Dataset Mapping’.
The Dataset Mapping looks like below:
It is quite evident from the mapping that one can deploy artifacts from one source dataset in the version to one target dataset in the environment.
The new feature enhances Deploy Data Set by allowing a user to deploy one source dataset to multiple dataset using the below mapping
Note that plugin steps can resolve references to properties that can be defined at various levels in Devops Deploy. We encourage our users to store the mapping in a property and refer to them.
When using multiple targets during deployment, this plugin step can now generate information for each target dataset. The simpler way of doing it will be to add filter artifacts at source container filter and leave the target container filter as blank.
Given a mapping like below:
Filter artifacts using container below filter :