This playbook demonstrates how to grow a ZFS aggregate on z/OS using Red Hat Ansible Certified Content for IBM Z.
This playbook uses:
- collection:
- ibm.ibm_zos_core
- modules:
- zos_data_set
It is a good practice to review the playbook contents before executing them. It will help you understand the requirements in terms of space, location, names, authority, and the artifacts that will be created and cleaned up.
This playbook requires:
- Configure the included inventory.yml with the
information from the managed z/OS host.
- Review inventory documentation
- Configure the included host_vars zos_host.yml
with the information from your z/OS system.
- Review host_vars documentation and any additional noted variables in the configuration.
This project has included a site.yml
playbook that serves as the primary playbook
that provides additional prerequisite checks then it invokes the grow_zfs_aggregate.yml
playbook.
If you want to run the primary playbook site.yml
it will check that your environment
has the correct version of Ansible as well as the collection needed to execute
correctly. To run the primary playbook, use command:
ansible-playbook -i inventories site.yml
You can skip the prerequisite check and run the grow_zfs_aggregate.yml
with
command:
ansible-playbook -i inventories grow_zfs_aggregate.yml
All changes are maintained chronologically by date found in the changelog.
© Copyright IBM Corporation 2022
Licensed under Apache License, Version 2.0.
Please refer to the support section for more details.