Background

Update operations, such as tagging virtual machines, require that state be passed as part of the operation. If batch automations are required, then that means the automator has to find a way to split automation between resources with different states so that the state does not change when tags are applied (i.e. starting a stopped VM or vice versa). Metadata changes should not require the state of the resource; this would greatly improve automated management of fleets of infrastructure.

User Stories

As an Ansible user attempting to automate Azure Virtual Machines I am able to perform metadata updates to the infrastructure records without passing the state of the resource at the same time. This will allow me to make changes to those resources without fear of making an involuntary state change, such as turning a VM on or off, when simply updating tags or other metadata that does not require the VM state to accompany the change.

Supporting documentation

<include links to technical docs, diagrams, etc>

Ready-Ready

https://docs.google.com/document/d/1dBW492UfCR5JxUvcx6efjYP_ibCdiwGONS-sxDb_SPU/edit# 

Complete Ready-Ready checklist

Definition of Done

Should be agreed upon per team; add/remove/update to reflect:{}

• CI is running, tests are automated and merged and successful
• DEV upstream code & tests merged
• DEV upstream documentation merged
• DEV downstream build attached to advisory
• QE - Test plans documented and attached to epic (or link to source)
• QE - automated tests merged and passing
• Docs - Downstream documentation is merged
• PM - all acceptance criteria are met (note: we want to start using the AC specific field, but please fill out below in the meantime)

Note: please also fill out the 'Acceptance Criteria' to indicate functional criteria aka scope, specific to this issue, that should be met in order to 'accept' the work to consider complete.