Windows
Patching
Patching
Apply Windows updates by category to Windows Server hosts and generate an HTML patch report. The playbook uses the demo.patching.patch_windows role to install updates filtered by category (Security, Critical, Feature Packs, etc.), with optional reboot control. A report server is deployed automatically to publish patching results. Runs in check mode by default.
Prerequisites
- Windows hosts in the Ansible Product Demos Inventory (deployed by Deploy Cloud Stack in AWS)
- WinRM connectivity via APD Machine Credential
- A Windows report server (
aws_win1) in theos_windowsinventory group
Survey prompts
| Prompt | Variable | Type | Required |
|---|---|---|---|
| Server Name or Pattern | _hosts |
text | No |
| Update categories | win_update_categories |
multiselect | No |
| Reboot after install? | allow_reboot |
multiplechoice | No |
Job templates
| Template | Playbook | Description |
|---|---|---|
| WINDOWS ǀ Patching | windows/patching.yml |
Apply Windows updates by category, optionally reboot, and publish an HTML patch report |
Why it matters
- Windows patching is a top customer pain point — this demo proves AAP handles it natively
- Category-based filtering (SecurityUpdates, CriticalUpdates, etc.) shows precision control over what gets installed
- Check mode lets you preview updates before applying — critical for change management approval
- The HTML report provides audit evidence without relying on WSUS or SCCM reporting
- Reboot control via survey shows operational safety for production Windows servers
Presenter walkthrough
- Show the survey: Walk through the category selector. ‘We can target just security updates, or cast a wider net. The operator chooses — not the tool.’
- Launch in check mode: Run against the Windows hosts. ‘Check mode queries Windows Update Agent without installing anything. We see exactly what would change.’
- Review the output: Show which KBs are applicable per host. Point out the reboot-required indicators.
- Switch to run mode: Re-launch with job type set to Run. Highlight the reboot control option. ‘In production, you might patch during a maintenance window and defer reboots.’
- Show the report: Navigate to the report server and walk through the HTML patching report.
- Connect to the bigger picture: ‘For environments with both Windows and RHEL, our Patch Cloud Stack workflow handles both in parallel with EBS snapshot rollback.’
Talking points
- Ansible manages Windows natively through WinRM — no agent installation required on the target hosts.
- Category filtering means you can apply just security updates during an emergency patch cycle without pulling in feature packs.
- Check mode is your dry run. Show the change board exactly what will happen before you touch production.
- The reboot survey option is a small detail that matters in production — operators control when reboots happen, not the automation.
- This same patching pattern works on-prem, in AWS, or in Azure. The playbook does not care where the Windows host lives.