How to Build an Internal Patch Management Workflow with Open-Source Tools: Keeping Montana Business Systems Patched Without a Subscription

A Montana small-business office at dusk with servers and workstations and a patch-management dashboard, dark tech style with mountain silhouettes

A repeatable patch workflow helps you find exposed systems, prioritize fixes, deploy updates in controlled stages, and confirm that vulnerabilities are resolved. This guide outlines how to build one with open-source, on-premises tools and native operating-system utilities, without paying a per-device fee for a patch-management platform.

Unpatched software is one of the common ways attackers gain initial access, particularly when vulnerable services are exposed to the internet. CISA’s StopRansomware Guide recommends regular updates and prioritizing internet-facing systems and known exploited vulnerabilities. A maintenance-first cadence reduces the time a vulnerability remains available to attackers, while staged deployments help limit update-related downtime.

A vulnerability remediation workflow from asset inventory through scanning, staged patching, and verification

1. Establish an accurate patch inventory

Start with a spreadsheet or self-hosted inventory database. Record each device’s hostname, owner, operating system and version, purpose, internet exposure, business criticality, maintenance window, reboot tolerance, backup status, and assigned patch group. Include servers, workstations, network appliances, and business applications, not just devices that staff use every day.

For ongoing inventory and vulnerability correlation, Wazuh collects information about operating systems and installed packages through endpoint agents. Its vulnerability-detection feature compares inventory with vulnerability data and can reflect certain Windows hotfixes. Use it to spot findings and track endpoint status; it does not install patches.

Add an independent scan with Greenbone Community Edition, which includes the OpenVAS scanner. Place the scanner on a restricted management network and use authenticated scans when appropriate. Authenticated scans can see more than a basic network probe, but they require carefully scoped credentials. Scanner results also need review: a finding is not always proof that a particular package is vulnerable, and a clean scan alone is not proof that a patch installed correctly.

2. Rank findings by risk and define patch groups

Do not sort work by severity score alone. Prioritize whether the system is internet-facing, whether the vulnerability appears in CISA’s Known Exploited Vulnerabilities Catalog, the asset’s business impact, and whether the affected service is reachable. An actively exploited issue on a public-facing server belongs in an emergency lane; routine updates can follow the regular maintenance schedule.

Create three groups: a small canary group for early testing, a broad group for most devices, and a sensitive group for systems with legacy applications or tight uptime requirements. Define a target for each lane, who can approve exceptions, and when exceptions expire. If a vulnerable service cannot be patched immediately, document temporary controls, such as restricting network access, and set a date to revisit the risk.

3. Automate Linux updates, but control reboots

On Ubuntu and Debian, unattended-upgrades installs updates from configured repositories. Ubuntu documents that it runs daily by default and records activity under /var/log/unattended-upgrades. Review /etc/apt/apt.conf.d/50unattended-upgrades to confirm trusted origins and any package exclusions, then check /etc/apt/apt.conf.d/20auto-upgrades to verify that package-list refresh and unattended upgrades are enabled. Test configuration changes with:

sudo unattended-upgrade --dry-run --verbose

For business-critical servers, avoid enabling unattended reboots without an explicit plan. The tool can be configured to reboot when required, but a surprise restart can interrupt staff or stop an application. Instead, alert on pending reboots, schedule them within an approved window, and confirm services return afterward.

On DNF-based systems, such as Fedora or supported Enterprise Linux distributions, configure /etc/dnf/automatic.conf. Set upgrade_type = security to limit automatic updates to packages with security advisories. Choose whether the host only notifies, downloads, or installs updates; then enable the matching systemd timer. For example:

sudo systemctl enable --now dnf-automatic.timer

Use Ansible to apply consistent settings, update a canary group, collect results, and coordinate planned reboots across Linux systems. Keep deployment playbooks under version control so changes can be reviewed and reversed. Automation reduces repetitive work, but it does not replace testing applications that depend on specific libraries or kernel versions.

Linux and Windows systems receiving updates in controlled pilot and production groups

4. Choose a Windows update source and keep it consistent

For Windows 10 and 11 devices that can use Microsoft’s update service, Windows Update client policies, often called Windows Update for Business policies, can set update timing and deployment behavior through Group Policy or local policy. Use a pilot group before wider deployment. These policies do not create a local update repository: devices still obtain update content from Microsoft.

If local content distribution or centralized approvals are required, Windows Server Update Services (WSUS) remains an option. Microsoft says WSUS is deprecated, meaning it is no longer receiving new features, but it continues to be supported for production deployments and receives security and quality updates under its product lifecycle. Configure clients with Group Policy, set a clear pilot group, and approve updates for broader groups only after testing. Avoid overlapping policies that send the same devices to different update sources.

For controlled execution, Ansible’s ansible.windows.win_updates module searches for and installs Windows updates through the Windows Update Agent. Ansible connects to Windows through PowerShell Remoting, commonly using WinRM. A basic task for a pilot group looks like this:

- name: Apply selected updates to Windows pilot systems
  ansible.windows.win_updates:
    category_names:
      - CriticalUpdates
      - SecurityUpdates
      - UpdateRollups
    reboot: true

Run it only during an approved window, with a tested connection and a plan for hosts that fail to return after a restart. Choose one authoritative update policy and use Ansible for coordinated execution and evidence, rather than letting multiple tools compete to control update source and restart behavior.

5. Include macOS without mixing up update tools

For Mac fleets, Apple’s softwareupdate command-line utility can list and install available system updates. A separate device-management tool may be needed to enforce deadlines and update behavior across enrolled Macs. Munki is an open-source option for managing application packages and staged software catalogs; it complements rather than replaces Apple’s native operating-system update mechanisms. Test the exact macOS versions and update policies in use before automating broad deployment.

6. Use a repeatable change, rollback, and verification cycle

For every patch group, follow the same sequence:

  1. Prepare: Create a change record with affected devices, update identifiers, owner, maintenance window, backup status, and rollback steps. Check that backups are recent and that recovery has been tested.
  2. Test: Patch a representative canary device. Confirm login, business applications, backups, monitoring, and network connectivity work as expected.
  3. Deploy: Expand in waves. For clustered or capacity-sensitive systems, patch one node at a time and confirm capacity before continuing.
  4. Recover if needed: Stop the rollout if a canary fails. Use the documented restore or vendor-supported recovery procedure; do not assume every update can be cleanly uninstalled.
  5. Verify: Confirm installed package versions or Windows update history, check service health, and rescan the affected systems. Keep the scan result and patch logs with the change record.

A maintenance window, backup, update, and restore plan for business systems

7. Track coverage, cost, and practical limits

Review a concise report each week: percentage of known assets reporting in, systems overdue for updates, unresolved exploited vulnerabilities, failed installations, pending reboots, and exceptions nearing expiry. An endpoint missing from the inventory is not a patched endpoint. Compare scanner findings with agent inventory and operating-system update logs before marking remediation complete.

Open-source tools can avoid per-device platform subscriptions, but they are not cost-free to operate. Budget for a management server, storage, backups, staff time, feed maintenance, testing, and secure administration. Some products also offer paid editions or support. Test legacy applications carefully: compatibility checks, maintenance windows, reboots, and staff disruption all add overhead, but skipping them can leave a known vulnerability exposed. These practices apply whether a Montana company runs updates internally or evaluates managed IT services in Montana, IT support in Great Falls, IT consulting in Montana, or IT outsourcing in Montana. Strong patch records also contribute to network security and data protection in Montana organizations, alongside backups, access controls, and other safeguards.

For background on infrastructure tradeoffs, see this guide to cloud versus in-house IT infrastructure. Subscription-free patching works best when the process, not any single tool, stays documented, measured, and consistently followed.

Call us at 406-866-0128 or visit https://itgreatfalls.com.