Detecting new patches is the first phase. Depending on your RMM tool, you may have the option to detect patches whenever you want based on a custom schedule, or you may be forced into a fixed schedule.
Here are some of the things we typically see our partners do if they have the flexibility to customize it:
- check several times per day (even hourly)
- check daily
- check a few times per week or weekly
- check every month or quarter, whenever they intend to do them
Detection is something you may want to do on a different schedule depending on your needs and depending on devices. If we go back a few years, Windows XP and Windows 7 patch detection was fairly CPU intensive, so patching less frequently was preferable. From Windows 8, patch detection has become much more efficient, and usually not noticeable by the end user—so running it more frequently is not as much of an issue.
My personal recommendation is to detect daily on desktop and twice a week or daily on servers. Some people will say this is overkill. But with Microsoft releasing more and more updates outside the patch Tuesday schedule, a daily check will enable you to detect new patches as they come out. This also allows you to detect patches on devices that are not often online.
Patch approvals
Approaches to approving patches vary widely. Here are some approaches we’ve seen:
- Approve all patches by hand
- Approve security and critical patches automatically and others manually
- Approve all patches automatically
In any of the above scenarios, techs either decline drive updates, approve them, or approve some of them.
Knowing this, it’s tricky to recommend a unique way to move forward. However, most people auto-approve critical and security updates and manually approve several other patches. While lots of MSPs decline drivers, I do not typically recommend this practice as drivers often contain very important security fixes.
If your RMM supports it, you can choose to setup a delayed approval, which is also recommended. It means you can approve patches and install them a minimum of “X” days after discovering them. We recommend delaying approval on most devices. That way you let the early birds get patches on day one and report any bugs, giving Microsoft and third parties the time to pull the patch if it’s not working, and giving you time to manually decline it if desired.
If you have a test group, the test group probably shouldn’t delay approvals so the updates install as quickly as possible.
With manual approval, I often hear people approve them by hand. But upon deeper inspection, the same people simply approve everything—or almost everything—giving a false sense of control over the process. If that is what you do, you may want to consider auto approving some or all of them to save yourself the time of pressing the button to approve them.
I still recommend reading up on the updates each month from an industry expert like our security nerd Gill Langston to ensure you know what updates are coming out and what you may want to delay, install immediately, or decline.
Downloading the patches
Downloading patches is usually done ahead of time. Most RMM platforms support pre-downloading patches at a convenient time, either to the end device or to a central device to minimize bandwidth on the customer’s network.
If your RMM supports it, use a central device (probe) to download the patches once per site. In recent years, some patches have been over 4GB, so caching them centrally can make a big difference on a site with 100 devices.
Next, consider when you want to pre-download. Some RMMs will download as soon as they’re approved, but if your RMM supports scheduling, try to schedule during lower consumption time, like at night—which is usually outside backup windows and peak network usage.
While we’ve covered a lot of information, there’s still more best practices to go over so stay tuned for my next article. We’ll cover installing patches, rebooting the end devices, and monitoring and reporting on patches.
If you’ve created an automation policy and would like to share it with the community, please feel free to email me at [email protected].
As always, don’t forget to go look in the automation cookbook at www.solarwindsmsp.com/cookbook if you’re interested in other automation policies, script checks, and custom services.
Marc-Andre Tanguay is Head Automation Nerd. You can follow him on Twitter at @automation_nerd
