Skip to content
PATCH MANAGEMENT

Installed is not done. Done is restarted and healthy.

An update that waits for a restart protects nobody. Secuno follows every install through the restart, checks how the device came back up, and calls it verified only when the vulnerabilities it fixes are closed. Windows and Linux.

  1. Approved
  2. Preflight
  3. Installed
  4. Restarted
  5. Healthy
  6. Verified
REMEDIATION PLAN

Start with the fixes that remove the most risk.

A client with hundreds of missing updates does not need a longer list. The plan puts the fixes in the order that removes the most of that client's open risk, and tells you how many actions reach 80%.

How the plan is calculated
REMEDIATION PLAN · ONE CLIENTEXAMPLE
  1. Windows cumulative update41 devices
    38%
  2. Browser update through winget63 devices
    57%
  3. Linux package from the distribution6 servers
    68%
  4. Remove an app nobody used in 90 days12 devices
    76%
  5. Vendor workaround, no patch exists yet4 devices
    81%

80% of open risk removed in five actions

  1. Each row counts only what is new

    A row is credited with the risk no row above it already removes, so the running total is real.

  2. Verified means the CVEs closed

    An install is not the result. The plan checks afterwards that the vulnerabilities behind it are gone.

  3. A workaround when there is no patch

    Twelve built-in vendor workarounds cover the gap until the fix ships, and count only while they hold.

THE RESTART LADDER

Five ways to restart, from never to forced.

Every device that needs a restart gets one plan in one of five modes. Updates that arrive while a plan is open join it, so a device restarts once per set of installs, not once per update.

  • Quiet hours from 22:00 to 07:00 hold in every mode, in the device's own time
  • The prompt carries the client's own logo, colour and language
  • Every pending restart is on one screen
AFTER THE RESTART

Did it come back healthy?

The agent reports every time it starts, with evidence of how the machine came up. A device that does not return is a finding, and no evidence is never read as good news.

  1. Forced

    EMERGENCY LANE

    The restart follows a countdown the user cannot dismiss.

    Emergency-lane installs, which still allow one 15-minute postponement.

  2. Scheduled

    The restart happens at a set time.

    Shared devices and estates run on fixed windows. The kiosk profile restarts nightly at 03:00.

  3. Deferrable

    WORKSTATION DEFAULT

    The user may postpone within a budget. At the deadline it restarts anyway.

    The default for workstations: four postponements, a 72-hour deadline and a 30-minute final countdown.

  4. Notify only

    The user is told a restart is needed. Nothing restarts.

    For machines you would rather not touch, while the pending restart stays visible.

  5. Never

    SERVER DEFAULT

    Secuno does not restart the device.

    The default for servers. They restart when you run it, or in order through a restart group.

LANES AND SLA CLOCKS

Urgency set by exploitation, not by a severity label.

Every Windows update, app update and Linux package sits in one of four lanes, set by the CVEs it fixes. Each lane starts a clock at detection.

24hours
from detection

Emergency

Exploited against this client, listed in CISA's Known Exploited Vulnerabilities catalogue, an EPSS score of 0.5 or higher, or a CVE in an open incident.

72hours
from detection

Accelerated

CVSS 9 or higher, EPSS 0.1 or higher, missing on an internet-facing device, or rated Critical by the vendor on a business-critical asset.

14days
from detection

Standard

Every other security update.

60days
from detection

Deferred

Drivers, feature, definition and tool updates, and anything declined or rolled back. Decided first: a driver with an exploited CVE is still a driver.

DEPLOYMENT POLICY

One policy per client, readable as four sentences.

Delivery, reboot, preflight and notification. Each clause is a sentence built from its own values, so an account manager can read a client's policy back to the client without translating settings.

DEPLOYMENT POLICY · WORKSTATIONSDEFAULTS
Delivery
Approved updates install inside the client's maintenance window, in batches sized to fit the time it has left.
Reboot
A user may postpone a restart four times. After 72 hours the device restarts anyway, with a 30-minute final countdown, and never between 22:00 and 07:00 device time.
Preflight
Nothing installs when disk space or power is short, on a metered connection, while someone presents, with another restart pending, or in the first 10 minutes after boot.
Notification
Users hear about it 15 minutes before the first batch, and the restart prompt carries the client's own logo, colour and language.
SIX CHECKS BEFORE EVERY BATCH
  • Disk space
  • Power
  • Metered connection
  • User presenting or in fullscreen
  • Restart already pending from another tool
  • First 10 minutes after boot

When one fails, the work goes back to the queue with its attempt refunded, and the reason is on screen.

A hold older than 72 hours becomes a high-severity risk, so a stuck device surfaces on its own.

QUESTIONS

The things worth asking first.

Will users be forced to restart in the middle of a meeting?

Not by default. Workstations get a deferrable restart: four postponements, a 72-hour deadline and a 30-minute final countdown, with quiet hours from 22:00 to 07:00. Someone presenting or in fullscreen holds the install before it even starts. The only forced restarts are Emergency-lane installs, which still offer one 15-minute postponement, and a Restart now that an operator chooses to send, which gives five minutes' warning.

What happens to servers?

Servers never restart on their own by default. Updates install and wait, and the restart happens when you run it or through a restart group, which restarts a set number of servers at a time, in order, with a health check between each. With the backup check on, a server's installs and restarts are held until a recent backup is proven, read from VSS, Windows Server Backup or Veeam on the server, or from Veeam Backup & Replication, Datto BCDR or Azure Backup through an integration. If a server does not report back after a restart, judged against the hours it is normally online, a high-severity finding is raised.

Can we roll back?

For apps, yes, across the fleet: roll a release back for a ring, a client or a device group, and each device returns to its previous version and is pinned, so the next update does not undo it. Linux packages roll back on apt, dnf, yum and zypper. A halted rollout ring rolls back in one click. For Windows updates, only what Windows itself allows to be removed, and the agent reports which updates those are. A rolled-back patch cannot be queued again until someone reverses the rollback.

Where do we start on a client with hundreds of missing updates?

With the Remediation Plan. It lists the fixes that remove the most of that client's open risk, in order, and says how many actions reach 80%. Each row counts only what the rows above it do not already remove, so the running total is honest. A fix only counts as Verified once a later inventory shows its CVEs closed.

06:00 · DAYBREAK

Installed, restarted, healthy. Before the first coffee.

We will take one client's patch backlog and show the plan, the policy and the restarts it would have run overnight.