Patched but Still Vulnerable: Restart and Reboot Discipline for Linux Fleets
  • PDF
  • 11 pages
  • 2.0 MB
  • November 2026
White Paper · Linux Operations

Patched but Still Vulnerable: Restart and Reboot Discipline for Linux Fleets

Restart and Reboot Discipline for Linux Fleets

Installing a security fix does not protect a Linux system until running processes restart or the host reboots. This white paper explains why the gap exists, how to detect it with tools such as needs-restarting and needrestart, where live patching fits, and how to run restarts as a routine with evidence. It is written for IT leaders, architects and engineers responsible for Linux fleets.

What you will learn

  • Why updated libraries, runtimes and kernels stay vulnerable in memory until a restart or reboot
  • What CVE-2026-98432 and CVE-2026-111497 showed about the gap between installed and running fixes
  • How to find pending restarts and reboots with dnf needs-restarting, needrestart, zypper ps and /proc
  • What kpatch, Canonical Livepatch, Oracle Ksplice and SUSE Live Patching cover, and their limits
  • A six-step operating model, from install to recorded evidence, for each class of host
  • Automatic updates, container image rebuilds and rolling Kubernetes node reboots with kured

Download the white paper

Free PDF. Fill in the form and the download starts straight away.

    Last Name

    First Name

    Company Name

    Email Address

    Phone Number

    Address

    Contents

    1. Two vulnerabilities, one lesson
    2. Why installed is not running
    3. Finding what still runs old code
    4. Live patching: where it fits
    5. An operating model for restarts
    6. Implementation considerations
    7. Practical use cases and risks
    8. Governance, metrics and cost
    9. Conclusions and references

    Questions about your environment?

    LOGIQ plans and runs Linux, virtualization and Java projects for enterprise customers.

    Contact LOGIQ