Boot loop

Also known as: bootloop

On this page 7

A boot loop (also written bootloop) is a failure state in which a device restarts repeatedly during startup without ever reaching a stable, usable operating system session. In the study of operating systems, and specifically of Boot & firmware, the term covers any cycle in which one stage of the boot chain fails, triggers a reset, and hands control back to the beginning of the same sequence.

The loop can begin at very different layers: in firmware before the power-on self-test completes, in the bootloader that cannot find or verify a kernel, or inside the kernel and early user space when a driver or service crashes. Diagnosing a boot loop therefore means finding where in the chain the reset happens, because the cure depends entirely on that location.

How a boot loop forms in the startup chain

A boot loop forms when a stage fails in a way that causes a reset instead of a halt with an error message. Modern systems are designed to recover automatically: a watchdog timer reboots a hung controller, Windows restarts after a stop error by default, and Linux can be configured to reboot after a kernel panic. These recovery mechanisms are useful for transient faults, but a persistent fault turns them into an endless cycle.

The typical PC boot chain has five stages, and each one has its own characteristic way of looping. The table below maps the stage to the trigger and the visible symptom, which is the first step of any diagnosis.

Boot stageTypical triggerVisible symptom
Power and POSTUnstable memory, failing power supply, bad overclockFans spin, no logo, restart after seconds
UEFI or BIOS firmwareCorrupted firmware image, invalid settingsLogo appears, then immediate reset
BootloaderMissing boot entry, broken configuration, Secure Boot rejectionBrief boot menu or error, then restart
Kernel initializationFaulty driver, mismatched initramfs, storage controller changeSpinner or kernel messages, then crash
Early user space and servicesCrashing system service, failed update, damaged registry hiveLogin screen never appears, repeated restart
Where in the boot chain a loop can start

A loop that never shows the manufacturer logo usually points to hardware or firmware, while a loop that shows the operating system logo almost always points to software above the bootloader. This simple observation removes half of the possible causes before any tools are used.

Common causes of a boot loop

The most frequent cause on desktop systems is an interrupted or faulty update, followed by driver incompatibility and storage corruption. Hardware faults are less frequent but harder to fix, because no software repair will hold while the underlying component is failing.

CategoryExample causeWhy it loops instead of stopping
Update failurePower loss during a feature updateRollback logic retries the same broken state
Driver faultNew graphics or storage driver crashes at loadAutomatic restart after each stop error
File system damageCorrupted system files or boot volumeKernel cannot mount root and resets
Firmware changeSwitching storage mode from AHCI to RAIDKernel lacks the matching controller driver
Security policyUnsigned bootloader under Secure BootFirmware rejects image and retries boot order
HardwareDefective RAM module or swollen batteryWatchdog or power circuit forces resets
Causes grouped by category

Overclocking deserves special mention. Many motherboards detect repeated failed POST attempts and restore default settings after three to five cycles, which is a deliberate loop designed to end itself. When that safeguard is absent or disabled, an unstable memory profile can keep a system cycling indefinitely.

A boot loop is defined by repetition without user input, which separates it from failures that stop and wait. Precise naming matters because support documentation, logs and recovery tools are organized around these distinctions.

Failure typeBehaviorUsual layer
Boot loopRestarts automatically and repeatedlyAny stage from POST to services
No POSTPowers on but never completes self-testHardware or firmware
Boot hangFreezes at one point with no restartBootloader, kernel or services
No bootable deviceFirmware reports no boot media and stopsBoot order or boot sector
Stop error without restartDisplays crash screen and waitsKernel or driver

One practical consequence is that disabling automatic restart converts a Windows boot loop into a stop error that stays on screen, revealing the stop code and the failing driver name. On Linux, removing the quiet and splash kernel parameters exposes the last messages before the reset in the same way.

Diagnosing and breaking a boot loop

Breaking a boot loop requires interrupting the cycle and entering an environment that does not depend on the failing stage. The safest approach moves from the least invasive action to the most invasive, preserving user data at every step.

  1. Disconnect new peripherals and external drives, then try a cold boot.
  2. Enter firmware setup and load default settings, checking the boot order.
  3. Use the boot menu to select a known good entry or an older kernel.
  4. Start the recovery environment: Windows Recovery Environment, macOS Recovery or a Linux rescue target.
  5. Remove the last update or driver, or run startup repair and file system checks.
  6. Boot from a Live USB to copy data before any reinstallation.

Windows enters its recovery environment automatically after two or three consecutive failed boots, so a forced power-off during the logo can also be used intentionally. The decision table below summarizes the recommended first action for common situations.

SituationRecommended first actionReason
Loop began after an updateUninstall the latest quality or feature updateRestores the last known working state
Loop began after new hardwareRemove the hardware and retestIsolates driver or power conflict
Loop after firmware setting changeReset firmware to defaultsReverts storage mode or overclock
Loop with no logo visibleTest memory and power supplyPoints to hardware below firmware
Loop with Secure Boot messageCheck signed bootloader and key databaseFirmware is rejecting the boot image
Decision table for the first repair action

Preventing boot loops and common mistakes

Prevention relies on redundancy in the boot chain. Many devices now use A/B system partitions, keeping a second copy of the operating system that the bootloader selects when the active copy fails a set number of times. Desktop users achieve a similar effect with restore points, system snapshots and full image backups.

  • Never interrupt a firmware update or operating system update in progress.
  • Keep a recovery drive or Live USB prepared before major changes.
  • Avoid changing storage controller mode without preparing the drivers first.
  • Do not assume a reinstall will help when hardware is the root cause.
  • Record stop codes and kernel messages instead of repeatedly power cycling.

A frequent mistake is reinstalling the operating system too early. Reinstallation destroys diagnostic evidence and often fails to fix loops caused by firmware settings or faulty memory, so the repair should always follow the location of the failure.

In plain words

A boot loop is when a computer keeps restarting while it is trying to start up. It shows a logo, maybe spins for a moment, and then begins again from scratch, over and over, without ever getting to the desktop.

Think of a relay race where one runner keeps dropping the baton, and the rules send everyone back to the starting line each time. The trick is to find which runner drops the baton. Once that step is known, it can be fixed or skipped, and the race can finish normally.

A real-world example

An IT technician at a small accounting office receives a laptop that restarts every few seconds after showing the Windows logo. The owner explains that a storage driver update was installed the previous evening. Because the logo appears, the technician rules out the power supply and memory and focuses on software above the bootloader.

After the third failed start, the laptop opens the Windows Recovery Environment. The technician selects startup settings, chooses to disable automatic restart, and the next attempt shows a stop code naming the storage driver. Back in recovery, the technician uninstalls the latest update and restarts.

The laptop reaches the desktop in under two minutes, with no data lost and no reinstallation. The technician then pauses driver updates for two weeks until the vendor releases a corrected version.