> Source: [sk183889](https://support.checkpoint.com/results/sk/sk183889)

# sk183889 - CPView - "I/O - "Overview" - section "Average" shows unusually high values for the devices "md0", "md1", "md2"

| Property | Value |
|----------|-------|
| Solution ID | sk183889 |
| Date Created | 2025-09-08 |
| Last Modified | 2025-09-19 |
| Technical Level | Advanced |
| Products | Hardware |
| Versions | Not Version-Specific |
| OS | Gaia |

## Symptoms

- * CPView - "`I/O`" section - "`Overview`" page - the section "`Average`" shows unusually high values in the columns "`Queue Size`", "`Wait`", and "`Service`" for the Devices called "`md0`", "`md1`", "`md2`".

  Example:

  ```
  
  ... ...
  | Average:
  |
  | Device  Request Size    Queue Size      Wait                     Service
  | sda     2.33            0.00            0.04                     0.04
  | sdb     2.33            0.00            0.04                     0.04
  | md0     0.00            1,091.00        2,777,686.00             2,546.00
  | md1     0.00            642.00          1,634,532.00             2,546.00
  | md2     0.00            1,300,587.50    204,963,714,033,355.90   283.11
  | dm-0    0.00            0.00            0.08    0.16
  ... ...
  ```

* The overall CPU utilization does not increase - the CPU remains mostly idle.

* There is no functional impact or performance degradation.

* The Hardware Diagnostic Tests (HDT) do not report any issues.

## Cause

The high CPU utilization values shown in CPView for `/dev/md<X>` devices are a cosmetic issue.

/dev/mdX devices are a software RAID device managed by the mdadm subsystem. Unlike a physical disk, it is a virtual block device that aggregates multiple underlying physical disks into a single logical unit. Because of this, it doesn't handle I/O itself but instead passes I/O operations to its member disks.  

If any of those physical disks are slow, overloaded, or undergoing a resync, I/O on the RAID device may appear to be constantly pending, resulting in 100% utilization. Additionally, software RAID can issue parallel I/Os across the member disks. When those disks are busy, the entire RAID layer (/dev/md0) appears fully utilized. Therefore, high utilization on /dev/mdX simply indicates that the RAID software had continuous I/O work to process during the sample period---it does not necessarily mean the system's CPU was fully consumed or that there is a single hardware bottleneck.  

This does not indicate actual CPU exhaustion or hardware bottlenecks. This is an expected behavior and not a malfunction. The calculation in CPView may mix units or misinterpret I/O statistics for these virtual devices, but there is no impact on system performance or RAID health.

## Solution

This solution requires authentication. Please log in to view the full solution.

---

# Agent Instructions

This content is from the Check Point Support Center (https://support.checkpoint.com), the official knowledge base for Check Point cybersecurity products.

## Navigating This Knowledge Base

- **Complete index**: [llms.txt](https://support.checkpoint.com/llms.txt)
- **All SK articles**: [SecureKnowledge Sitemap](https://support.checkpoint.com/sitemaps/secureknowledge-sitemap-index.xml)
- **SK article URL pattern**: `https://support.checkpoint.com/results/sk/{skId}`
- **Markdown responses**: AI bot User-Agents automatically receive `text/markdown` content