> Source: [sk183525](https://support.checkpoint.com/results/sk/sk183525)

# sk183525 - High CPU usage on one SND core

| Property | Value |
|----------|-------|
| Solution ID | sk183525 |
| Date Created | 2025-06-03 |
| Last Modified | 2025-12-31 |
| Technical Level | Advanced |
| Products | Security Gateway |
| Versions | R82, R81.20, R81.10 (EOS) |
| OS | Gaia |

## Symptoms

- * On Security Gateway with Multi-Queue enabled there is high CPU usage on one SND core while other SND cores are underutilized.

* In CPVIEW output (CPVIEW \> CPU \> Overview) you can see that system interrupts appear as the top CPU consumers, which indicates high traffic on the SND.

* CPU Spike Detective logs (see [sk166454](https://support.checkpoint.com/results/sk/sk166454)) shows long-duration spikes on the affected core, for example:  

  ```
  
  spike info: type: cpu, cpu core: 0, top consumer: system interrupts, start time: [DATE] [TIME] , 
  spike duration (sec): 4262, initial cpu usage: 81, average cpu usage: 99, perf taken: 1  
  ```

  <br />

* The `"mq_mng -ov"` command output indicates large RX packet counts on the affected SND core.

* The output of the `"ethtool -i `" command shows:  

  ```
  driver: ice
  ```

  <br />

## Cause

The *ice* NIC driver does not effectively distribute ESP (Encapsulating Security Payload) packets across multiple CPU cores. Instead, it maps them to a default queue, which overloads the corresponding core.

## 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