> Source: [sk106614](https://support.checkpoint.com/results/sk/sk106614)

# sk106614 - Cluster constantly fails over due to incorrect SIM Affinity configuration

| Property | Value |
|----------|-------|
| Solution ID | sk106614 |
| Date Created | 2015-06-28 |
| Last Modified | 2021-01-20 |
| Technical Level | Advanced |
| Products | Security Gateway |
| Versions | R82.10, R82, R81.20 |
| OS | Gaia |

## Symptoms

- * SmartView Tracker logs show that all cluster interfaces are constantly flapping one after another.

* Output of "watch -d -n 1 cphaprob state" command on each member shows that state of members constantly changes between "Active" and "Down".

* Output of "cphaprob list" command shows that Critical Device "Active Interface Check" reports its status as "problem".

* Output of "top" command repeatedly shows spike in CPU utilization by different CoreXL FW instances ("fw_worker" processes).

* Output of "fw ctl affinity -l" command and of "sim affinity -l" command shows both CoreXL SNDs and CoreXL FW instances run on the same CPU core.

  *Example - CoreXL FW instances "fw_\<X\>" run on the same CPU cores, to which the interfaces were assigned (i.e., the same CPU cores, on which CoreXL SNDs run)*:

  ```
  
  [Expert@CP2:0]# fw ctl affinity -l
  eth2-01: CPU 0
  eth2-02: CPU 1
  eth2-03: CPU 2
  eth2-04: CPU 3
  eth2-05: CPU 4
  eth2-06: CPU 5
  eth2-07: CPU 6
  eth2-08: CPU 7
  eth2-09: CPU 8
  eth3-02: CPU 9
  eth3-04: CPU 10
  fw_0: CPU 11
  fw_1: CPU 10
  fw_2: CPU 9
  fw_3: CPU 8
  fw_4: CPU 7
  fw_5: CPU 6
  fw_6: CPU 5
  fw_7: CPU 4
  fw_8: CPU 3
  fw_9: CPU 2
  [Expert@CP2:0]# sim affinity -l
  eth2-01 : 0
  eth2-02 : 1
  eth2-03 : 2
  eth2-04 : 3
  eth2-05 : 4
  eth2-06 : 5
  eth2-07 : 6
  eth2-08 : 7
  eth2-09 : 8
  eth3-02 : 9
  eth3-04 : 10
  ```

## Cause

Incorrect affinity of interfaces creates a bottle-neck and does not allow to distribute the traffic load between CoreXL FW Instances. As a result, CPU cores are overloaded, which in turn affects the processing of CCP packets. As a result, interfaces are declared by the cluster as "down", which by design leads to the cluster member going down.

## 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