> Source: [sk183793](https://support.checkpoint.com/results/sk/sk183793)

# sk183793 - ClusterXL Member changes its state to "Down" due to "Critical Device DSD"

| Property | Value |
|----------|-------|
| Solution ID | sk183793 |
| Date Created | 2025-08-19 |
| Last Modified | 2025-08-31 |
| Technical Level | Advanced |
| Products | Security Gateway |
| Versions | R82.10, R82, R81.20, R81.10 (EOS), R81 (EOS) |
| OS | Gaia |

## Symptoms

- * Output of the "`cphaprob state`" command shows that the ClusterXL Member changed its cluster state to "Down" as follows:

  ```
  
  ... ...
  Active PNOTEs: DSD
  
  Last member state change event:
     Event Code:                 CLUS-117500
     State change:               INIT -> DOWN
     Reason for state change:    PNOTE DSD
  ... ...
  ```

* The affected ClusterXL Member recovers only after a reboot.

* Output of the "`show dynamic-balancing state`" command ([sk164155](https://support.checkpoint.com/results/sk/sk164155)) on the affected ClusterXL Member shows that the CoreXL Dynamic Balancing is "off" (disabled).

* Restarting the CoreXL Dynamic Balancing on the affected ClusterXL Member does not help:

  `set dynamic-balancing state stop`

  `set dynamic-balancing state start`
* The `/var/log/dump/usermode/`` directory on the affected ClusterXL Member contains core dump files for the ``dsd`` process.`

* `The ``$FWDIR/log/dynamic_split.elg`` file on the affected ClusterXL Member shows:`

  `
  ``Dynamic Balancing is currently Off (Stopped due to State Verification failure)`

## Cause

Incorrect CoreXL Affinity configuration.

Multiple CoreXL Firewall instances were assigned to the same CPU cores (for example, 4-7 and 6-7), instead of a one-to-one mapping between instances and CPU cores.

Example output of the "`fw ctl multik stat`" command that shows the wrong configuration:

```
-----------------------------------------------
ID  | Active  | CPU    | Connections | Peak
-----------------------------------------------
0   | Yes     | 7      |         435 |      894
1   | Yes     | 6      |         467 |      890
2   | Yes     | 6-7    |         427 |      908
3   | Yes     | 6-7    |         416 |      892
4   | Yes     | 4-7    |         433 |      866
5   | Yes     | 4-7    |         422 |      873
```

This wrong configuration led to an issue with the CoreXL Dynamic Balancing - the DSD daemon, which is a Critical Device on ClusterXL Members.

The core dumps for the DSD daemon indicated that the unexpected exit was not due to resource exhaustion but rather to the incorrect affinity setup and the resulting failure in Dynamic Balancing and state verification.

## 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