> Source: [sk163413](https://support.checkpoint.com/results/sk/sk163413)

# sk163413 - CPD, FWD, and FWM processes on the Security Management Server or the Security Gateway randomly crash and cannot start

| Property | Value |
|----------|-------|
| Solution ID | sk163413 |
| Date Created | 2019-11-11 |
| Last Modified | 2021-12-14 |
| Technical Level | General |
| Products | Security Gateway, Security Management Server, Multi-Domain Security Management Server |
| Versions | R81.10 (EOS), R81 (EOS), R81.10 (EOS), R81 (EOS), R81 (EOS), R81.10 (EOS) |

## Symptoms

- * CPD, FWD and FWM processes on the Security Management Server or the Security Gateway randomly crash and cannot start.
* "`T_get_event: cannot register socket XXX (1024 sockets already registered for in)`" line in the log file of the failed process (`$CPDIR/log/cpd.elg`, `$FWDIR/log/fwd.elg`, and `$FWDIR/log/fwm.elg`).

## Cause

Processes cannot handle more than 1024 connections and fail.

## Solution

This problem was fixed. The fix is included starting from:

* [Check Point R81.10](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doGoviewsolutiondetails=&solutionid=sk170416)
* [Jumbo Hotfix Accumulator for R81](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doGoviewsolutiondetails=&solutionid=sk170114#Take%2042) starting from Take 42
* [Jumbo Hotfix Accumulator for R80.40](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doGoviewsolutiondetails=&solutionid=sk165456#Take%20119) starting from Take 119
* [Jumbo Hotfix Accumulator for R80.30](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doGoviewsolutiondetails=&solutionid=sk153152#Take%20237) starting from Take 237
* [Jumbo Hotfix Accumulator for R80.20](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doGoviewsolutiondetails=&solutionid=sk137592#Take%20203) starting from Take 203

<br />

**Code was improved: The Log server now supports up to 2700 Gateways.**   

<br />

Check Point recommends to always upgrade to the most recent version ([Security Gateway](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doShowproductpage&productTab=downloads&product=435) / [Security Management Server](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doShowproductpage&productTab=downloads&product=184) / [Multi-Domain Security Management Server](https://supportcenter.checkpoint.com/supportcenter/portal?eventSubmit_doShowproductpage&productTab=downloads&product=166)).

<br />

<br />

<br />

If you are not able to install the Release or Jumbo Hotfix Accumulator that resolves this issue, then as a workaround, enable the Linux EPOLL API for the affected processes.

With this API, a process monitors one of a set of file descriptors to become ready to perform I/O (for more information, see <https://linux.die.net/man/7/epoll> and <https://linux.die.net/man/2/select>).

**Important:** Schedule a maintenance window.

1. Connect to the command line on the affected Security Management Server or Security Gateway.

2. Log in to the Expert mode.

3. Edit the *$CPDIR/tmp/.CPprofile.sh* file:

   `vi $CPDIR/tmp/.CPprofile.sh`
4. Add this line at the end:

   * To enable the Linux EPOLL API for all the affected processes:

     `EPOLL_DISABLE=FALSE; export EPOLL_DISABLE`
   * To enable the Linux EPOLL API for the CPD process only:

     `EPOLL_DISABLE_CPD=FALSE; export EPOLL_DISABLE_CPD`
5. Save the changes in the file and exit the Vi editor.

6. Restart the Check Point services:

   * On a Security Management Server:

     `cpstop ; cpstart`

     **Important:** This will disconnect all SmartConsole clients.
   * On a Multi-Domain Security Management Server:

     `mdsstop ; mdsstart`

     **Important:** This will disconnect all SmartConsole clients.
   * On a Security Gateway (Cluster Members):

     `cpstop ; cpstart`

     **Important:** In cluster, this can cause a failover.
7. Confirm that the new environment variable is loaded:

   * If you configured the variable for all the affected processes:

     `echo $EPOLL_DISABLE`
   * If you configured the variable for the CPD process only:

     `echo $EPOLL_DISABLE_CPD`

   The output must show:

   `FALSE`

**Related SK Articles:**

* [sk168692 - The FWD process stops working randomly on Security Group Members on Scalable Chassis and Maestro](https://supportcontent.checkpoint.com/solutions?id=sk168692)

---

# 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