Skip to main content
mbas
Explorer II
May 13, 2026 7:11 PM UTC
Solved

It is not possible to delete rogue devices from the FortiNAC database if there are too many

Hi all,

 

I am installing FortiNAC in a large environment and I added the Aruba WLC to FortiNAC without any issues. 3 days later, FortiNAC performed L2 pooling and discovered over 400,000 rogue hosts. After that, I disabled L2 pooling for the WLC, which stopped it from discovering.

I use FortiNAC version 7.6.6 and have increased the VM to 16 CPUs and 32 GB of RAM.

After that, I set the aging period to 1 day for the rogue hosts, but it did not work. I found this command, but it returned an error;

> execute enter-shell

# client -rog -op delete

 

error

During this time, the FortiNAC RAM usage was at 92%, and the server was extremely slow.

I could not find a way to get rid of these rogue hosts. While I was working, it did not affect the appliance, but when I configured PA for endpoint compliance checks, it did. While testing, I saw that many hosts were at risk, so I tried to mark them all as safe, but FortiNAC got stuck, so we had to restart the appliance.

After that, I listed all the risky hosts and deleted them. I also disabled the endpoint compliance check, but the customer doesn't want to continue working without deleting the rogue hosts.

 

Do you know of a way to delete all hosts from the database? Should I increase the FortiNAC resources to 32 CPUs and 96 RAM?

mbas
mbasAnswer
Explorer II
Jun 1, 2026 7:49 AM UTC

@AEK After restoring the database to 1 May, I saw that the rogue host count was 90,000. I set the aging period to 1 day and then ran the command below.

# client -rog -op delete

After that, the host count started to decrease, and 3 days later, there were 1,500 rogue hosts that had been newly registered.

My problem has been solved. Thank you so much for your help and time.

7 replies

AEK
SuperUser
SuperUser
May 13, 2026 8:55 PM UTC

Hi Mbas

I guess support may have the SQL access and commands to delete them directly from the DB.

Other option is to restore your FNAC from the last healthy backup (personally I’d most probably choose this option).

Translate
AEK
ebilcari
Staff
Staff
May 14, 2026 7:23 AM UTC

Or restoring the database from the auto-backup that is done every midnight:

I do not believe that the number of discovered rogue devices would cause a performance issue, as long as they are not online and are not being continuously evaluated.
If aging is set to one day, they will most likely be removed the following day.

Translate
Emirjon
mbas
mbasAuthor
Explorer II
May 17, 2026 3:01 PM UTC

The performance issue was related to device profiling. When I created and ran a rule, the process got stuck because more than 400,000 devices were identified as rogue.

 

Even when I set 'Aging' to one day, 420,000 devices were still listed as rogue after 1 week. 3 days ago, the number started to decrease slowly. Today, I checked and saw that there were 46,000 rogue hosts. As the server's CPU and RAM usage was at 90%, I think a process such as the device profiling rules was causing the issue. 

 

It looks like the problem has been solved. Thank you for your time and suggestion.

 

Best regards.

Translate
Risk and Resilience in the Age of AI. An OT Thought Leadership Summit for Houston's Critical Infrastructure Leaders. Date: Thursday, December 3, 2026. Time: 8:30 am - 1:30 pm (CT). Location: The Westin Houston, Memorial City | 945 Gessner Road, Houston, TX 77024