Skip to main content
kcheng
Staff & Editor
Staff & Editor
April 1, 2026

Troubleshooting Tip: Potential Reason for FortiGate intermittently rebooting without major hardware or software issue

  • April 1, 2026
  • 0 replies
  • 367 views
Description This article describes the chronological steps to investigate an issue where the FortiGate was observed being rebooted every 2 hours, although the Hardware Quick Inspection Package (HQIP) test on the FortiGate passes all tests, and there are no software bugs observed.
Scope FortiOS.
Solution

In a scenario where the FortiGate randomly went unresponsive and recovered without any performed actions, it is always recommended to check the following information:

  • Last reboot reason: get system status | grep "Last".
  • Crashlog: diagnose debug crashlog read.
  • Console logs.
  • System Event logs.

 

While there could be many reasons that could cause a reboot on the FortiGate, this article focuses on the behavior where FortiGate was being rebooted every 2 hours due to misconfiguration on Automation Stitch. In the FortiGate command line, the information would be illustrated as follows for symptoms that it is being rebooted by an administrator or triggered under certain conditions:

 

danum-kvm65 # get system status | grep "Last" Last reboot reason: warm reboot

 

Refer to the following article for different types of reboots on FortiGate: Technical Guide: Identifying the FortiGate reboot reasons and what to do next.

 

Besides the reboot reason, the crashlog indicates that the FortiGate interfaces were brought down by the process_name 'autod':

 

danum-kvm65 # diagnose debug crashlog read | grep 2026-04 55: 2026-04-01 08:29:43 the killed daemon is /bin/sflowd: status=0x0 56: 2026-04-01 08:36:19 the killed daemon is /bin/updated: status=0x0 57: 2026-04-01 08:37:29 Interface port1 is brought down. process_id=2079, process_name="autod" 58: 2026-04-01 08:37:29 Interface port2 is brought down. process_id=2079, process_name="autod" 59: 2026-04-01 08:37:29 Interface port3 is brought down. process_id=2079, process_name="autod" 60: 2026-04-01 08:37:29 Interface port4 is brought down. process_id=2079, process_name="autod" 61: 2026-04-01 08:37:29 Interface port5 is brought down. process_id=2079, process_name="autod" 62: 2026-04-01 08:37:29 Interface port6 is brought down. process_id=2079, process_name="autod" 63: 2026-04-01 08:37:29 Interface port7 is brought down. process_id=2079, process_name="autod" 64: 2026-04-01 08:37:29 Interface port8 is brought down. process_id=2079, process_name="autod" 65: 2026-04-01 08:37:29 Interface port9 is brought down. process_id=2079, process_name="autod" 66: 2026-04-01 08:37:29 Interface port10 is brought down. process_id=2079, process_name="autod" 67: 2026-04-01 08:38:08 the killed daemon is /bin/sflowd: status=0x0 

 

From the information above, it is possible to narrow down to the abnormality caused by the autod daemon. The autod daemon is the daemon that handles Automation Stitches functionality. 

 

Connecting a laptop or host directly to the FortiGate console port is another way to identify what was happening during the time of the incident. In this incident, it has been observed that there were no kernel crashes that occurred before the system was instructed to reboot. This scenario usually indicates a warm reboot:

 

danum-kvm65 #  The system is going down NOW !!  Please stand by while rebooting the system. Restarting system  System is starting... 

 

In the System event logs, it is also possible to identify the root cause of the reboot if it's triggered under the same condition explained in this article. The following example illustrates the system event where automation stitch 'IPS_AV_DB' is being triggered, and it follows with the system event that states 'User rebooted the device from autod':

 

image.png

 

Navigate to 'Security Fabric -> Automation' to check on the configuration of the automation stitch named 'IPS_AV_DB'. It has been confirmed that an automation stitch was configured to reboot the FortiGate when the antivirus and IPS DB gets updated:

 

image.png

 

Deleting or modifying the action on this automation stitch resolves the issue of the FortiGate being rebooted every 2 hours. The reason for the duration is that the default scheduled update settings on the FortiGate are being set to every 2 hours. 

Technical Tip: How to check most updated security version database on FortiGate

 

The default auto-update schedule has been changed to daily from FortiOS v7.6.5 and above under bug ID 1204277:

Changes in default behavior. If this misconfiguration is done on FortiOS v7.6.5 and above, the FortiGate will be observed being rebooted on a daily basis when FortiGate performs IPS and antivirus database updates.

 

Automation stitches is a powerful tool to automate certain information collection, perform certain actions, and/or trigger an alert to the administrator when a certain scenario is being detected. Hence, it is important to understand the impact of an automation stitch before configuring the relevant settings. 

 

Related documents:

Technical Tip: Short list of processes on the FortiGate 

Creating automation stitches 

Technical Tip: Use FortiGate automation stitches for alert emails 

Technical Tip: Creating automation stitches 

Technical Tip: How to configure an Automation Stitch to execute a packet capture in a desired time 

Technical Tip: How to configure Automation stitch for Downstream FortiGate and confirm it is working 

Technical Tip: Using automation stitches to Enable/Disable IPV4 Firewall Policy based on SD-WAN logs 

Technical Tip: How FortiGate Trigger Automation-stitch with Multiple Events