CVE-2025-68686 Splunk · SPL

Detect Fortinet FortiOS Sensitive Information Exposure (CVE-2025-68686) in Splunk

CVE-2025-68686 is an Exposure of Sensitive Information to an Unauthorized Actor vulnerability (CWE-200) in Fortinet FortiOS, added to CISA's Known Exploited Vulnerabilities catalog. Unauthorized actors may be able to access sensitive information exposed by affected FortiOS instances, potentially including configuration data, credentials, session tokens, or internal system details that could facilitate further compromise. This detection focuses on identifying anomalous access to FortiOS management interfaces, unusual API/CLI enumeration behavior, and downstream indicators of information disclosure abuse such as suspicious authentication following reconnaissance against exposed FortiOS endpoints.

MITRE ATT&CK

Tactic
Reconnaissance Initial Access Discovery

SPL Detection Query

Splunk (SPL)
spl
index=network_firewall sourcetype=fortinet:fortigate
| eval mgmt_action=case(match(_raw,"(?i)login"),"login", match(_raw,"(?i)config"),"config", match(_raw,"(?i)backup"),"backup", match(_raw,"(?i)download"),"download", 1=1,"other")
| where mgmt_action!="other"
| bin _time span=1h
| stats count as request_count values(mgmt_action) as actions dc(user) as unique_users by src_ip, dest_ip, _time
| where request_count > 20
| sort - request_count
high severity medium confidence

Flags bursts of FortiOS administrative/management log activity per source IP within an hourly window, consistent with attempts to enumerate or extract sensitive information related to CVE-2025-68686.

Data Sources

Fortinet FortiGate logs

Required Sourcetypes

fortinet:fortigate

False Positives & Tuning

  • Authorized config backup automation
  • IT staff performing scheduled maintenance
  • Security scanning traffic from approved scanners
  • High-traffic MSP management consoles managing multiple tenants

Other platforms for CVE-2025-68686


Testing Methodology

Validate this detection against 3 adversary techniques from Atomic Red Team. Each test below lists the behaviour to exercise and the telemetry you should expect to see. Executable commands and cleanup steps are available with Pro.

  1. Test 1Simulate high-volume FortiOS admin API access

    Expected signal: Multiple HTTP GET requests to /api/v2/cmdb/* endpoint from a single source IP within a short time window, logged in firewall/proxy logs.

  2. Test 2Simulate FortiOS config backup enumeration

    Expected signal: Repeated backup/config-related HTTP requests logged in FortiOS event logs and network proxy logs.

  3. Test 3Simulate SSL-VPN portal session enumeration

    Expected signal: Multiple failed/attempted SSL-VPN login events logged with consistent source IP within a short window.


Response Playbook

Triage

  1. Confirm the affected FortiOS device version against Fortinet's advisory FG-IR-25-934 to verify exposure to CVE-2025-68686.
  2. Review management interface (HTTPS admin, SSL-VPN portal) access logs for the affected device over the past 30 days for anomalous source IPs or unusual request volumes.
  3. Cross-reference alerting source IPs against known scanner/threat intel feeds and internal asset inventory to distinguish authorized administration from suspicious external access.
  4. Check whether the FortiOS management interface is internet-facing; if so, treat as high priority given active KEV exploitation status.

Containment

  1. Restrict or disable internet-facing access to FortiOS administrative interfaces (HTTPS admin, SSL-VPN management) via local-in policies or trusted host restrictions.
  2. Rotate all credentials, API tokens, and certificates that may have been exposed via the affected FortiOS instance, then apply Fortinet's patch per FG-IR-25-934.

Evidence Collection

  1. Export FortiOS event logs, admin login logs, and configuration change history for the affected time window for forensic review.
  2. Capture full packet captures or proxy logs for sessions to/from suspicious source IPs interacting with the management interface, preserving timestamps and payload sizes.

Escalation Criteria

  • !Escalate immediately if evidence shows successful extraction of configuration data, credentials, or VPN session data from an internet-facing FortiOS device.
  • !Escalate to incident response if the affected device is unpatched, internet-facing, and access patterns match known KEV exploitation TTPs referenced in CISA BOD 26-04 guidance.

Investigation Guide

Related Techniques

Forensic Artifacts

  • >FortiOS admin/webadmin access logs (event.log, elog entries)
  • >SSL-VPN portal authentication and session logs
  • >Configuration change history (diff logs) and backup file access timestamps

Tuning Guidance

Baseline normal administrative access patterns per FortiOS device (expected admin source IPs, typical request volume, scheduled backup windows) before enabling alerting thresholds. Exclude known MSP management IPs and vulnerability scanning ranges. Adjust the request-count threshold (default 20/hour) based on organizational admin activity volume, and prioritize alerts on internet-facing management interfaces per CISA BOD 26-04 guidance.


Hunting Queries

Hunts for repeated or bulk access to FortiOS configuration/API endpoints that may indicate information disclosure activity related to CVE-2025-68686.

Hunting — KQL
kql
CommonSecurityLog
| where DeviceVendor has "Fortinet"
| where isnotempty(RequestURL)
| where RequestURL has_any ("cmdb", "api", "backup")
| summarize count() by SourceIP, RequestURL, bin(TimeGenerated, 1h)
| order by count_ desc
Hunting — SPL
spl
index=network_firewall sourcetype=fortinet:fortigate url=*cmdb* OR url=*api* OR url=*backup*
| stats count by src_ip, url, _time
| sort - count

Atomic Red Team Tests

Test 1 Simulate high-volume FortiOS admin API access
linux

Generates repeated HTTP requests to a lab FortiOS-like management API path to simulate enumeration/extraction behavior for detection validation.

Command

bash
for i in $(seq 1 30); do curl -sk -o /dev/null -w '%{http_code}\n' https://<lab-fortios-ip>/api/v2/cmdb/system/status; done

Cleanup

bash
No persistent changes made; clear local curl history if desired: history -c

Expected Telemetry

Multiple HTTP GET requests to /api/v2/cmdb/* endpoint from a single source IP within a short time window, logged in firewall/proxy logs.

Expected Detection

KQL/SPL/QRadar rules should trigger on request_count > 20 within the 1-hour bin for the source IP.

Test 2 Simulate FortiOS config backup enumeration
linux

Simulates repeated attempts to access configuration backup endpoints on a lab FortiOS management interface.

Command

bash
for i in $(seq 1 25); do curl -sk -o /dev/null https://<lab-fortios-ip>/api/v2/monitor/system/config-backup; sleep 1; done

Cleanup

bash
No persistent state changes; no cleanup required beyond clearing shell history.

Expected Telemetry

Repeated backup/config-related HTTP requests logged in FortiOS event logs and network proxy logs.

Expected Detection

Sumo Logic and Chronicle YARA-L rules should flag the source IP for exceeding the config/backup action threshold.

Test 3 Simulate SSL-VPN portal session enumeration
windows

Simulates repeated login attempts against a lab FortiOS SSL-VPN portal to emulate credential/session enumeration tied to information exposure.

Command

powershell
1..25 | ForEach-Object { Invoke-WebRequest -Uri 'https://<lab-fortios-ip>/remote/login' -Method POST -Body @{username='testuser';password='testpass'} -SkipCertificateCheck -UseBasicParsing | Out-Null; Start-Sleep -Seconds 1 }

Cleanup

powershell
No persistent accounts or sessions created on the lab device; verify no residual sessions remain via FortiOS admin console.

Expected Telemetry

Multiple failed/attempted SSL-VPN login events logged with consistent source IP within a short window.

Expected Detection

CrowdStrike CQL and QRadar AQL rules should alert on the elevated event count for admin/login activity from the source IP.

Related Detections