The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Friday added a critical security flaw impacting F5 BIG-IP Access Policy Manager (APM) to its Known Exploited Vulnerabilities (KEV) catalog, citing evidence of active exploitation.
The vulnerability in question is CVE-2025-53521 (CVSS v4 score: 9.3), which could allow a threat actor to achieve remote code execution.
“When a BIG-IP APM access policy is configured on a virtual server, specific malicious traffic can lead to Remote Code Execution (RCE),” according to a description of the flaw in CVE.org.
While the shortcoming was initially categorized and remediated as a denial-of-service (DoS) vulnerability with a CVSS v4 score of 8.7, F5 said it has been reclassified as a case of RCE in light of “new information obtained in March 2026.”
The company has since updated its advisory to confirm that the vulnerability “has been exploited in the vulnerable BIG-IP versions.” It did not share any additional details on who may be behind the exploitation activity.
However, F5 published a number of indicators that can be used to assess if the system has been compromised –
- File-related indicators –
- Presence of /run/bigtlog.pipe and/or /run/bigstart.ltm.
- Mismatch of file hashes when compared to known good versions of /usr/bin/umount and/or /usr/sbin/httpd.
- Mismatch of file sizes or timestamps when compared to known good versions of /usr/bin/umount and/or /usr/sbin/httpd.
- Each release and EHF may have different file sizes and timestamps.
- Log-related indicators –
- An entry in “/var/log/restjavad-audit.
.log” showing a local user accessing the iControl REST API from localhost. - An entry in “/var/log/auditd/audit.log.
” showing a local user accessing the iControl REST API from localhost to disable SELinux. - Log messages in “/var/log/audit” show the results of a command being run in the audit log.
- An entry in “/var/log/restjavad-audit.
- Other TTPs observed include –
- Modifications to the underlying components that the system integrity checker, sys-eicheck, relies on, resulting in a failure of the tool, specifically /usr/bin/umount and/or /usr/sbin/httpd, indicating unexpected changes to the system software as mentioned above.
- HTTP/S traffic from the BIG-IP system that contains HTTP 201 response codes and CSS content-type to disguise the attacker’s activities.
- Changes to the following three files, although their presence alone does not signal a security issue –
- /var/sam/www/webtop/renderer/apm_css.php3
- /var/sam/www/webtop/renderer/full_wt.php3
- /var/sam/www/webtop/renderer/webtop_popup_css.php3
“We have observed cases of webshell being written to disk; however, the webshells have been observed to work in memory only, meaning the files listed above might not be modified,” F5 cautioned.
The issue impacts the following versions –
- 17.5.0 – 17.5.1 (Fixed in version 17.5.1.3)
- 17.1.0 – 17.1.2 (Fixed in version 17.1.3)
- 16.1.0 – 16.1.6 (Fixed in version 16.1.6.1)
- 15.1.0 – 15.1.10 (Fixed in version 15.1.10.8)
In light of active exploitation, Federal Civilian Executive Branch (FCEB) agencies have been given until March 30, 2026, to apply the fixes to secure their networks.
“When F5 CVE-2025-53521 first emerged last year as a denial-of-service issue, it didn’t immediately signal urgency, and many system administrators likely prioritized it accordingly,” watchTowr CEO and founder Benjamin Harris said in a statement shared with The Hacker News.
“Fast forward to today’s big ‘yikes’ moment: the situation has changed significantly. What we’re observing now is pre-auth remote code execution and evidence of in-the-wild exploitation, with a CISA KEV listing to back it up. That’s a very different risk profile than what was initially communicated.”
Defused Cyber, in an X post, has also confirmed that it’s seeing “acute scanning activity” for vulnerable F5 BIG-IP devices following the addition of CVE-2025-53521 to the KEV catalog.
“This actor is hitting /mgmt/shared/identified-devices/config/device-info which is a F5 BIG-IP REST API endpoint used to retrieve system-level information, such as hostname, machine ID, and base MAC address,” it said.
