Hackers are exploiting a critical remote code execution vulnerability in Citrix NetScaler Application Delivery Controller and NetScaler Gateway to target appliances with web shells and malware.
Tracked as CVE-2026-88771, the flaw allows unauthenticated attackers to run arbitrary commands without enabling any optional features.
Citrix disclosed the vulnerability on September 27, 2026, and confirmed exploitation against unmitigated deployments.
The company assigned a CVSS v4.0 score of 9.5. CISA added the flaw to its Known Exploited Vulnerabilities catalog the same day, highlighting the need for urgent action.nvd.nist+1
According to the supplied Zscaler ThreatLabz analysis, attackers combined access-log poisoning, forged authentication records, appliance fingerprinting, and attempts to install several web shells. Researchers also observed delivery of a Rust-based FreeBSD implant.
Hackers Exploit NetScaler RCE
The attack begins with HTTP GET requests to common NetScaler login pages. These requests contain a User-Agent value starting with INDEX:, followed by a Base64-encoded payload.
The appliance records the value in its HTTP access logs, storing the malicious instructions without immediately executing them.
Attackers then submit login requests containing fake Packet Processing Engine failure messages. These messages imitate legitimate alerts, including “PPE missed too many heartbeats” and “PPE unexpectedly died,” but include shell commands.
The trigger arrives when the legitimate diagnostic script /netscaler/ns_monuploadd_err.pl processes matching failure records.
It inserts attacker-controlled log content into a shell command without safely handling special characters. This allows the embedded instructions to execute with the script’s privileges.
The commands retrieve the staged payload from HTTP logs, decode it, and execute it. CERT-EU found that the script selects the last matching record, explaining why attackers repeatedly submit forged messages to keep their entry most recent.
ThreatLabz reported attempts to create fingerprint files and deploy PHP and Perl web shells across several web-accessible directories.
Attackers also tried changing /etc/httpd.conf to enable PHP and CGI execution. A separate delivery attempt downloaded a FreeBSD x64 implant from 209.141.43[.]95.
ThreatLabz linked that server to infrastructure previously observed during StyleSmuggler exploitation against Adobe Commerce and Magento. However, shared infrastructure alone does not establish common ownership or attacker attribution.
Administrators should identify internet-facing appliances, confirm their software builds, and preserve forensic evidence before updating. CERT-EU recommends capturing evidence before promptly applying patches.
Citrix’s bulletin lists fixed releases beginning with 14.1-73.37, 13.1-64.23, 14.1-73.37 FIPS, and 13.1.37.279 for FIPS and NDcPP deployments. Organizations should install the appropriate supported update or a later release.
Defenders should examine HTTP logs for INDEX: markers and encoded User-Agent content, then correlate them with suspicious PPE failure messages. Unexpected changes to /etc/httpd.conf can reveal attempts to activate web shells, zscaler said.
Indicators of Compromise
| Indicator type | Indicator | Description |
|---|---|---|
| Payload delivery IP | 209.141.43[.]95 | Server used to deliver the FreeBSD x64 implant |
| Payload URL | hxxp://209.141.43[.]95/files/kw2-freebsd-x64-s11 | Observed implant download location |
Note: IP addresses and domains are intentionally defanged (e.g., [.]) to prevent accidental resolution or hyperlinking. Re-fang only within controlled threat intelligence platforms such as MISP, VirusTotal, or your SIEM.
Join 16,000+ SOC teams using ANY.RUN to streamline threat investigations and reduce manual effort. Explore for your team