Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Citrix Bleed, CVE-2023-4966, leaked session tokens from NetScaler appliance memory, letting attackers hijack authenticated sessions and walk straight past multi-factor authentication.

What Is Citrix Bleed?
Citrix Bleed is the informal name for CVE-2023-4966, a sensitive information disclosure vulnerability in Citrix NetScaler ADC and NetScaler Gateway appliances configured as a gateway or an AAA virtual server. Citrix published a patch on 10 October 2023 and rated the issue 9.4 on the CVSS scale. Mandiant subsequently established that exploitation had been under way since late August 2023, several weeks before any advisory existed.
The flaw is deceptively simple. An over-long HTTP Host header caused the appliance to read past the end of a buffer when constructing its response, returning adjacent memory to the attacker. That memory routinely contained valid session tokens belonging to other, already authenticated users.
A session token represents an authentication that has already completed. Replaying one into the appliance gives the attacker the victim’s session without a password, without an account lockout, and — critically — without any further multi-factor authentication prompt, because the second factor was satisfied when the real user logged in.
This is the general problem with session-based authentication, familiar from ordinary cookie theft, but here the tokens were being read directly out of the memory of a perimeter device with no authentication required. Citrix Bleed is best understood as a session hijacking vulnerability with an unauthenticated remote trigger.
Exploitation became widespread through October and November 2023. LockBit affiliates were the most visible users, and CISA, the FBI, and partner agencies published a joint advisory on 21 November 2023 describing their tradecraft. Reported victims included Boeing, the US arm of Industrial and Commercial Bank of China, whose trading operations were disrupted in November 2023, Comcast Xfinity, and the legal and healthcare sectors in several countries. The NCSC and CISA both issued UK and US guidance urging immediate patching.
The typical intrusion pattern was consistent: harvest tokens, replay a session, enumerate Active Directory from the virtual desktop environment, extract credentials, move laterally, exfiltrate data, and deploy ransomware. Appliances provide a useful foothold precisely because they sit between the internet and the internal network and are often excluded from endpoint monitoring.
The detail that caught out many organisations is that applying the firmware update does not invalidate sessions that were stolen beforehand. An attacker holding a token captured in September 2023 could still use it against a patched appliance. Citrix’s advisory therefore instructed administrators to terminate all active and persistent sessions after upgrading:
kill icaconnection -all
kill rdp connection -all
kill pcoipConnection -all
kill aaa session -all
clear lb persistentSessions
Organisations that patched but skipped this step were compromised anyway. It is one of the most useful case studies available for the broader point that remediation of an exploited vulnerability includes eviction, not only the patch.
ns.log and web server logs for unusually large Host header values and for sessions whose source address changes mid-session.Internet-facing appliances remain one of the most productive initial access routes for ransomware operators, as the MOVEit, Fortinet, and Ivanti cases of the same period showed. Practical measures include subscribing to vendor advisories and treating appliance patching as an emergency change path rather than a quarterly one, restricting management interfaces to an administrative network, logging appliance telemetry into your SIEM, shortening session lifetimes, and binding sessions to a device or client certificate where the product supports it.
If a perimeter device in your estate was exposed during one of these windows, our investigation and incident response services can establish whether the exposure became an intrusion.
No. The stolen token represents a session in which authentication, including the second factor, has already completed. Replaying it skips the login flow entirely, so no prompt is generated. This is why the vulnerability was so widely exploited against organisations that considered themselves protected, and why session hijacking deserves attention separately from credential theft.
Because the firmware update stopped new tokens leaking but did nothing to the tokens already stolen. Citrix’s advisory included commands to terminate all active and persistent sessions after upgrading, and organisations that omitted that step left valid attacker sessions in place. Eviction is part of remediation whenever a vulnerability has been exploited before you patched it.
Citrix NetScaler ADC and NetScaler Gateway, in builds before 14.1-8.50, 13.1-49.15, and 13.0-92.19, and in the 12.1 branch, which was already end of life and received no fix. The appliance had to be configured as a gateway, such as a VPN virtual server, ICA proxy, or RDP proxy, or as an AAA virtual server, for the flaw to be reachable.
Look for oversized Host headers in appliance logs, sessions whose source address changes unexpectedly, and concurrent sessions for one user from distant locations. Appliance logs rotate quickly, so corroborate with Active Directory authentication records, virtual desktop session history, and internal network telemetry from late August 2023 onward. Engage incident responders if evidence is inconclusive.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.