Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessGain control of a guest VM
-
DeliveryNegotiate oversized multi-page blkif ring
-
ExploitOverflow stack in tap_ctl_connect_xenblkif()
-
ExecutionExecute code as root in dom0
-
ImpactCompromise host and co-resident guests
Vulnerability Assessment
| Exploitation Conditions | The host must run the XAPI toolstack with tapdisk (blktap) as the block backend, and the attacker must already control a guest VM whose paravirtualized block frontend connects to that backend. The specific trigger is the guest negotiating a multi-page ring whose size exceeds 8 pages during connection - tapdisk performs no boundary check and overflows a stack buffer. No special host configuration is needed (all versions are vulnerable and there are no mitigations), so any default XAPI/tapdisk deployment is affected. The principal limiting factor is that exploitation requires control of a guest VM on the host (a guest-to-host attack), not mere network reachability; there is no user interaction required on the host side. |
| Impact | This is a credible, high-severity guest-to-host privilege-escalation issue despite the complete absence of quantitative scoring signals. No CVSS vector, EPSS score, KEV listing, SSVC decision, or POC is present in the input - so prioritization must rest entirely on the vendor's own description, which is authoritative (Xen Security Team, XSA-521). The vendor states a buggy or malicious guest can trigger a stack overflow in tapdisk, which normally runs as root in the control domain, 'potentially leading to code execution' - i.e. a possible guest escape to dom0. The lack of any boundary check on the frontend-negotiated multi-page ring size (assumed ≤8) means the attacker controls the overflow trigger with low complexity. Key signals are MISSING rather than conflicting: there is no EPSS to gauge exploitation likelihood, no KEV entry, and no public exploit, so real-world exploitation volume cannot be estimated from the data. Note the impact is qualified ('potentially leading to code execution'), not confirmed RCE, so treat code execution as plausible-but-unproven; a stack overflow at minimum yields a denial of service (tapdisk crash) affecting guest storage. Because 'all versions are vulnerable' and 'there are no mitigations,' patching is the only remediation and should be prioritized on any multi-tenant XAPI/blktap host exposed to untrusted guests, even though no active exploitation is indicated. |
| Exploit ScenarioDerived | An attacker who controls a guest VM on a Xen host using the XAPI toolstack drives the paravirtualized block frontend to negotiate a multi-page ring larger than the 8-page maximum tapdisk assumes, overflowing a stack buffer in tap_ctl_connect_xenblkif(). Because tapdisk runs as root in the control domain, a successful overflow can crash the storage backend or potentially execute attacker-controlled code with root privileges, escaping guest isolation. No public exploit code was identified in the provided data. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 13, 2026
Guest-to-host privilege escalation in Xen's tapdisk (blktap) userspace xen-blkback backend lets a malicious or buggy guest VM crash or potentially execute code as host root: all versions of tapdisk are affected because the device-connect path performs no boundary check on the multi-page ring size requested by the frontend. Exploitation requires only that the host runs the XAPI toolstack with tapdisk as the block backend and that the attacker already controls a guest VM attached to it (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H), with no user interaction on the host side; the vulnerability is reachable under default configurations, and the Xen Security Advisory states there are no mitigations. No public exploit code or active exploitation has been identified at time of analysis, and the fix is an upstream patch (xsa521.patch) for blktap master rather than a named released version.
RemediationAI
Apply the vendor patch: Xen Security Advisory 521 attaches xsa521.patch for blktap master (SHA-256 72ece150e9d0836f217f3382c8dc76cd51bffbe4b7610c00f219b90e5600cbc5), which adds the missing boundary check on the frontend-requested ring size and resolves the issue; because the patch targets master rather than a tagged release, treat this as an upstream fix and obtain the released, patched package from your XAPI distribution vendor (XenServer/Citrix Hypervisor, XCP-ng, or equivalent) once it is published. The advisory is available at https://seclists.org/oss-sec/2026/q4/149 with project policy context at http://www.xenproject.org/security-policy.html. Note that XSA-521 explicitly states there are no mitigations, so there is no configuration change that removes the risk while continuing to run untrusted guests on tapdisk: the practical compensating controls are to apply the patch at the earliest opportunity and, in the interim, restrict hosts running the vulnerable tapdisk to guests that are fully trusted, or avoid running untrusted guest workloads on those hosts. Avoiding tapdisk altogether (for example by changing the block backend for affected virtual disks) removes exposure to this specific code path but may reduce functionality or performance and is not a supported substitute for patching; revocation or quarantine of any guest whose backend connection is suspected of triggering the overflow, combined with host crash-dump analysis, can help confirm attempted exploitation.
Affected ProductsAI
All versions of tapdisk are vulnerable according to Xen Security Advisory 521 (CVE-2026-102701). Tapdisk is the userspace blktap xen-blkback implementation shipped and used by the XAPI toolstack, so affected deployments are XAPI-based platforms and distributions that run guests with tapdisk as the block backend, including XenServer/Citrix Hypervisor, XCP-ng and other XAPI-derived stacks; no CPE strings were supplied with this intelligence, so exact package and build identifiers must be confirmed against the relevant distribution's advisory. The authoritative upstream reference is the oss-security posting of XSA-521 at https://seclists.org/oss-sec/2026/q4/149, with project process detail at http://www.xenproject.org/security-policy.html. Because the flaw lies in the connection path rather than in version-specific code, no tapdisk release can be considered unaffected prior to the xsa521.patch fix.
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Tapdisk is a userspace implementation of the xen-blkback protocol used by the XAPI toolstack to provide paravirtualized block devices to guests, running as a privileged (root) process on the host. During backend connection, the frontend and backend negotiate the shared ring, and the Xen block protocol permits multi-page rings; the vulnerable function tap_ctl_connect_xenblkif() accepted the ring size supplied by the frontend without validating it, assuming it could never exceed 8 pages. Because the oversized value is written into a fixed-size stack buffer, a guest that negotiates an out-of-range ring triggers a stack overflow (out-of-bounds write class, e.g. CWE-121/CWE-787; no CWE identifier is assigned in the supplied data), which the advisory assesses as potentially leading to code execution in this root-privileged process. This is a classic guest-to-host trust-boundary failure in a paravirtualized I/O backend, reflected in the assessment's Scope:Changed vector: the malicious input originates in the untrusted guest and the resulting impact lands on the host. Discovery is credited to Claude Opus, with diagnosis by Teddy Astie of Vates; all versions of tapdisk are vulnerable and the advisory explicitly notes there are no mitigations.
Know when the next one hits your stack
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
Attacker must control a guest (PR:L) and acts over the local virtualization ring interface (AV:L); the unchecked oversized ring is trivially triggered (AC:L); root code execution in dom0 crosses a trust boundary (S:C, C/I/A:H).
Estimated by vuln.today — no official severity rating has been published for this CVE yet.
Description PRE-NVD
Executive ViewAI
Xen environments using the XAPI toolstack with tapdisk as the block backend face a guest-to-host privilege escalation risk that can allow a controlled guest VM to crash the host or potentially execute code as host root. Because the issue is reachable under default configurations and no vendor-released patch or mitigation is available, organizations hosting untrusted or multi-tenant workloads on affected Xen hosts should treat it as a high-priority infrastructure risk. Leadership should care because a successful exploit could compromise the hypervisor and all resident workloads, even though no public exploit has been identified at time of analysis.
No vendor-recommended interim mitigations identified at time of analysis; restrict untrusted guests on affected XAPI/tapdisk hosts and monitor for host crashes.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
Attack Technique
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
CVE.org
GitHub PoC
Exploit-DB