Vulnerabilities > CVE-2021-1098 - Improper Resource Shutdown or Release vulnerability in Nvidia Virtual GPU

047910
CVSS 4.6 - MEDIUM
Attack vector
LOCAL
Attack complexity
LOW
Privileges required
NONE
Confidentiality impact
PARTIAL
Integrity impact
PARTIAL
Availability impact
PARTIAL
local
low complexity
nvidia
CWE-404

Summary

NVIDIA vGPU software contains a vulnerability in the Virtual GPU Manager (vGPU plugin), where it doesn't release some resources during driver unload requests from guests. This flaw allows a malicious guest to perform operations by reusing those resources, which may lead to information disclosure, data tampering, or denial of service. This affects vGPU version 12.x (prior to 12.3), version 11.x (prior to 11.5) and version 8.x (prior 8.8).

Common Weakness Enumeration (CWE)

Common Attack Pattern Enumeration and Classification (CAPEC)

  • Flooding
    An attacker consumes the resources of a target by rapidly engaging in a large number of interactions with the target. This type of attack generally exposes a weakness in rate limiting or flow control in management of interactions. Since each request consumes some of the target's resources, if a sufficiently large number of requests must be processed at the same time then the target's resources can be exhausted. The degree to which the attack is successful depends upon the volume of requests in relation to the amount of the resource the target has access to, and other mitigating circumstances such as the target's ability to shift load or acquired additional resources to deal with the depletion. The more protected the resource and the greater the quantity of it that must be consumed, the more resources the attacker may need to have at their disposal. A typical TCP/IP flooding attack is a Distributed Denial-of-Service attack where many machines simultaneously make a large number of requests to a target. Against a target with strong defenses and a large pool of resources, many tens of thousands of attacking machines may be required. When successful this attack prevents legitimate users from accessing the service and can cause the target to crash. This attack differs from resource depletion through leaks or allocations in that the latter attacks do not rely on the volume of requests made to the target but instead focus on manipulation of the target's operations. The key factor in a flooding attack is the number of requests the attacker can make in a given period of time. The greater this number, the more likely an attack is to succeed against a given target.
  • Excessive Allocation
    An attacker causes the target to allocate excessive resources to servicing the attackers' request, thereby reducing the resources available for legitimate services and degrading or denying services. Usually, this attack focuses on memory allocation, but any finite resource on the target could be the attacked, including bandwidth, processing cycles, or other resources. This attack does not attempt to force this allocation through a large number of requests (that would be Resource Depletion through Flooding) but instead uses one or a small number of requests that are carefully formatted to force the target to allocate excessive resources to service this request(s). Often this attack takes advantage of a bug in the target to cause the target to allocate resources vastly beyond what would be needed for a normal request. For example, using an Integer Attack, the attacker could cause a variable that controls allocation for a request to hold an excessively large value. Excessive allocation of resources can render a service degraded or unavailable to legitimate users and can even lead to crashing of the target.
  • Resource Leak Exposure
    An attacker utilizes a resource leak on the target to deplete the quantity of the resource available to service legitimate requests. Resource leaks most often come in the form of memory leaks where memory is allocated but never released after it has served its purpose, however, theoretically, any other resource that can be reserved can be targeted if the target fails to release the reservation when the reserved resource block is no longer needed. In this attack, the attacker determines what activity results in leaked resources and then triggers that activity on the target. Since some leaks may be small, this may require a large number of requests by the attacker. However, this attack differs from a flooding attack in that the rate of requests is generally not significant. This is because the lost resources due to the leak accumulate until the target is reset, usually by restarting it. Thus, a resource-poor attacker who would be unable to flood the target can still utilize this attack. Resource depletion through leak differs from resource depletion through allocation in that, in the former, the attacker may not be able to control the size of each leaked allocation, but instead allows the leak to accumulate until it is large enough to affect the target's performance. When depleting resources through allocation, the allocated resource may eventually be released by the target so the attack relies on making sure that the allocation size itself is prohibitive of normal operations by the target.