CVE-2023-24941: Microsoft Network File System Remote Code Execution
June 01, 2023 | Trend Micro Research TeamIn this excerpt of a Trend Micro Vulnerability Research Service vulnerability report, Quinton Crist, Guy Lederfein, and Lucas Miller of the Trend Micro Research Team detail a recently patched remote code execution vulnerability in the Microsoft Network File Service (NFS). This bug was originally discovered by Wei in Kunlun Lab with Cyber KunLun. The vulnerability is triggered when handling incoming NFSv4.1 calls containing utf8strings when the server is low on memory. A remote, unauthenticated attacker could exploit this vulnerability by sending a crafted call to an affected server. The following is a portion of their write-up covering CVE-2023-24941, with a few minimal modifications.
A remote code execution vulnerability has been reported in Microsoft Network File System (NFS). The vulnerability is triggered when handling incoming NFSv4.1 calls.
The Vulnerability
Microsoft Windows ships with several network features, some of which can be used to communicate with non-Windows file shares. One of these modules is NFS.
Network File System (NFS) is a network file system protocol originally developed by Sun Microsystems in 1984. Version 2 is documented in RFC 1094. Version 3 is documented in RFC 1813. Version 4 was developed by the IETF and is documented in RFC 3010 (released December 2000) and RFC 3530 (released April 2003). NFS allows users to access remote file shares in the same way that the local file system is accessed. Different access levels and permissions can be set on the share, such as read-write and read-only. Additionally, IP/UID/ GID/Kerberos security can be used. NFS uses Open Network Computing (ONC) Remote Procedure Call (RPC) to exchange control messages. ONC RPC was originally developed by Sun Microsystems, it can also be referred to as Sun RPC.
When ONC RPC messages are transferred over TCP, they are prepended with a Fragment header structure (as illustrated in the following table) that specifies the length of the message. This allows the receiver to distinguish multiple messages sent over a single TCP session. Other protocols such as UDP do not use this field. Note that all multi-byte values are encoded in big-endian byte order.
In the NFS4 protocol, a utf8string is transferred in the following format:
A remote codes execution vulnerability has been reported in Microsoft Network File System. The vulnerability is triggered when handling incoming NFSv4.1 calls containing utf8strings when the server is low on memory. When a server is parsing a received string, a buffer is allocated to store the string data. The code does not properly handle the allocation failing, resulting in the null termination byte still being written to the end of an invalid buffer.
A remote, unauthenticated attacker could exploit this vulnerability by sending a crafted call to the victim Network File System service. Successful exploitation of this vulnerability can result in remote code execution.
Source Code Walkthrough
The following code snippet was taken from nfssvr.sys version 10.0.17763.4252. Comments added by Trend Micro have been highlighted.
Detection Guidance
To detect an attack exploiting this vulnerability, the detection device must monitor and parse traffic on ports 2049/TCP and 2049/UDP.
When ONC RPC messages are transferred over TCP, they are prepended with a Fragment header structure (as illustrated in the table above) that specifies the length of the message. This allows the receiver to distinguish multiple messages sent over a single TCP session. Other protocols such as UDP do not use this field. The vulnerable method XdrDecodeString can only trigger this vulnerability when called from Nfs4SvrXdrpDecode_STRING.
The following fields are parsed using Nfs4SvrXdrpDecode_STRING and are defined in RFC 3530:
The detection device should monitor the above fields in all NFS4 messages. Any valid field value could trigger this vulnerability, but an attacker would request a large buffer to increase the probability of the failed allocation happening during the string processing. String lengths greater than 0x1000 should be considered suspicious. If found, an attack exploiting this vulnerability is likely underway.
Note that the detection of strings larger than 0x1000 is based on the typical allowed limit of file path strings and can be adjusted higher or lower to account for various server configurations.
Conclusion
Microsoft patched this bug in May as CVE-2023-24941. In their write-up, they note that disabling NFSv4.1 and downgrading to NFSv2 or NFSv3 can be used as a temporary mitigation for this vulnerability. However, they also note that you should not employ this mitigation unless you have already installed CVE-2022-26937 from the May 2022 Windows security updates. The better option is to test and deploy the latest patch for NFSv1 to completely address this vulnerability.
Special thanks to Quinton Crist, Guy Lederfein, and Lucas Miller of the Trend Micro Research Team for providing such a thorough analysis of this vulnerability. For an overview of Trend Micro Research services please visit http://go.trendmicro.com/tis/.
The threat research team will be back with other great vulnerability analysis reports in the future. Until then, follow the team on Twitter, Mastodon, LinkedIn, or Instagram for the latest in exploit techniques and security patches.