FreshPorts - VuXML

Samuli Seppänen reports:

OpenVPN v2.4.0 was audited for security vulnerabilities independently by Quarkslabs (funded by OSTIF) and Cryptography Engineering (funded by Private Internet Access) between December 2016 and April 2017. The primary findings were two remote denial-of-service vulnerabilities. Fixes to them have been backported to v2.3.15.

An authenticated client can do the 'three way handshake' (P_HARD_RESET, P_HARD_RESET, P_CONTROL), where the P_CONTROL packet is the first that is allowed to carry payload. If that payload is too big, the OpenVPN server process will stop running due to an ASSERT() exception. That is also the reason why servers using tls-auth/tls-crypt are protected against this attack - the P_CONTROL packet is only accepted if it contains the session ID we specified, with a valid HMAC (challenge-response). (CVE-2017-7478)

An authenticated client can cause the server's the packet-id counter to roll over, which would lead the server process to hit an ASSERT() and stop running. To make the server hit the ASSERT(), the client must first cause the server to send it 2^32 packets (at least 196 GB).

< 2.3.15

ge 2.4.0 lt 2.4.2

< 2.3.15

ge 2.4.0 lt 2.4.2

< 2.3.15

< 2.3.15

Samuli Seppänen reports:

OpenVPN 2.3.11 [...] fixes two vulnerabilities: a port-share bug with DoS potential and a buffer overflow by user supplied data when using pam authentication.[...]

< 2.3.11

< 2.3.11

James Yonan reports:

A malicious [authenticated] client in "dev tap" ethernet bridging mode could theoretically flood the server with packets appearing to come from hundreds of thousands of different MAC addresses, causing the OpenVPN process to deplete system virtual memory as it expands its internal routing table.

< 2.0.1

The OpenVPN project reports:

In late November 2014 Dragana Damjanovic notified OpenVPN developers of a critical denial of service security vulnerability (CVE-2014-8104). The vulnerability allows an tls-authenticated client to crash the server by sending a too-short control channel packet to the server. In other words this vulnerability is denial of service only.

< 2.0.11

ge 2.1.0 lt 2.2.3

ge 2.3.0 lt 2.3.6

Steffan Karger reports:

The bounds check in read_key() was performed after using the value, instead of before. If 'key-method 1' is used, this allowed an attacker to send a malformed packet to trigger a stack buffer overflow. [...]

Note that 'key-method 1' has been replaced by 'key method 2' as the default in OpenVPN 2.0 (released on 2005-04-17), and explicitly deprecated in 2.4 and marked for removal in 2.5. This should limit the amount of users impacted by this issue.

< 2.3.18

ge 2.4.0 lt 2.4.4

ge 2.4.0 lt 2.4.4

< 2.3.18

James Yonan reports:

If the TCP server accept() call returns an error status, the resulting exception handler may attempt to indirect through a NULL pointer, causing a segfault. Affects all OpenVPN 2.0 versions.

ge 2.0 lt 2.0.4

David Sommerseth reports:

OpenVPN 2.1 until v2.4.12 and v2.5.6 may enable authentication bypass in external authentication plug-ins when more than one of them makes use of deferred authentication replies, which allows an external user to be granted access with only partially correct credentials. This issue is resolved in OpenVPN 2.4.12 and v2.5.6.

< 2.5.6

< 2.5.6

James Yonan reports:

If two or more client machines try to connect to the server at the same time via TCP, using the same client certificate, and when --duplicate-cn is not enabled on the server, a race condition can crash the server with "Assertion failed at mtcp.c:411"

< 2.0.1

James Yonan reports:

A format string vulnerability in the foreign_option function in options.c could potentially allow a malicious or compromised server to execute arbitrary code on the client. Only non-Windows clients are affected. The vulnerability only exists if (a) the client's TLS negotiation with the server succeeds, (b) the server is malicious or has been compromised such that it is configured to push a maliciously crafted options string to the client, and (c) the client indicates its willingness to accept pushed options from the server by having "pull" or "client" in its configuration file (Credit: Vade79).

ge 2.0 lt 2.0.4

Lev Stipakov and Gert Doering report:

There is a time frame between allocating peer-id and initializing data channel key (which is performed on receiving push request or on async push-reply) in which the existing peer-id float checks do not work right.

If a "rogue" data channel packet arrives during that time frame from another address and with same peer-id, this would cause client to float to that new address.

The net effect of this behaviour is that the VPN session for the "victim client" is broken. Since the "attacker client" does not have suitable keys, it can not inject or steal VPN traffic from the other session. The time window is small and it can not be used to attack a specific client's session, unless some other way is found to make it disconnect and reconnect first.

< 2.4.8_3

< 2.4.8_3

< 202016

The OpenVPN project reports:

OpenVPN 2.3.0 and earlier running in UDP mode are subject to chosen ciphertext injection due to a non-constant-time HMAC comparison function.

< 2.0.9_4

ge 2.1.0 lt 2.2.2_2

ge 2.3.0 lt 2.3.1

Samuli Seppänen reports:

In May/June 2017 Guido Vranken threw a fuzzer at OpenVPN 2.4.2. In the process he found several vulnerabilities and reported them to the OpenVPN project. [...] The first releases to have these fixes are OpenVPN 2.4.3 and 2.3.17.

This is a list of fixed important vulnerabilities:

• Remotely-triggerable ASSERT() on malformed IPv6 packet
• Pre-authentication remote crash/information disclosure for clients
• Potential double-free in --x509-alt-username
• Remote-triggerable memory leaks
• Post-authentication remote DoS when using the --x509-track option
• Null-pointer dereference in establish_http_proxy_passthru()

< 2.3.17

ge 2.4.0 lt 2.4.3

< 2.4.3

< 2.3.17

James Yonan reports:

DoS attack against server when run with "verb 0" and without "tls-auth". If a client connection to the server fails certificate verification, the OpenSSL error queue is not properly flushed, which can result in another unrelated client instance on the server seeing the error and responding to it, resulting in disconnection of the unrelated client.

< 2.0.1

Hendrik Weimer reports:

OpenVPN clients are a bit too generous when accepting configuration options from a server. It is possible to transmit environment variables to client-side shell scripts. There are some filters in place to prevent obvious nonsense, however they don't catch the good old LD_PRELOAD trick. All we need is to put a file onto the client under a known location (e.g. by returning a specially crafted document upon web access) and we have a remote root exploit. But since the attack may only come from authenticated servers, this threat is greatly reduced.

ge 2.0 lt 2.0.6

James Yonan reports:

If the client sends a packet which fails to decrypt on the server, the OpenSSL error queue is not properly flushed, which can result in another unrelated client instance on the server seeing the error and responding to it, resulting in disconnection of the unrelated client.

< 2.0.1

Gert Döring reports:

OpenVPN 2.5.1 and earlier versions allows a remote attackers to bypass authentication and access control channel data on servers configured with deferred authentication, which can be used to potentially trigger further information leaks.

< 2.5.2

< 2.5.2