SSH vulnerabilities

May 2026: This security advisory provides the details and fix information for a vulnerability that was discovered and fixed after the was disclosed in February 2026. This new advisory is for a new vulnerability in the control connection handshaking. The section of this advisory includes Show Control Connections guidance to help with system checks.  A vulnerability in the peering authentication in Cisco Catalyst SD-WAN Controller, formerly SD-WAN vSmart, and Cisco Catalyst SD-WAN Manager, formerly SD-WAN vManage, could allow an unauthenticated, remote attacker to bypass authentication and obtain administrative privileges on an affected system. This vulnerability exists because the peering authentication mechanism in an affected system is not working properly. An attacker could exploit this vulnerability by sending crafted requests to the affected system. A successful exploit could allow the attacker to log in to an affected Cisco Catalyst SD-WAN Controller as an internal, high-privileged, non-root user account. Using this account, the attacker could access NETCONF, which would then allow the attacker to manipulate network configuration for the SD-WAN fabric.

In the Linux kernel, the following vulnerability has been resolved: crypto: algif_aead - Revert to operating out-of-place This mostly reverts commit 72548b093ee3 except for the copying of the associated data. There is no benefit in operating in-place in algif_aead since the source and destination come from different mappings. Get rid of all the complexity added for in-place operation and just copy the AD directly.

OpenSSH before 10.3 mishandles the authorized_keys principals option in uncommon scenarios involving a principals list in conjunction with a Certificate Authority that makes certain use of comma characters.

OpenSSH before 10.3 omits connection multiplexing confirmation for proxy-mode multiplexing sessions.

OpenSSH before 10.3 can use unintended ECDSA algorithms. Listing of any ECDSA algorithm in PubkeyAcceptedAlgorithms or HostbasedAcceptedAlgorithms is misinterpreted to mean all ECDSA algorithms.

In OpenSSH before 10.3, command execution can occur via shell metacharacters in a username within a command line. This requires a scenario where the username on the command line is untrusted, and also requires a non-default configurations of % in ssh_config.

In OpenSSH before 10.3, a file downloaded by scp may be installed setuid or setgid, an outcome contrary to some users' expectations, if the download is performed as root with -O (legacy scp protocol) and without -p (preserve mode).

Vulnerability in the OpenSSH GSSAPI delta included in various Linux distributions. This vulnerability affects the GSSAPI patches added by various Linux distributions and does not affect the OpenSSH upstream project itself. The usage of sshpkt_disconnect() on an error, which does not terminate the process, allows an attacker to send an unexpected GSSAPI message type during the GSSAPI key exchange to the server, which will call the underlying function and continue the execution of the program without setting the related connection variables. As the variables are not initialized to NULL the code later accesses those uninitialized variables, accessing random memory, which could lead to undefined behavior. The recommended workaround is to use ssh_packet_disconnect() instead, which does terminate the process. The impact of the vulnerability depends heavily on the compiler flag hardening configuration.

A vulnerability allowing a low-privileged user to extract saved SSH credentials.

A vulnerability was determined in mkj Dropbear up to 2025.89. Impacted is the function unpackneg of the file src/curve25519.c of the component S Range Check. This manipulation causes improper verification of cryptographic signature. The attack can be initiated remotely. The attack is considered to have high complexity. The exploitability is considered difficult. The actual existence of this vulnerability is currently in question. Patch name: fdec3c90a15447bd538641d85e5a3e3ac981011d. To fix this issue, it is recommended to deploy a patch. The project maintainer explains: "Signature Malleability is not exploitable in SSH protocol. (...) [A] PoC doesn't exist for SSH implementation, but rather it's against the internal API."

An attacker may exploit the use of outdated and weak MAC algorithms in the device’s SSH service to potentially compromise the integrity of the SSH session, allowing manipulation of transmitted data if the attacker can interact with the network traffic.

An attacker may exploit the use of weak CBC-based cipher suites in the device’s SSH service to potentially observe or manipulate parts of the encrypted SSH communication, if they are able to intercept or interact with the network traffic.

In OpenClaw before 2026.2.23, tools.exec.safeBins validation for sort could be bypassed via GNU long-option abbreviations (such as --compress-prog) in allowlist mode, leading to approval-free execution paths that were intended to require approval. Only an exact string such as --compress-program was denied.

A vulnerability in Cisco Catalyst SD-WAN Software could allow an unauthenticated, remote attacker to view sensitive information on an affected system. This vulnerability is due to insufficient file system restrictions. An authenticated attacker with netadmin privileges could exploit this vulnerability by accessing the vshell of an affected system. A successful exploit could allow the attacker to read sensitive information on the underlying operating system.

A vulnerability in the Data Collection Agent (DCA) feature of Cisco Catalyst SD-WAN Manager could allow an unauthenticated, remote attacker to gain DCA user privileges on an affected system. This vulnerability is due to the presence of a credential file for the DCA user on an affected system. An attacker could exploit this vulnerability by sending a crafted HTTP request and reading the file that contains the DCA password from that affected system. A successful exploit could allow the attacker to access another affected system and gain DCA user privileges. Note: Cisco Catalyst SD-WAN Manager releases 20.18 and later are not affected by this vulnerability.

A vulnerability in the API of Cisco Catalyst SD-WAN Manager could allow an authenticated, remote attacker to overwrite arbitrary files on the local file system. To exploit this vulnerability, the attacker must have valid read-only credentials with API access on the affected system. This vulnerability is due to improper file handling on the API interface of an affected system. An attacker could exploit this vulnerability by uploading a malicious file on the local file system. A successful exploit could allow the attacker to overwrite arbitrary files on the affected system and gain vmanage user privileges.

BeyondTrust Remote Support (RS) and certain older versions of Privileged Remote Access (PRA) contain a critical pre-authentication remote code execution vulnerability. By sending specially crafted requests, an unauthenticated remote attacker may be able to execute operating system commands in the context of the site user.

Terraform / OpenTofu Provider adds support for Proxmox Virtual Environment. Prior to version 0.93.1, in the SSH configuration documentation, the sudoer line suggested is insecure and can result in escaping the folder using ../, allowing any files on the system to be edited. This issue has been patched in version 0.93.1.

OpenClaw is a personal AI assistant. Prior to version 2026.1.29, there is an OS command injection vulnerability via the Project Root Path in sshNodeCommand. The sshNodeCommand function constructed a shell script without properly escaping the user-supplied project path in an error message. When the cd command failed, the unescaped path was interpolated directly into an echo statement, allowing arbitrary command execution on the remote SSH host. The parseSSHTarget function did not validate that SSH target strings could not begin with a dash. An attacker-supplied target like -oProxyCommand=... would be interpreted as an SSH configuration flag rather than a hostname, allowing arbitrary command execution on the local machine. This issue has been patched in version 2026.1.29.

OpenClaw (aka clawdbot or Moltbot) before 2026.1.29 obtains a gatewayUrl value from a query string and automatically makes a WebSocket connection without prompting, sending a token value.

Soft Serve is a self-hostable Git server for the command line. Versions 0.11.2 and below have a critical authentication bypass vulnerability that allows an attacker to impersonate any user (including admin) by "offering" the victim's public key during the SSH handshake before authenticating with their own valid key. This occurs because the user identity is stored in the session context during the "offer" phase and is not cleared if that specific authentication attempt fails. This issue has been fixed in version 0.11.3.

wheel is a command line tool for manipulating Python wheel files, as defined in PEP 427. In versions 0.40.0 through 0.46.1, the unpack function is vulnerable to file permission modification through mishandling of file permissions after extraction. The logic blindly trusts the filename from the archive header for the chmod operation, even though the extraction process itself might have sanitized the path. Attackers can craft a malicious wheel file that, when unpacked, changes the permissions of critical system files (e.g., /etc/passwd, SSH keys, config files), allowing for Privilege Escalation or arbitrary code execution by modifying now-writable scripts. This issue has been fixed in version 0.46.2.

A vulnerability in the SSH service of Cisco IEC6400 Wireless Backhaul Edge Compute Software could allow an unauthenticated, remote attacker to cause the SSH service to stop responding. This vulnerability exists because the SSH service lacks effective flood protection. An attacker could exploit this vulnerability by initiating a denial of service (DoS) attack against the SSH port. A successful exploit could allow the attacker to cause the SSH service to be unresponsive during the period of the DoS attack. All other operations remain stable during the attack.

A vulnerability in the SSH server of TP-Link TL-WR820N v2.80 allows the use of a weak cryptographic algorithm, enabling an adjacent attacker to intercept and decrypt SSH traffic. Exploitation may expose sensitive information and compromise confidentiality.

SSH servers parsing GSSAPI authentication requests do not validate the number of mechanisms specified in the request, allowing an attacker to cause unbounded memory consumption.

A vulnerability in the SSH restricted shell interface of the network management services allows improper access control for authenticated read-only users. If successfully exploited, this vulnerability could allow an attacker with read-only privileges to gain administrator access on the affected system.

SSH clients receiving SSH_AGENT_SUCCESS when expecting a typed response will panic and cause early termination of the client process.

ssh in OpenSSH before 10.1 allows the '\0' character in an ssh:// URI, potentially leading to code execution when a ProxyCommand is used.

ssh in OpenSSH before 10.1 allows control characters in usernames that originate from certain possibly untrusted sources, potentially leading to code execution when a ProxyCommand is used. The untrusted sources are the command line and %-sequence expansion of a configuration file. (A configuration file that provides a complete literal username is not categorized as an untrusted source.)

SSH Tectia Server before 6.6.6 sometimes allows attackers to read and alter a user's session traffic.

A flaw was found in libssh versions built with OpenSSL versions older than 3.0, specifically in the ssh_kdf() function responsible for key derivation. Due to inconsistent interpretation of return values where OpenSSL uses 0 to indicate failure and libssh uses 0 for success—the function may mistakenly return a success status even when key derivation fails. This results in uninitialized cryptographic key buffers being used in subsequent communication, potentially compromising SSH sessions' confidentiality, integrity, and availability.

Teleport provides connectivity, authentication, access controls and audit for infrastructure. Community Edition versions before and including 17.5.1 are vulnerable to remote authentication bypass. At time of posting, there is no available open-source patch.

An OpenSSH daemon listens on TCP port 22. There is a hard-coded entry in the "/etc/shadow" file in the firmware image for the "root" user. However, in the default SSH configuration the "PermitRootLogin" is disabled, preventing the root user from logging in via SSH. This configuration can be bypassed/changed by an attacker through multiple paths though.

Erlang/OTP is a set of libraries for the Erlang programming language. Prior to versions OTP-27.3.3, OTP-26.2.5.11, and OTP-25.3.2.20, a SSH server may allow an attacker to perform unauthenticated remote code execution (RCE). By exploiting a flaw in SSH protocol message handling, a malicious actor could gain unauthorized access to affected systems and execute arbitrary commands without valid credentials. This issue is patched in versions OTP-27.3.3, OTP-26.2.5.11, and OTP-25.3.2.20. A temporary workaround involves disabling the SSH server or to prevent access via firewall rules.

In Ubuntu, gnome-control-center did not properly reflect SSH remote login status when the system was configured to use systemd socket activation for openssh-server. This could unknowingly leave the local machine exposed to remote SSH access contrary to expectation of the user.

In sshd in OpenSSH before 10.0, the DisableForwarding directive does not adhere to the documentation stating that it disables X11 and agent forwarding.

VyOS 1.3 through 1.5 (fixed in 1.4.2) or any Debian-based system using dropbear in combination with live-build has the same Dropbear private host keys across different installations. Thus, an attacker can conduct active man-in-the-middle attacks against SSH connections if Dropbear is enabled as the SSH daemon. I n VyOS, this is not the default configuration for the system SSH daemon, but is for the console service. To mitigate this, one can run "rm -f /etc/dropbear/*key*" and/or "rm -f /etc/dropbear-initramfs/*key*" and then dropbearkey -t rsa -s 4096 -f /etc/dropbear_rsa_host_key and reload the service or reboot the system before using Dropbear as the SSH daemon (this clears out all keys mistakenly built into the release image) or update to the latest version of VyOS 1.4 or 1.5. Note that this vulnerability is not unique to VyOS and may appear in any Debian-based Linux distribution that uses Dropbear in combination with live-build, which has a safeguard against this behavior in OpenSSH but no equivalent one for Dropbear.

A flaw was found in the OpenSSH package. For each ping packet the SSH server receives, a pong packet is allocated in a memory buffer and stored in a queue of packages. It is only freed when the server/client key exchange has finished. A malicious client may keep sending such packages, leading to an uncontrolled increase in memory consumption on the server side. Consequently, the server may become unavailable, resulting in a denial of service attack.

SSH servers which implement file transfer protocols are vulnerable to a denial of service attack from clients which complete the key exchange slowly, or not at all, causing pending content to be read into memory, but never transmitted.

A vulnerability was found in OpenSSH when the VerifyHostKeyDNS option is enabled. A machine-in-the-middle attack can be performed by a malicious machine impersonating a legit server. This issue occurs due to how OpenSSH mishandles error codes in specific conditions when verifying the host key. For an attack to be considered successful, the attacker needs to manage to exhaust the client's memory resource first, turning the attack complexity high.

SSH Communication Security PrivX versions between 18.0-36.0 implement insufficient validation on public key signatures when using native SSH connections via a proxy port. This allows an existing PrivX "account A" to impersonate another existing PrivX "account B" and gain access to SSH target hosts to which the "account B" has access.

Applications and libraries which misuse connection.serverAuthenticate (via callback field ServerConfig.PublicKeyCallback) may be susceptible to an authorization bypass. The documentation for ServerConfig.PublicKeyCallback says that "A call to this function does not guarantee that the key offered is in fact used to authenticate." Specifically, the SSH protocol allows clients to inquire about whether a public key is acceptable before proving control of the corresponding private key. PublicKeyCallback may be called with multiple keys, and the order in which the keys were provided cannot be used to infer which key the client successfully authenticated with, if any. Some applications, which store the key(s) passed to PublicKeyCallback (or derived information) and make security relevant determinations based on it once the connection is established, may make incorrect assumptions. For example, an attacker may send public keys A and B, and then authenticate with A. PublicKeyCallback would be called only twice, first with A and then with B. A vulnerable application may then make authorization decisions based on key B for which the attacker does not actually control the private key. Since this API is widely misused, as a partial mitigation golang.org/x/cry...@v0.31.0 enforces the property that, when successfully authenticating via public key, the last key passed to ServerConfig.PublicKeyCallback will be the key used to authenticate the connection. PublicKeyCallback will now be called multiple times with the same key, if necessary. Note that the client may still not control the last key passed to PublicKeyCallback if the connection is then authenticated with a different method, such as PasswordCallback, KeyboardInteractiveCallback, or NoClientAuth. Users should be using the Extensions field of the Permissions return value from the various authentication callbacks to record data associated with the authentication attempt instead of referencing external state. Once the connection is established the state corresponding to the successful authentication attempt can be retrieved via the ServerConn.Permissions field. Note that some third-party libraries misuse the Permissions type by sharing it across authentication attempts; users of third-party libraries should refer to the relevant projects for guidance.

SSH authenticated user when access the PAM server can execute an OS command to gain the full system access using bash. This issue affects Privileged Access Manager before 3.7.0.1.

A signal handler in sshd(8) may call a logging function that is not async-signal-safe. The signal handler is invoked when a client does not authenticate within the LoginGraceTime seconds (120 by default). This signal handler executes in the context of the sshd(8)'s privileged code, which is not sandboxed and runs with full root privileges. This issue is another instance of the problem in CVE-2024-6387 addressed by FreeBSD-SA-24:04.openssh. The faulty code in this case is from the integration of blacklistd in OpenSSH in FreeBSD. As a result of calling functions that are not async-signal-safe in the privileged sshd(8) context, a race condition exists that a determined attacker may be able to exploit to allow an unauthenticated remote code execution as root.

A Cross-Site Scripting vulnerability in Roundcube through 1.5.7 and 1.6.x through 1.6.7 allows a remote attacker to steal and send emails of a victim via a crafted e-mail message that abuses a Desanitization issue in message_body() in program/actions/mail/show.php.

A race condition vulnerability was discovered in how signals are handled by OpenSSH's server (sshd). If a remote attacker does not authenticate within a set time period, then sshd's SIGALRM handler is called asynchronously. However, this signal handler calls various functions that are not async-signal-safe, for example, syslog(). As a consequence of a successful attack, in the worst case scenario, an attacker may be able to perform a remote code execution (RCE) as an unprivileged user running the sshd server.

OpenSSH 9.5 through 9.7 before 9.8 sometimes allows timing attacks against echo-off password entry (e.g., for su and Sudo) because of an ObscureKeystrokeTiming logic error. Similarly, other timing attacks against keystroke entry could occur.

A security regression (CVE-2006-5051) was discovered in OpenSSH's server (sshd). There is a race condition which can lead sshd to handle some signals in an unsafe manner. An unauthenticated, remote attacker may be able to trigger it by failing to authenticate within a set time period.

Improper Authentication vulnerability in Progress MOVEit Transfer (SFTP module) can lead to Authentication Bypass.This issue affects MOVEit Transfer: from 2023.0.0 before 2023.0.11, from 2023.1.0 before 2023.1.6, from 2024.0.0 before 2024.0.2.

sshproxy is used on a gateway to transparently proxy a user SSH connection on the gateway to an internal host via SSH. Prior to version 1.6.3, any user authorized to connect to a ssh server using `sshproxy` can inject options to the `ssh` command executed by `sshproxy`. All versions of `sshproxy` are impacted. The problem is patched starting in version 1.6.3. The only workaround is to use the `force_command` option in `sshproxy.yaml`, but it's rarely relevant.