This update for openssh fixes the following issues:
- CVE-2025-61984: Fixed code execution via control characters in usernames when a ProxyCommand is used (bsc#1251198)
- CVE-2025-61985: Fixed code execution via '\0' character in ssh:// URI when a ProxyCommand is used (bsc#1251199)
This update for openssh fixes the following issues:
- Security issues fixed:
* CVE-2025-32728: Fixed a logic error in DisableForwarding option (bsc#1241012)
- Other bugs fixed:
* Allow KEX hashes greater than 256 bits (bsc#1241045)
* Fixed hostname being left out of the audit output (bsc#1228634)
* Fixed failures with very large MOTDs (bsc#1232533)
This update for openssh fixes the following issues:
- CVE-2025-26465: Fixed MitM attack against OpenSSH's VerifyHostKeyDNS-enabled client (bsc#1237040).
This update for openssh fixes the following issues:
- CVE-2023-51385: Limit the use of shell metacharacters in host- and
user names to avoid command injection. (bsc#1218215)
This update for openssh fixes the following issues:
- CVE-2023-48795: Fixed prefix truncation breaking ssh channel integrity (bsc#1217950).
the following non-security bug was fixed:
- Fix the 'no route to host' error when connecting via ProxyJump
This update for openssh fixes the following issues:
- CVE-2023-38408: Fixed a condition where specific libaries loaded via
ssh-agent(1)'s PKCS#11 support could be abused to achieve remote code
execution via a forwarded agent socket if those libraries were present on the
victim's system and if the agent was forwarded to an attacker-controlled
system. [bsc#1213504, CVE-2023-38408]
- Close the right filedescriptor and also close fdh in read_hmac to avoid file
descriptor leaks. [bsc#1209536]
- Attempts to mitigate instances of secrets lingering in memory after a session
exits. [bsc#1186673, bsc#1213004, bsc#1213008]