On Fri, 31 May 2024 08:50:02 -0700 Kent Borg <kentb...@borg.org> wrote:
> But how in the hell could a compromise of xz put a backdoor into > sshd‽‽ Because systemd patches sshd…because systemd. It didn't. There is no vulnerability in OpenSSH. There is no vulnerability in OpenSSH patched to work with systemd's logging facilities. The vulnerability lies in systemd's use of xz. OpenSSH is the vector used to invoke the back door embedded in xz. I'm oversimplifying things, because the "simple" description is anything but simple: https://x.com/fr0gger_/status/1774342248437813525 https://www.linkedin.com/posts/rekunkel_great-infographic-about-the-xz-outbreak-activity-7180237206685409281-ITXL And in fact, systemd was about to *remove* the xz dependency when the backdoor was discovered. It's possible that this announcement caused the actors behind the backdoor to accelerate their plans, which in turn may have contributed to its discovery. -- \m/ (--) \m/ _______________________________________________ Discuss mailing list Discuss@driftwood.blu.org https://driftwood.blu.org/mailman/listinfo/discuss
