2026-11-05 14:13 UTC
Replies (1)
-
@m0xEE@breloma.m0xee.net 2026-11-05 15:28
@theorytoe@ak.kyaruc.moe I believe almost unscathed. Even in case with well-known vulnerabilities, building a reliable practical exploit is hardly feasible because of re-linking on each reboot, safe default allocator with address randomisation, etc. A couple of days ago I ran into a story of a person who started patching their systems, but then realised that it might not be that bad: https://portal.mozz.us/gemini/gemini.clehaxze.tw/gemlog/2026/11-01-openbsd-is-right-on-mitigations-in-light-on-recent-wave-of-serious-vulnerabilities.gmi Or direct Gemini link: gemini://gemini.clehaxze.tw/gemlog/2026/11-01-openbsd-is-right-on-mitigations-in-light-on-recent-wave-of-serious-vulnerabilities.gmi That is also why I feel safer hosting my stuff on "weird" machines — even in case with recent widely publicised AF_ALG root escalation in Linux, the proof of concept code was targeting x86_64, it simply won't work on a 32-bit Big Endian PowerPC machine :marseylaughwith: If there was someone motivated enough to target my very opinionated setups, I'd be fucked either way — whether I install updates or not. I mean I try to patch my systems, but I'm usually not too eager. This might be one of the few cases when security by obscurity might actually work — those willing to exploit a known vulnerability, would likely be targeting binaries with known offsets shipped by major distros, not your Gentoo system optimised for a particular microarchitecture.