I swear some systems setup advice on the internet is literally written by intelligence agencies as a psyop One of these methods for generating a LUKS (full disk encryption) password is not like the other....
one honk maybe more
joelh@infosec.exchan..
replied 26 Jan 2026 09:27 +0000
in reply to: https://benjojo.co.uk/u/benjojo/h/nN53M9Wqk5SxB4J2c7
@benjojo It reads like an LLM, where the model may have been concerned that the system didn’t have OpenSSL so provided an alternative, and the author had no idea and just copy pasta for AdSense revenue.
benjojo
replied 26 Jan 2026 10:16 +0000
in reply to: https://infosec.exchange/users/joelh/statuses/115960685686420582
@joelh this was written way before LLMs even entered the scene, Rather than an artificial intelligence, I think this is just natural stupidity
willpower232@phpc.so..
replied 26 Jan 2026 21:17 +0000
in reply to: https://benjojo.co.uk/u/benjojo/h/nN53M9Wqk5SxB4J2c7
@benjojo I like how the rubbishness of those password generation options outweighs the truly wild choice of lastpass
benjojo
replied 26 Jan 2026 21:24 +0000
in reply to: https://phpc.social/users/willpower232/statuses/115963476986256411
@willpower232 to give them some credit, the article has a date of 2012 and a lot of stuff (oh my God a lot) has changed since then The date derived password is still insane though
varx@infosec.exchang..
replied 27 Jan 2026 00:35 +0000
in reply to: https://benjojo.co.uk/u/benjojo/h/nN53M9Wqk5SxB4J2c7
@benjojo I was shocked to discover that this came out of Percona's docs. ...I had a much better impression of them before this moment.
benjojo
replied 27 Jan 2026 09:52 +0000
in reply to: https://infosec.exchange/users/varx/statuses/115964258504967676
@varx right??? I still assume that database product is fine. Just maybe they should stick around to maintaining the databases rather than maintaining the Linux systems around them