327

cat (lemmy.zip)

submitted 1 month ago by Maven@lemmy.zip to c/programmer_humor@programming.dev

22 comments fedilink hide all child comments

you are viewing a single comment's thread
view the rest of the comments

[-] dan@upvote.au 21 points 1 month ago

id_rsa shouldn't exist any more.

[-] bjoern_tantau@swg-empire.de 38 points 1 month ago

If I had a nickel for every time I had to change my ssh key algorithm I'd have two nickels.

Which isn't much but it's concerning that it happened twice.

[-] friend_of_satan@lemmy.world 24 points 1 month ago* (last edited 1 month ago)

A few days ago I was messing with my ubiquiti dream router and its ssh config option said the key should start with ssh-rsa 🙄

[-] dan@upvote.au 19 points 1 month ago

It probably accepts other key types and it's just the UI that's outdated. I doubt they're using an SSH implementation other than Dropbear or OpenSSH, and both support ed25519.

[-] dbx12@programming.dev 2 points 1 month ago

Could be stupid input validation which requires ^ssh-rsa

[-] Scoopta@programming.dev 11 points 1 month ago

Fact of the matter is RSA is perfectly secure still...and ECDSA/ED25519 should also be extinct given the rising need for post quantum cryptography

[-] Supermariofan67@programming.dev 16 points 1 month ago

The problem is not the RSA math itself but that it is both extremely slow and implementing it is particularly susceptible to bugs and side channel attacks https://blog.trailofbits.com/2019/07/08/fuck-rsa/

[-] Scoopta@programming.dev 1 points 1 month ago

Most of the situations I encounter RSA are in projects where I hope RSA is implemented correctly. I have a lot of Let's Encrypt certs that are still RSA and my main SSH keys are still RSA. All of these were generated quite some time ago. I understand the problem with projects that implement it incorrectly but I'd hope OpenSSH and certbot aren't those projects 😥