On Thursday, 29 October 2026 05:38:35 CET 3hn92a--- via tor-relays wrote: > What about the -sk version of mldsa44-ed25519? In the case of NitroKey 3, OnlyKey, and SoloKey, the hardware unfortunately does not support these cryptographic operations. The computation of ML-DSA signatures and the storage of massive post-quantum keys (exceeding 1.3 KB) overwhelm the memory and cryptographic architecture of current FIDO2 tokens. I am waiting for the new NitroKeys and OnlyKeys. They will also have large hardware-based password storage. (similar to the old Signet HC) For now, I am happy with ed25519-sk (non-resident = Default) and ed25519-sk_rk (resident) keys. Additionally: In nftables, SSH is permitted only for a few IP addresses. GrapheneOS + PGPPony¹ can do post-quantum cryptography. ¹OpenKeychain successor https://pgpony.app/guides/migrate-openkeychain.php > On Wed, 28 Oct 2026, George via tor-relays wrote: > > Since the topic has come up again recently, I put up a short guide on > > some best practices for OpenSSH. > > > > https://nycbug1.nycbug.org/openssh.html > > > > Note that as of OpenSSH 10.3, the first quantum computing-resistant key > > cipher is available, mldsa44-ed25519, and if both the client and server > > versions are capable, you should replace your keys with the new cipher. > > (https://www.openssh.org/releasenotes.html#10.3) > > > > Comments offline appreciated, but the guide is meant to be short and > > actionable. > > > > George -- ╰_╯ Ciao Marco! Debian GNU/Linux It's free software and it gives you freedom!
Attachment:
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ tor-relays mailing list -- tor-relays@xxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to tor-relays-leave@xxxxxxxxxxxxxxxxxxxx