Daniel Kahn Gillmor <d...@fifthhorseman.net> writes: > Thanks for this review, Russ! Can you give a more detailed breakdown of > these keys? for example, at least algorithm choice and size? (iiuc, > all PGP-2 keys are RSA keys, but i don't think their sizes are > constrained).
This is for all keys, just just for the obsolete keys. I suspect the last two lines are all modern keys. 2 pub 512R 1 pub 768R 7 pub 1024D 67 pub 1024R 1 pub 1535R 1 pub 2047R 18 pub 2048R 3 pub 4096R I'm happy to provide more detailed information but I don't know the flags to gpg1 very well, so I'm not sure what would produce useful information. Most of these are not in active use, and I don't object to using this as a driving force to do a bunch of spring cleaning and tell hierarchy administrators they need to generate new keys if they want their control messages to still be honored. > I don't know enough about how Usenet uses these keys, but I think > they're only relevant for continued use if they involve decryption. They are exclusively used to sign and verify control messages using the pgpverify protocol [1]. Some of these old PGP-2 keys are still in active use to sign newly-issued control messages because getting sites to update the keys is hard, and Usenet is very low on resources. Usenet keys are basically never used for encryption or decryption. Part of the problem with convincing people to upgrade is that this isn't a very high-security problem and it's not horribly difficult to correct for any attacks. It's very unlikely that anyone would spend thousands of dollars to forge a Usenet control message. I'm dubious that anyone would even spend $100. (Those 512-bit RSA keys may be even cheaper to compromise than that at this point, though. I haven't kept up with the state of the art.) Still, we should modernize. (I issued a new Big Eight key and am dual-issuing control messages now with both the old and new key, and plan to continue dual-issuing control messages until the software to issue signatures with the old key is no longer supported.) [1] https://www.eyrie.org/~eagle/usefor/other/pgpverify I kept meaning to write an RFC but never got around to it. -- Russ Allbery (r...@debian.org) <https://www.eyrie.org/~eagle/>