According to https://bugzilla.mozilla.org/show_bug.cgi?id=2083449 they actually did use an HSM? Though some of the jargon is a bit beyond me, maybe this will finally get me to learn more PKI stuff.

> The hardware used for each of the issuers is in the attached file. These HSMs are FIPS 140-2 Level 3 compliant.

▲

oasisbob 3 hours ago | parent | next [-]

My vague and inexperienced read on their description boils down to "we were doing something weird and non-standard as part of our key ceremony because we intended these keys to be exported, and because of that process weirdness we blundered and gave the HSM faulty instructions."

▲

zimmerfrei 3 hours ago | parent | prev [-]

The most likely explanation is that they generated the weak key externally, imported it into the HSM, generated the CSR from it, and finally had Root CA sign it.

A FIPS HSM does not forbid one to import keys into it.

▲

cstdr 2 hours ago | parent [-]

But shouldn't such a HSM support key attestation (and the parent require such attestation)? Attestation would also track whether a key was imported or generated on the HSM, and of sufficient length.

▲

zimmerfrei an hour ago | parent [-]

Attestation is not required by the compliance standard for browser CAs (WebTrust) and it is also quite vendor-specific and not very interoperable, so it is not commonly done.

The Root CA must sign requests after authentication, but a "manual" approval is also accepted if done with the appropriate oversight with witnesses (clearly not the case here).

	▲	cstdr an hour ago \| parent [-]
		Of course it's not required, as we see here. The question is why not? Getting a simple code-signing cert seems to have more stringent requirements in that it either requires attestation or CA-issued hardware.