4 ms·
The use-case for DNSSEC is still far from clear to me. DANE is the closest thing to a valid use case and it looks like Chrome and others are exploring better so
by davidu 11y ago
The use-case for DNSSEC is still far from clear to me. DANE is the closest thing to a valid use case and it looks like Chrome and others are exploring better solutions than DANE.
Moreover -- The lack of encryption is still a major issue IMHO. Validation is great, except that when a record fails to validate the end user experience is essentially NXDOMAIN which is not a very friendly user experience.
At least with SSL certificate failures, the application has visibility into the cause of the failure. In DNSSEC, it's hard to distinguish between a failed validation and any other kind of DNS resolution failure.
I just can't see a world where people enable validating resolvers that reject non-validating responses. And if that's true, then the entire point of DNSSEC is largely compromised -- with the possible exception of using DNSSEC to validate in-line verifications (a la DANE).
- zurn 11y agoYou can do all kinds of nice DANE-like things. For example keys for SSH, IPSec, OpenVPN, NFS, etc. "We already have something for web browsers" isn't a big counterargument. I agree they should add confidentiality. That shouldn't be too hard since all the key handling is already taken care of.