[tptacek]

Which is exactly what happened to Slack, and took them offline for most of a business day for a huge fraction of their customers. This is such a big problem that there's actually a subsidiary DNSSEC protocol (DNSSEC NTA's) that addresses it: tactically disabling DNSSEC at major resolvers for the inevitable cases where something breaks.

[indolering]

As if DNS isn't a major contributing to A LOT of downtime. That doesn't mean it's not worth doing not investing in making deployment more seamless and less error prone.

[growse]

> As if DNS isn't a major contributing to A LOT of downtime. That doesn't mean it's not worth doing not investing in making deployment more seamless and less error prone.

Ah yes. Let's take something that's prone to causing service issues and strap more footguns to it.

It's not worth it, because the cost is extremely quantifiable and visible, whereas the benefits struggle to be coherent.

[indolering]

The benefits are huge: there are lots of attacks that DNSSEC trivially prevents and it would help secure more than just web browsers.

[ekr____]

Can you expand on this a bit, under the assumption that the traffic is using some form of transport security (e.g., TLS, SSH, etc.)?

[indolering]

DNS underlies domain authority and the validity of every connection to every domain name ultimately traces back to DNS records. The amount of infra needed to shore up HTTPS is huge and thus SSH and other protocols rely on trust-on-first-use (unless you manually hard-code public keys yourself - which doesn't happen). DNS offers a standard, delegable PKI that is available to all clients regardless of the transport protocol.

With DNSSEC, a host with control over a domain's DNS records could use that to issue verifiable public keys without having to contact a third party.

I ran into this while working on decentralized web technologies and building a parallel to WebPKI just wasn't feasible. Whereas we could totally feed clients DNSSEC validated certs, but it wasn't supported.

[ekr____]

Thanks for the explanation. It seems like there are two cases here:

1. Things that use TLS and hence the WebPKI 2. Other things.

None of what you've written here applies to the TLS and WebPKI case, so I'm going to take it that you're not arguing that DNSSEC validation by clients provides a security improvement in that case.

That leaves us with the non-WebPKI cases like SSH. I think you've got a somewhat stronger case there, but not much of one, because those cases can also basically go back to the WebPKI, either directly, by using WebPKI-based certificates, or indirectly, by hosting fingerprints on a Web server.

[twelvedogs]

The difference is DNS provides a fairly obvious up side

[gzread]

Actually, does it? Yes, the obvious upside when I type in slack.com instead of 123.45.56.67 is very good. Does this same upside apply to addresses I don't type in? What's actually the advantage of addressing one of foobarcorp's infinitude of servers uasing the string "123-45-57-78.slp05.mus.foobar.com" instead of "123.45.57.78"? It seems to just waste bytes. And most communication is of the latter sort - an app talking to its own servers managed by the same company.

[themafia]

BGP can be hijacked. Anycast IPs exist. Rolling out a new release when one of your IPs is unavailable could be a severe challenge. SVC records are actually kinda neat.

[gzread]

All of that's a problem with DNS too, even updating the IP. You could still use it to get the initial entry point if you wanted. But when you serve a webpage with an automatically generated pointer to image3.yourdomain, the only reason not to make that an IP is HTTPS, and LE just started issuing IP address certificates. Think about it - it saves a few round trips.

If the IP is anycast, all the better.