Content deleted Content added
m Hyperlink some RFCs |
Artoria2e5 (talk | contribs) →History: KeyTrap, oops. |
||
Line 183:
Unfortunately, the IETF RFC 2535 specification had very significant problems scaling up to the full Internet; by 2001 it became clear that this specification was unusable for large networks. In normal operation DNS servers often get out of sync with their parents. This isn't usually a problem, but when DNSSEC is enabled, this out-of-sync data could have the effect of a serious self-created denial of service. The original DNSSEC required a complex six-message protocol and a lot of data transfers to perform key changes for a child (DNS child zones had to send all of their data up to the parent, have the parent sign each record, and then send those signatures back to the child for the child to store in a SIG record). Also, public key changes could have absurd effects; for example, if the ".com" zone changed its public key, it would have to send 22 million records (because it would need to update all of the signatures in all of its children). Thus, DNSSEC as defined in RFC 2535 could not scale up to the Internet.
The IETF fundamentally modified DNSSEC, which is called ''DNSSEC-bis'' when necessary to distinguish it from the original DNSSEC approach of RFC 2535. This new version uses "delegation signer (DS) resource records" to provide an additional level of indirection at delegation points between a parent and child zone. In the new approach, when a child's master public key changes, instead of having six messages for every record in the child, there is one simple message: the child sends the new public key to its parent (signed, of course). Parents simply store one master public key for each child; this is much more practical. This means that a little data is pushed to the parent, instead of massive amounts of data being exchanged between the parent and children. This does mean that clients have to do a little more work when verifying keys. More specifically, verifying a DNS zone's KEY RRset requires two signature verification operations instead of the one required by RFC 2535 (there is no impact on the number of signatures verified for other types of RRsets). Most view this as a small price to pay, since it makes DNSSEC deployment more practical. The new version is published in RFC4033-4035.
In January 2024, a "KeyTrap" denial-of-service attack was announced for all specification-respecting DNSSEC resolvers. The DNSSEC specification (RFC4033-4035) specifies that a resolver, when receiving a signed packet from the upstream, should try ''all'' keys with the correct "tag" on ''all'' signatures until one of the combinations successfully verifies. By putting many keys with the same "tag" and many signatures corresponding to that "tag" in a packet, the researchers can slow down a resolver by a factor of 2 million. In response, resolvers began to place limits on the amount of verification errors, key tag collisions, and hash calculations.<ref>{{cite web |title=The KeyTrap Denial-of-Service Algorithmic Complexity Attacks on DNS Version: January 2024 |author1=Elias Heftrig|author2= Haya Schulmann |author3= Niklas Vogel |author4= Michael Waidne |url=https://www.athene-center.de/fileadmin/content/PDF/Keytrap_2401.pdf |website=ATHENE}} ([https://www.athene-center.de/en/keytrap press release])</ref>
==Authenticating NXDOMAIN responses and NSEC==
| |||