CVE-2018-5745

If a BIND resolver sends a query for a DNSSEC-signed authoritative zone, and the authoritative server replies with a valid wildcard answer and signed NSEC3 proof, followed by an unsigned NSEC at the same owner name, it will trigger an unexpected program exit. This issue affects BIND 9 versions 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.11.3-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.

In BIND 9, a validly signed NSEC3 record from an unrelated sibling zone may be accepted as an insecurity proof. This downgrades a secure delegation and allows a forged unsigned answer to pass through.

If BIND is loaded with a named.conf file that contains no global options block, an attacker can send a query of QTYPE TKEY which may cause an assertion failure and subsequent unexpected program exit. This issue affects BIND 9 versions 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, and 9.20.9-S1 through 9.20.27-S1.

In a query response, an attacker may send named multiple copies of a record that should only exist once (such as an SOA record). If the RDATA is identical on all copies, the record is appended to the in-memory RDATA set, increasing negative cache memory usage.

A BIND recursive resolver may experience excessive resource consumption if it encounters large numbers of a particular kind of invalid DNSSEC record. Default limits on max-records-per-type and max-types-per-name help mitigate the exposure.

On a resolver configured to use dns64, if an applicable answer from the authoritative server is malformed in a specific way, the resolver named process will exit unexpectedly. This issue affects BIND 9 versions 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.11.3-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.

For a secondary zone with transfers restricted by TSIG, named may start to serve the data provided in a zone transfer before the final message with the TSIG signature arrives. This could allow an attacker without a valid TSIG signature to send unauthorized zone contents to a secondary server, and named does not roll back to the pre-transfer state.

If a BIND resolver has cached a tree of SVCB/HTTPS AliasMode records, and is then queried for the root of that tree, the resolver will spend disproportionate CPU time constructing the response. This issue affects BIND 9 versions 9.18.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.18.11-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.

A BIND resolver encountering an SVCB/HTTPS AliasMode record referencing 14 or more SVCB/HTTPS ServiceMode records may fail to properly deallocate internal resources. If this happens repeatedly, resource exhaustion will eventually prevent the resolver from performing new recursive lookups. This issue affects BIND 9 versions 9.18.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.18.11-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.

A malformed zone may contain an NS or DNAME node above its origin, which `named` treats as a zone cut. If an attacker inserts a malformed zone into a BIND authoritative server (e.g., via zone transfer), queries for names inside the configured zone then lose authoritative status and return an out-of-zone delegation. On a server that also provides recursion BIND can follow this locally sourced cut and cache attacker-supplied data, affecting names outside the configured zone. This situation persists as long as the malformed zone remains in the zone database. This issue affects BIND 9 versions 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, 9.21.0 through 9.21.25, 9.11.3-S1 through 9.18.50-S1, and 9.20.9-S1 through 9.20.27-S1.