If named is configured to use Response Policy Zones (RPZ) an error processing some rule types can lead to a condition where BIND will endlessly loop while handling a query.

A server is potentially vulnerable to degradation of service if
    1. the server is configured to use RPZ,
    2. the server uses NSDNAME or NSIP policy rules, and
    3. an attacker can cause the server to process a specific query

Successful exploitation of this condition will cause named to enter a state where it continues to loop while processing the query without ever reaching an end state. While in this state, named repeatedly queries the same sets of authoritative nameservers and this behavior will usually persist indefinitely beyond the normal client query processing timeout. By triggering this condition multiple times, an attacker could cause a deliberate and substantial degradation in service.

Versions affected: 9.9.10, 9.10.5, 9.11.0->9.11.1, 9.9.10-S1, 9.10.5-S1

Workaround:

Only the NSDNAME and NSIP RPZ rule types can cause this condition to occur. You can work around this vulnerability if you are able to express your desired policy while avoiding NSDNAME or NSIP rules, otherwise it is advised that you upgrade to a version which corrects the defect.

References:

https://kb.isc.org/article/AA-01495/74/CVE-2017-3140

Upstream patch:

https://source.isc.org/cgi-bin/gitweb.cgi?p=bind9.git;a=commit;h=2648c49be78