CVE-2019-6477

TCP-pipelined queries can bypass tcp-clients limit

Published Nov 26, 2019 CVSS 3.1 · 7.5 High EPSS 4.1% KEV · Not listed
Description
With pipelining enabled each incoming query on a TCP connection requires a similar resource allocation to a query received via UDP or via TCP without pipelining enabled. A client using a TCP-pipelined connection to a server could consume more resources than the server has been provisioned to handle. When a TCP connection with a large number of pipelined queries is closed, the load on the server releasing these multiple resources can cause it to become unresponsive, even for queries that can be answered authoritatively or from cache. (This is most likely to be perceived as an intermittent server problem).

Remediation

Vendor Solution

Upgrade to the patched release most closely related to your current version of BIND: BIND 9.11.13 BIND 9.14.8 BIND 9.15.6 BIND Supported Preview Edition is a special feature preview branch of BIND provided to eligible ISC support customers. BIND 9.11.13-S1 Note that the fix for CVE-2019-6477 addresses only the server memory leak issue. TCP-pipelining may still malfunction by dropping some responses on a TCP connection where a client query pattern generates excessive outstanding queries, but the malfunction will affect that TCP connection alone and will not cause any degradation of service to other clients. An affected client connection might also appear to hang, but will clear when either the client or the server initiates a close or reset and will not remain in that state indefinitely. Disabling TCP-pipelining entirely is completely effective at mitigating the vulnerability with minimal impact to clients that use pipelined TCP connections and with no impact to clients that do not support TCP-pipelining. The majority of Internet client DNS queries are transported over UDP or TCP without use of TCP-pipelining.


Vendor Workaround

The vulnerability can be avoided by disabling server TCP-pipelining: keep-response-order { any; }; and then restarting BIND. The server restart is necessary because neither a 'reload' nor a 'reconfig' operation will properly reset currently pipelining TCP clients.

Tracking

Sign in to view the affected projects.

Advisories

Source ID Title
Debian DSA Debian DSA DSA-4689-1 bind9 security update
EUVD EUVD EUVD-2019-16037 With pipelining enabled each incoming query on a TCP connection requires a similar resource allocation to a query received via UDP or via TCP without pipelining enabled. A client using a TCP-pipelined connection to a server could consume more resources than the server has been provisioned to handle. When a TCP connection with a large number of pipelined queries is closed, the load on the server releasing these multiple resources can cause it to become unresponsive, even for queries that can be answered authoritatively or from cache. (This is most likely to be perceived as an intermittent server problem).
Ubuntu USN Ubuntu USN USN-4197-1 Bind vulnerability
History

No history.

Weaknesses

Sources

Status
PUBLISHED
Assigner
isc
Published
Updated
Sep 16, 2024, 4:47 PM
Reserved
Jan 16, 2019, 12:00 AM
Status
Modified
Published
Nov 26, 2019, 4:15 PM
Modified
Jun 17, 2026, 2:39 AM
Severity
Moderate
Public date
Nov 20, 2019, 12:00 AM