curl / Docs / Security Problems / NTLM password overflow via integer overflow

NTLM password overflow via integer overflow

Project curl Security Advisory, September 5th 2018 - Permalink

VULNERABILITY

libcurl contains a buffer overrun in the NTLM authentication code.

The internal function Curl_ntlm_core_mk_nt_hash multiplies the length of the password by two (SUM) to figure out how large temporary storage area to allocate from the heap.

The length value is then subsequently used to iterate over the password and generate output into the allocated storage buffer. On systems with a 32 bit size_t, the math to calculate SUM triggers an integer overflow when the password length exceeds 2GB (2^31 bytes). This integer overflow usually causes a very small buffer to actually get allocated instead of the intended very huge one, making the use of that buffer end up in a heap buffer overflow.

(This bug is almost identical to CVE-2017-8816.)

We are not aware of any exploit of this flaw.

INFO

This bug was introduced in commit be285cde3f, April 2006.

The Common Vulnerabilities and Exposures (CVE) project has assigned the name CVE-2018-14618 to this issue.

CWE-131: Incorrect Calculation of Buffer Size

AFFECTED VERSIONS

This issue is only present on 32 bit systems. It also requires the password field to use more than 2GB of memory, which should be rare.

curl is used by many applications, but not always advertised as such.

THE SOLUTION

In libcurl version 7.61.1, the integer overflow is avoided.

A patch for CVE-2018-14618 is available.

RECOMMENDATIONS

We suggest you take one of the following actions immediately, in order of preference:

A - Upgrade curl to version 7.61.1

B - Apply the patch to your version and rebuild

C - Put length restrictions on the password you can pass to libcurl

TIME LINE

It was publicly reported to the curl project on July 18, 2018. We contacted distros@openwall on August 27.

curl 7.61.1 was released on September 5 2018, coordinated with the publication of this advisory.

CREDITS

Reported by Zhaoyang Wu. Patch by Daniel Stenberg.

Thanks a lot!