curl / Docs / curl CVEs / invalid URL parsing with '#'


invalid URL parsing with '#'

Project curl Security Advisory, November 2, 2016 - Permalink


curl doesn't parse the authority component of the URL correctly when the host name part ends with a '#' character, and could instead be tricked into connecting to a different host. This may have security implications if you for example use a URL parser that follows the RFC to check for allowed domains before using curl to request them.

Passing in would wrongly make curl send a request to while your browser would connect to given the same URL.

The problem exists for most protocol schemes.


The Common Vulnerabilities and Exposures (CVE) project has assigned the name CVE-2016-8624 to this issue.

CWE-172: Encoding Error

Severity: Medium


This flaw exists in the following curl versions.

libcurl is used by many applications, but not always advertised as such!


In version 7.51.0, the parser function is fixed.

As a side-effect of this fix, using the # character as part of the user or password fields in the URL is no longer supported. According to RFC 3986 section 2.3 it isn't allowed. See issue #1216


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

A - Upgrade curl and libcurl to version 7.51.0

B - Apply the patch to your version and rebuild

C - Strip out the parts of the URLs containing '#' before passing them to curl


It was first reported to the curl project on October 10.

We contacted distros@openwall on October 19.

curl 7.51.0 was released on November 2 2016, coordinated with the publication of this advisory.


Thanks a lot!