Hello friends,

Unfortunately we found another flaw with a potential security inpact and here
is the fully story:

                    libcurl URL decode buffer boundary flaw
                    =======================================

Project cURL Security Advisory, June 22nd 2013
http://curl.haxx.se/docs/security.html

1. VULNERABILITY

   libcurl is vulnerable to a case of bad checking of the input data which may
   lead to heap corruption.

   The function curl_easy_unescape() decodes URL encoded strings to raw binary
   data. URL encoded octets are represented with %HH combinations where HH is a
   two-digit hexadecimal number. The decoded string is written to an allocated
   memory area that the function returns to the caller.

   The function takes a source string and a length parameter, and if the length
   provided is 0 the function will instead use strlen() to figure out how much
   data to parse.

   The "%HH" parser wrongly only considered the case where a zero byte would
   terminate the input. If a length-limited buffer was passed in which ended
   with a '%' character which was followed by two hexadecimal digits outside of
   the buffer libcurl was allowed to parse alas without a terminating zero,
   libcurl would still parse that sequence as well. The counter for remaining
   data to handle would then be decreased too much and wrap to become a very
   large integer and the copying would go on too long and the destination
   buffer that is allocated on the heap would get overwritten.

   We consider it unlikely that programs allow user-provided strings unfiltered
   into this function. Also, only the not zero-terminated input string use case
   is affected by this flaw. Exploiting this flaw for gain is probably possible
   for specific circumstances but we consider the general risk for this to be
   low.

   The curl command line tool is not affected by this problem as it doesn't use
   this function.

   There are no known exploits available at this time.

   The Common Vulnerabilities and Exposures (CVE) project has assigned the name
   CVE-2013-2174 to this issue.

2. AFFECTED VERSIONS

   Affected versions: from libcurl 7.7 to and including 7.30.0
   Not affected versions: libcurl before 7.7 and >= 7.31.0

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

3. THE SOLUTION

   libcurl 7.31.0 implements a proper check that the following hexdigits are
   within the provided input length.

4. RECOMMENDATIONS

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

   A - Upgrade to curl and libcurl 7.31.0

   B - Apply this patch and rebuild libcurl

       http://curl.haxx.se/libcurl-unescape.patch

   C - Double-check your curl_easy_unescape() usage (for example use it with
       the 'length' argument set to 0), or avoid using that function

5. TIME LINE

   Vulnerability found by Timo Sirainen. Patched by Daniel Stenberg.

   It was reported to the curl project on May 19th 2013. We contacted
   linux-distros on June 12th.

   curl 7.31.0 was released on June 22nd 2013, coordinated with the
   publication of this advisory.

6. CREDITS

   Reported by Timo Sirainen. Thanks a lot!

-- 
  / daniel.haxx.se
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-users
FAQ:        http://curl.haxx.se/docs/faq.html
Etiquette:  http://curl.haxx.se/mail/etiquette.html