Solution
The attacker learned the usernames from the login form. It answered “user does not exist” or “incorrect password” depending on the input, so failed attempts revealed whether a username was valid.
What happened?
The platform first looked up the username, then checked the password if the account existed. The error differed between stages. This flaw is known as username enumeration .
Given the public names and the naming pattern, the attacker could generate candidates. For María García, those ranged from mgarcia00 to mgarcia99.
A programme submitted each candidate with an arbitrary password. Burp Intruder , part of Burp Suite, can automate requests and compare responses.
“User does not exist” ruled out a candidate. “Incorrect password” confirmed a real username.
With dozens of teachers and a hundred numbers per person, a few thousand attempts were enough, matching the failures in the logs.
The attacker then made the phishing emails more convincing by including each teacher’s real username.
The difference can also be in the HTTP status or the response time. If the application checks the password only for existing users, those requests may take longer.
How could it be fixed and prevented?
To respond , warn the teachers, block the source of the attempts and identify anyone who entered their password on the fake site so they can reset it.
To prevent a repeat:
Make failures look the same. Use a generic error such as “Incorrect username or password”, with the same HTTP status and similar timing. The server can also perform a password check when no user was found.
Apply the rule across forms. Registration and password recovery can also leak account names. A recovery form can always say “If the account exists, you’ll receive an email.”
Limit attempts. Set limits per account and per IP address and alert on large numbers of failures.
Use multifactor authentication to reduce the risk if someone obtains a password through phishing.
‹
Case 11/11
›
A project by Eniun