Testing for CCS Injection

Question

I'm trying to test for a CCS Injection. I have found two prevailant methods to do so

1. Use nmap's ssl-ccs-injection nse script
2. Tripwire's OpenSSL-CCS-Inject-Test

Both are pretty simple to use, however, they are giving me different results. nmap's script gives me a STATUS: Vulnerable while Tripwire's script gives me a "No need to patch" message. What do I believe/what do I do?

Many thanks for the help.

Answer 1

In order to exploit the vulnerablity, a MITM attacker would effectively do the following:

o Wait for a new TLS connection, followed by the ClientHello ServerHello handshake messages.

o Issue a CCS packet in both the directions, which causes the OpenSSL code to use a zero length pre master secret key. The packet is sent to both ends of the connection. Session Keys are derived using a zero length pre master secret key, and future session keys also share this weakness.

o Renegotiate the handshake parameters.

o The attacker is now able to decrypt or even modify the packets in transit.

The script works by sending a 'ChangeCipherSpec' message out of order and checking whether the server returns an 'UNEXPECTED_MESSAGE' alert record or not. Since a non-patched server would simply accept this message, the CCS packet is sent twice, in order to force an alert from the server. If the alert type is different than 'UNEXPECTED_MESSAGE', we can conclude the server is vulnerable.

Source: https://nmap.org/nsedoc/scripts/ssl-ccs-injection.html

You can use openssl s_client in linux for sending packets to the server and confirming the vulnerability manually.

Moreover I will suggest you to use testssl.sh available on https://testssl.sh because the tool is way more comprehensive and accurate.