Monday, November 25, 2013

Daily Blog #155: Sunday Funday 11/24/13 Winner!

Hello Reader,
     Another Sunday Funday come and gone and some good entries this week!I really liked this weeks winner because he went into both physical access control (badge logs), network access controls (firewall logs), and host access controls (event logs and artifacts) to narrow down his suspects and fully examine the facts at hand. Great job Steve M!

The Rules:
  1. You must post your answer before Monday 11/25/13 2AM CST (GMT -5)
  2. The most complete answer wins
  3. You are allowed to edit your answer after posting
  4. If two answers are too similar for one to win, the one with the earlier posting time wins
  5. Be specific and be thoughtful 
  6. Anonymous entries are allowed, please email them to
  7. In order for an anonymous winner to receive a prize they must give their name to me, but i will not release it in a blog post

The Challenge:
Your board of directors have received an email from a Gmail address sent from Thunderbird mailer at 9pm at night with insider information about the company with a demand for action or the sender will go to the press.  IT security has found the IP address of the companies firewall of one of the smaller company branches in the email header and passed the data to you. The branch has only 8 employees and normal office hours end at 5pm.

Please detail how you will:
1. Determine which system sent the email
2. Determine which user of the system sent the email

The Winning Answer:
Steve M
Considering this case may involve litigation, I would stress collecting good case notes and following proven processes throughout the investigation.

To initially narrow the search of suspects, I would first check physical access badge logs to determine if anyone was in the office at 9pm.  If employees can access the Internet via this firewall when connected via VPN, I would check VPN access logs as well.

On the network side, I would begin by looking at the perimeter firewall traffic logs for mail connections to Google/gmail, specifically on ports 993/TCP (IMAP), 995/TCP (POP3), and 25/TCP or 587/TCP (SMTP) which would be used by Thunderbird.  Google may switch services between various IP ranges, but as of right now it appears most mail related addresses resolve within the subnet.  It is probably a safe assumption to say Google won't move their mail services outside of this network/port range in the foreseeable future.  Once I have identified an internal IP address as communicating to Google over these ports during the window the email was received, I would look for DHCP logs or an asset inventory system to determine which internal system had the IP address at that time.

Once I have identified a suspect internal system, I would proceed with host level forensics to find the user with that account's information in his/her Thunderbird profile (assuming Win7):

- First, I would take an image of the system for investigation and evidence purposes.  I would also duplicate it for a working copy.

- I would sweep the system for directories matching "C:\Users\\AppData\Roaming\Thunderbird\Profiles\", to see which Windows logon accounts use Thunderbird.

- At this point, I would perform a raw text search of the entire contents of this directory (including "ImapMail" and "Mail") for the specific source email address in question.  If hits were found, I would focus the remainder of my investigation on this profile.

- If no hits were found, I may have to extract the usernames from all Thunderbird profiles on this system.  Considering both usernames and passwords can be decrypted with this method, I would NOT do this on a large number of potentially unrelated profiles due to privacy concerns.  Instead I would revisit my case notes to see if I can pinpoint the suspected user via other means (event log records and other filesystem activity within the same time period for example).

- Once I've identified a profile with hits, I would extract the profile folder from the disk for further investigation.  Mozilla stores the usernames and passwords encrypted in the "moz_logins" table of the "signons.sqlite" sqlite db and the encryption keys in the "key3.db" file.

- I would use a tool such as "ThunderbirdPassDecryptor" to decrypt the username/password from this profile using these files.  At that point, the source email address (and password) should be known, and the owner of the Windows profile can be considered the source.