Further Reading: Using the Audit Trail
1. The Setup Audit Trail: Tracking Policy Changes
This tool tracks “Who changed the security settings?” It is your best defense against accidental (or malicious) changes to your session security.
- What it tracks: Changes to MFA requirements, modifications to Login IP Ranges, and permission set assignments.
- The Scenario: You set up a strict Login IP Range for the “Waffles Over Everything” Org, but a week later, it’s gone.
- How to use it:
- Go to Setup > Security > View Setup Audit Trail.
- Download the last six months of data as a
.csv. - Search for: keywords like “Login IP,” “MFA,” or “Permission Set.”
- The Result: You can see exactly which admin deleted the IP range and at what time, allowing for a post-mortem coaching session.
2. Login History: Tracking Session Behavior
While the Setup Audit Trail tracks policy changes, the Login History tracks user behavior. This is where you find “Impossible Travelers” or credential stuffing attempts.
- What it tracks: Every login attempt (Success or Failure), the Source IP, the Browser/Platform, and the Login Type (e.g., Application, Service, or SAML for SSO).
- How to use it to secure sessions:
- Go to Setup > Users > Login History.
- Filter by Status: Look for “Invalid Password” or “Failed: IP Restricted.” A high volume of these for a single user suggests a brute-force attack.
- Check MFA Challenges: Look for the “Identity Verification” status. If a user has 50 “Success” logins but 0 “Identity Verification” challenges, your MFA policy might not be applied to them correctly.
- Identify Old Protocols: Look for logins using outdated TLS versions or browsers. You can use this data to force users to upgrade for a more secure session.
3. Login Forensics: The “Advanced” Audit
For Admins who need more than just a list of IPs, Login Forensics (part of Event Monitoring) provides deeper insights into session security without needing a full Shield license.
- The Benefit: It doesn’t just show where they logged in; it identifies anomalies.
- What it spots:
- Average Login Volume: If a user typically logs in 5 times a day but suddenly logs in 500 times, the system flags it.
- Suspicious Timeframes: Logins occurring at 3:00 AM for a user who only works 9-to-5.