CERT NZ's critical controls 2019

This is CERT NZ’s second annual list of ten critical controls for organisations. These controls would prevent, or better contain, the majority of attacks we’ve seen in the past year.

There are many controls that could improve an organisation’s information security. We created this list to help you prioritise your controls based on the threats and incident data we’ve seen. The incident data came from reports to CERT NZ, and our international threat feeds.

To reflect the changes we’ve seen over the last year, the list for 2019 includes two new controls:

  • implement network segmentation, and
  • manage cloud authentication.

These replace two of the controls we highlighted in 2018 — removing legacy systems and managing BYOD devices. Removing these controls from our list doesn't mean they're not still important. Rather, it shows the changing nature of the threat environment. You can still find details of these controls in our guides section.

Removing legacy systems

Managing BYOD devices

We recommend you also continue with your own best practices, like maintaining an effective password policy.

Download a copy of CERT NZ's critical controls 2019 [PDF, 572 KB]

This year’s critical controls

In the coming months, we’ll publish more details on the controls for 2019. For each control, we'll provide:

  • a page summarising the intent and success measures for business owners, and
  • a separate page with implementation advice for practitioners.

1. Enforce multi-factor authentication (MFA)

Credential dumps and credential harvesting attacks are common. They give attackers access to large numbers of usernames and passwords. Protect your business systems and data by enabling MFA on all privileged or remote access systems. This includes:

  • VPNs
  • administrative consoles
  • webmail
  • published applications like Citrix.

We saw several phishing campaigns focused on credential harvesting in 2018. One example was the Office365 campaign. In the cases we saw, enabling MFA would have prevented unauthorised access to the accounts with leaked credentials.

 woman standing at whiteboard

2. Patch your software

Keep software, like operating systems and applications, up-to-date. It's one of the most simple and effective steps you can take to secure your environment.

We've seen many organisations attacked by malware that exploits known vulnerabilities. Applying patches would have helped them avoid these attacks.

3. Disable unused services and protocols

Keeping your systems up-to-date isn’t always enough to keep attackers away. Older services and protocols often have their own vulnerabilities. Leaving them on your network gives attackers more opportunity to breach your network. To mitigate this, scan your network for services and protocols that are:

  • no longer used, or
  • known to be vulnerable.

If you identify any, carry out remediation based on your findings. The recent WannaCry incident demonstrated what can happen when attackers exploit out-of-date protocols.

4. Change default credentials

Security is sometimes overlooked in the rush to get new technology into production. A key step to take for any new application or device is to change or remove all default credentials. This will prevent an attacker accessing your network with known usernames and passwords.

We continue to see organisations compromised by attackers using unchanged default credentials.

5. Implement and test backups

Backups are critical for recovering from incidents like ransomware. Store your backups offline, and test them often. Organisations often need to restore data from their latest backup in response to threats like ransomware.

We've seen organisations lose data and incur significant operational costs because they didn't have up-to-date, well-maintained backups.

6. Implement application whitelisting

Two of the most common ways to infect a user’s workstation with malware are through:

  • email clients, and
  • web browsers.

To prevent this, identify a list of applications that your users need. Make sure they can only execute approved applications.

Most malware incidents reported to CERT NZ are likely to have originated from:

  • opening malicious email attachments, or
  • drive-by downloads.

Whitelisting the approved applications will help protect the system from these attacks. It's a key security control for your network.

man and woman behind screen

7. Enforce the principle of least privilege

Grant users the minimum level of access and control in your network that they need to do their job. Remove their accounts when they're no longer needed. This will limit the damage that intrusions into your network can cause. We also recommend enforcing separation of privilege. When a user requires administrative privileges, use a separate account.

We're aware of incidents where users held unnecessary administrative privileges. Attackers were able to exploit their accounts to make unauthorised changes to the environment.

8. Configure centralised logging and analysis

Storing and securing your logs in a central place makes log analysis and alerting easier. Logs are a key part of understanding what happened in an incident. Configuring alerts for key actions can help you:

  • detect abnormal behaviour
  • tell you what to investigate.

Without good logging, it’s very difficult to discover the nature and extent of a compromise. This makes your efforts to contain and recover from an incident much harder.

Logs weren't available for many of the incidents reported to CERT NZ. This meant it wasn't possible to do a complete post-incident investigation.

9. Implement network segmentation

Proper network segmentation relies on the implementation of other critical controls, in particular:

  • disabling unused services and protocols, and
  • enforcing the principle of least privilege.

We've seen incidents where attackers used common management tools and protocols to gain control of other machines on a network. There are also tools scripted to get credentials. The credentials are then used to access other devices and applications in a network.

You can prevent attackers spreading through your network by:

  • using network tools like firewalls
  • following the other critical controls.

10. Manage cloud authentication

We're aware of incidents where cloud authentication misconfigurations let attackers bypass security controls. They do this by using legacy authentication protocols. Organisations are also moving toward using more cloud-based services. It's easy to end up in a situation where you have multiple authentication systems.

Centralising authentication gives you better control and visibility over who has access to your systems and information. It also:

  • provides a unified experience
  • lets you configure MFA for applications that may not support it.