The security of an organization’s network is one of the most important things to prioritize. It’s crucial that it’s confidentiality, integrity, and availability are maintained. However, one of the most common security failures within an organization is its network security, despite the potential impact of a network security incident. Even though there are many ways to fall into a pit when it comes to network security, hope is not lost, thankfully. This blog post aims to bring light to some common pitfalls when it comes to organizational network security.
Network Device Misconfigurations
What makes up the network? Well, to put it simply, you have network devices, and you have endpoints. While everyone and their cat is focussing on endpoint security, it’s very easy to overlook some commnon network device misconfigurations.
Weak Credentials
Everyone always says it, but having weak credentials is a really good way to let somebody into the network that isn’t supposed to be. Ensure that all credentials used to login to devices follow a couple guidelines:
- Minimum 12 characters in length
- Complex (mix of upper and lowercase letters, numbers, and special characters)
- Is NOT easily guessable if someone knows attributes of you or your organization
- Example: If your organization’s name is
DEVCORP
, don’t make a passwordD3vC0Rp!
- Example: If your organization’s name is
Insecure Protocols and Encryption
Something not often thought of is what protocols are being used in the network. Protocols are the key to allowing device communication within a network, so ensuring that only secure protocols are used should be high priority for any organization.
Below is a table for general recommendations for network protocols. However, depending on business operational requirements, protocols such as NTLM may be required for compatability reasons. As such, use your best judgement when determining what protocols to allow on your networks.
Do | Do Not |
---|---|
SMBv3 | SMBv1, SMBv2 |
Kerberos (NTLMv2 if required) | NTLM (at least NTLMv1) |
HTTPS | HTTP |
SNMPv3 | SNMPv1, SNMPv2 |
SSHv2 | SSHv1, Telnet |
SFTP, FTPS | FTP, TFTP |
LDAPS | LDAP |
In general, organizations should steer clear of protocols that offer insecure, little, or no encryption. It’s up to the organization to determine what risks they’re willing to undertake. While the protocols mentioned above are generally good ideas to implement, they aren’t silver bullets. For example, attackers can still use Kerberos for lateral movement inside an Active Directory environment.
In general, however, I’d recommend completely abstaining from the following protocols
- LLMNR
- mDNS
- NetBIOS
Access Controls, What Access Controls?
Access controls are extremely important within any organization, but how does it apply to network security? Well, as the name suggests, you’re controlling access to resources on the network! One thing to keep in mind, however, is that it’s all a balancing act - you have to ensure that those who need access have it, and those who don’t need access don’t. Furthermore, how difficult it is to access a system is important. Sure, having 4 different forms of authentication would make it more difficult for an attacker to access a system, but it also hinders the speed at which authorized access is attained.
Least Privilege
The principle of least privilege is one of the most important things to implement to keep a network secure. Dave in accounting should not be able to use his account to log into network devices and make configuration changes. Thus, ensuring that only a select few people are able to access specific network resources should be highly prioritized.
Network Access Control (NAC)
Network Access Control (NAC) is an extremely potent tool in an organization’s network security toolbox. As Cisco puts it in their post regarding NAC:
A NAC system can deny network access to noncompliant devices, place them in a quarantined
area, or give them only restricted access to computing resources, thus keeping insecure nodes from infecting
the network.
Thus, having these controls significantly increases network security simply by allowing granular control over what is allowed to access it. If there’s a compromised endpoint, having the ability to quarantine it can prove to significantly hinder an attacker’s ability to penetrate the network and grant precious time to defenders.
Provisioning and Segmentation Issues
With everything said, probably one of the best way to prevent attackers from laterally moving in a network is simply to segment it - whether that be through VLANs, air gapping, or something else.
VLAN’s…What’re Those?
Although Virtual Local Area Networks (VLANs) were first described in 1998, they are still one of - if not - the most common method to segment a network. Don’t think that putting everything on one vlan will save you from a potential attack, however. Using VLANs still requires thought into how a network should be broken up into logical chunks. For example, it’s common to split a network into pieces based off of use-case. There might be vlans for POS, IOT, and even guest devices!
One thing that I would strongly advise when it comes to thinking out what VLANs to have is to ensure that device management interfaces are put onto their own seperate VLAN. As an example, having a VLAN specifically so that you can manage your switch management interfaces can be crucial to ensuring a secure network environment.
Addresses Addresses Everywhere, Not Any Bit Distinct
Subnetting is one of the first things taught in any network class, and is a foundational concept in any good network. You may be asking, why is this being talked about to secure a network? Well, first off, as you may know from your networking classes, unless it’s configured to, devices in different subnets can’t communicate with each other (at least at layer 3). While this is a fairly trivial control to get around, it can prove to hinder adversarial advances into a network.
As previously stated, subnetting is a relatively weak security control. However, one thing that
subnetting does really well is allowing defenders to better understand the context of a device
given an IP address. If an organization has an IP addressing scheme that incorporates information
into the address, then just having an IP address of a compromised host may greatly assist in
defensive action. For example, it is relatively common to include information such as a physical
location ID or VLAN ID into an octet of an IP address to better distinguish it from others. While
technologies such as DNS allow for translations of IP addresses into common names, the more
information that can be determined from small pieces of information is vital to quick response.
Furthermore, DNS can only translate IP address into common names such as translating 192.168.1.1
into FW-1.local
. Should an organization include VLAN IDs into an IP address, having 192.168.10.1
would allow a defender to immediately recognize the originating VLAN for a potential attack.
General
Storing Passwords in Insecure Places
Perhaps one of the oldest tricks in the book when it comes to a network compromise is storing
credentials in an insecure place. Do NOT have a passwords.txt
files with the password for
anything! One suggestion for solving this issue is to use a trusted password manager to store
credentials. Although this will solve the glaring insecurities of insecure credential
placement, it will not save you from one of the weakest links in cyber security - people.
Having proper user training is vital to bettering organizational security posture.
Updates People, Updates!
A very low hanging fruit when it comes to network security is ensuring that devices connecting to or facilitating the network are properly updated. That isn’t to say that you need to immediately update a device as soon as an update comes out, however, as it is not unheard of for software developers to push faulty updates leading to a whole host of problems.
Ultimately, it is up to the organization to determine what policy they want to have for pushing updates to devices. Having the latest security patch applied as soon as possible may be the best choice for an organization, while some may choose to wait until updates are proven to work in customer environments.
What’s a Documentation?
Writing documentation is useless, right? After all, we have John Doe who’s been here for 30 years and he knows everything! Well, John Doe may be a good source of information, but if something happens to him, nobody may know enough to maintain critical business operations. Suddenly John’s “duck tape and superglue” approach isn’t working anymore, and nobody has any idea how to fix it. Well, that’s why documentation is so vital. Having a centralized knowledge base is so important within an organization, especially for storing environment-specific information. Despite what is commonly said, you’re not going to find everything on the internet, so when something happens to your environment, it’s important to have specific documentation to hopefully fix it.
Now, how does that relate to network security? Well, having documentation may not necessarily help much in terms of active security posture, but it can significantly assist defenders in having all of the details of the network at their disposal in the event of an incident. In fact, one of the biggest advantages that defenders should have over a given adversary is knowledge regarding all of - if not - most of the network. This can be achieved through proper documentation.
Limited Device and Network Visibility
One of the worst things that can happen to an organization in terms of responding to an incident is when they have little to no visibility into what’s going on with their network or network devices. Thankfully, there’s a lot that can be done to solve this glaring issue. Below is a list of possible solutions to gain increased insight into network/device activities:
- SNMP monitoring (Use SNMPv3 please!)
- Uptime
- Temperature
- Interface errors and throughput
- CPU utilization
- Memory usage
- Logging
- Configuration and state changes
- Login and logoff events
I would also strongly recommend using some sort of SIEM or log aggregation solution to better aggerate and search through logs generated by network devices. Here are a few examples
- Elastic
- Splunk
- Graylog
Lastly, having a way to easily create a security dashboard for your network devices can provide a great way to visually monitor network devices at scale. For this I would recommend using tools like Prometheus and Grafana.
Too Complex
Once you learn a lot about networking it’s very easy to fall into the trap of overcomplicating a network. Remember, you may be trying to secure a network, but you’re also supporting business operations, so balance is crucial. Sombody may say that they want a gigantic network complete with SDWAN, EIGRP, BGP, and HSRP, but that may be too complicated for the organization, and thus places a greater strain on the network and security team(s). There may not always be a choice, but please do your best to balance simplicity, complexity, and security to best secure business operations.
Security Devices, but Where’s the Training?
“We spend $500,000 dollars on AI driven security appliances! How did we still get hacked?”
One of the worst pitfalls in network security is to not train the people who are defending your network. Even though security appliances these days tend to be very good at what they do, having people skilled enough to use them is just as - if not - more important that the security appliances themselves. Investing in properly educating employees on the tools being used in the organization’s environment is crucial to having an excellent security posture.
Conclusion
Network security remains one of the most important areas of security. Although I’ve provided a lot of general recommendations for network security, it’s still up to you to decide how you want to go about implementing them.
Remember, balance is key when it comes to implementing these.