CISSP Blog Post 26, Domain 8: Malware

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Finally! Congratulations on making it to the last CISSP post in this series. Today we will cover the most exciting topic of Malware! Here are some of the most common types of malware you need to know about:

1. Viruses are malicious code that replicate by creating, replacing, or attacking other programs or files. Viruses generally require some initiating action by the user. Virus Types include File Infectors and Boot Sector Infectors (which are read before the host operating system is started)
2. Worms are a malicious and continuous process that reproduces and eats up resources. Generally it does not require an initiating action by the user. They spread over networks by exploiting vulnerabilities in network protocols, or through application components (e.g. DLLs, etc.) Unlike viruses, worms do not require using infected files to spread (i.e. viruses require a file “host”).
3. Trojans are installed by a user because they think they want it. They are a form of social engineering.
4. Remote Access Tools aka RAT’s can be legitimate remote administration tool but they can also be an illegitimate remote access trojan.
5. Rootkits are often trojans or other malware that can replace critical system files or interfere with system kernel functions to seize control of a processor’s central ring (0 or 1) such that a whole system is compromised.
6. A Logic Bomb is malicious code, often planted by someone you know (i.e. an insider programmer) that is triggered by an event or specific schedule. Usually as an act of revenge.
7. Botnets are where multiple systems are compromised and turned into agents / bots / zombies.
8. Distributed Denial of Service (DDoS) attacks have 3 phases:
   1. Attacker infects many machines with agents (aka bots or zombies)
   2. Attacker uses a Master / Handler program to command agents
   3. Agents initiate denial of service or SPAM attack against attacker’s target ISPs and managed DNS can help stop a DDoS attack.
9. Zero-Day Exploits / Malware are attacks that take place shortly after a security vulnerability is discovered but before a vendor has a fix or patch available.

So how do you protect against Malware? Malware tools come with different types of capabilities including:

• Known Signature Scanning – the program scans based on known malware or attack signatures (e.g. Antivirus). These solutions are only as good as known, available signatures.
• Heuristic Scanning – the program looks for suspicious system behavior or activity. It does NOT use baseline learning, it only uses predefined rules.
• Change Detection Tools look for unauthorized changes to files, system configuration, or programs (e.g. File Integrity Monitoring solutions). These tools take baseline snapshots of files (via a file hash) and then creates new hashes periodically to see if they change.

CISSP Blog Post 25, Domain 8: Program Exploits

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Ok! Last domain of the CISSP – we have two topics to cover – we’ll cover one today and one next week.

In this post, we’ll cover some common Program Exploits at a high level to get you familiar with terminology.

The first is Memory Buffer Overflow an example is a website form running on a server where the attacker enters a longer string than the program that ingests the form can allow, causing the memory on the server to overflow which can corrupt data, crash the system, or provide access to things the attacker should not have access to. To fix, the programmer must put in validation checking for fields in the website form.

Covert Channel is a secret transfer or sharing of information that violates security. Examples could be a Covert Storage Channel which is a hidden data storage location, or hidden data that an attacker shouldn’t be accessing within a program. A Covert Timing Channel is secret signaling. For example, using screen flicker to exfiltrate data from a facility.

Cross-Site Scripting is a well known attack and is where a malicious user puts comments with a malicious script in a web form. A regular user then picks up content when they load the website in their browser and the comment causes the regular user’s browser to execute the script. This for example, could be used to harvest cookies. A user can safeguard against this by disabling scripting in their browser.

Cross-Site Request Forgery is similar where a user has two browser tabs open. In Tab 1 they might have an image with a reference link or a script with a request action on a specific banking site. In Tab 2, they may have open their banking site. The browser may allow a transaction or activity from Tab 1 to occur on the website in Tab 2 believing it’s legitimate because it’s occurring within the same browser.

Memory or Object Reuse is where you need to sanitize media before reusing it with a protected audit log trail.

Trapdoors / Back-doors / Maintenance Hooks are hidden mechanisms for bypassing access controls. They are put in by programmers – typically for convenience when debugging their code.

SQL Injection is where a front-end form passes input containing SQL code that runs on a back-end database and returns output or runs code. For example, if someone put the following into a "First Name" field of a web form: “Bobby ‘DROP TABLE”, it could cause the database to delete a table from the database if there are no validation checks or neutralization of form entries to cause them to not execute.

A Race Condition Attack is where two signals or processes race each other to influence the output first. A physical representation of this would be two joint bank account owners trying to make a withdrawal from the account at the same time. If the combination of both their withdrawals is larger than the account, the bank may not realize that they have overdrawn their account and allow both of them to complete their withdrawals at the same time.

Time of Check / Time of Use (TOCTOU) is where variables of a system are changed but there’s a delay in when the system honors the changes. The example here is changing your account permissions may not take affect right away or require you to refresh your browser, etc.

CISSP Blog Post 24, Domain 7: Disaster Planning and Restoration

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Ok, you’ve (hopefully) backed up your data in the last post, so now let’s talk Disaster Planning and restoration.

Some unforeseen factors when planning that you should be aware of may include:

• Your backup site is also impacted by the disaster
• You cannot get to your backup site
• If you have a hot site, it may not be able to accommodate multiple customers all having an issue at the same time
• Your employee’s families may need care as well, reducing available support staff

So what’s the goal of restoration? The goal is to return to your original site with original capacity and data.

Recovery vs. Restoration

Restoration phases include:

1. Is the incident ended?
2. Is it safe to return?
3. Document the losses
4. Salvage the assets
5. Repairs & replacement
6. Return to site (Tier 5 first, all the way up to Tier 1 support employees)
7. Closure – lessons learned, official end of disaster

Pro tip: When you’re documenting your plan put a 1-year expiration date on the plan to force updates and make it obvious which is most recent version.

Speaking of version control – obsolete plans should be:

1. Archived
2. Collected
3. Confirm collection
4. Issue new plan
5. Destroy old plans

Oh… also you – need to be testing your plan. You can do so in multiple ways including:

Testing Type	Method
Checklist or Desk Check	Give each business unit (BU) a copy of the plan and have them run through a checklist to ensure all relevant points are covered.
Structured Walk-through / Tabletop Exercise	Key players get together and review plan collectively.
Simulation Test	Practice drill mobilizing the personnel (e.g. Fire Drill) and rehearse going to assembly point.
Parallel Test	Operational test at alternate site running in parallel with main site (production).
Full Interruption Test	Shutdown production environment and run a live environment at alternate site. Need to have prior management written permission before parallel test conducted.

CISSP Blog Post 23, Domain 7: Digital Backups

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Welcome to February 2021! This month I plan to wrap up our CISSP blog post series.

Let’s start by talking about data backups! There are lots of ways to do data backups:

• A Full backup is exactly what it sounds like – all your data is copied to another location and backed up.
• A Differential backup is where all data that’s changed since the last full backup is copied.
• An Incremental backup is where all data that has changed since the last full OR incremental backup, is backed up. This is easier to restore from but you will need more time and media storage space.

Cost and Capability Comparison of Backup Sites

Some technology that can be useful for creating backups is a Redundant Array of Independent Disks (aka RAID). Again, lots of choices:

RAID 0 – Stripping of data – very fast, no recovery! 2 drives minimum required.

RAID 0

RAID 1 – Mirroring – double storage cost, slower, 2 drives minimum required.

RAID 1

RAID 3 & 4 – RAID 3 reads and writes data at the byte level. RAID 4 reads and writes at the block level. You can only lose 1 active drive at a time. If the parity drive fails, the RAID falls back to RAID 0 or you can rebuild the parity drive back on a spare drive. This requires 3 drives minimum and only gives 2 drive capacity.

RAID 3 & 4

RAID 5 is faster because parity info is written in parallel. If there is no spare drive, it will reconstruct lost data and parity info into system memory in chunks. It needs 3 drives minimum, with a 2 drive capacity.

RAID 6 (Enhanced RAID 5) provides 2-dimensional parity, allowing for the loss of 2 drives simultaneously. It needs 1 extra drive than a RAID 5. Requires 4 drives minimum, with a 2 drive capacity.

RAID 5 & 6

You can also combine RAID’s: e.g. 0+1, 1+0, 1+5, 5+1, etc.

CISSP Blog Post 22, Domain 5: Identity and Access Management (IAM)

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

In this post, we are going to cover some basic terminology, then review different types of multi-factor authentication, review Biometric authentication, talk a little bit about federated identity management and then end with a summary review of different types of access controls.

So, starting with terminology:

• Identification is to announce oneself to a system or a facility.
• Authentication is to verify users of a system or a facility
• Authorization is the act of enforcing permissions for each user
• Accountability is to track a user’s activities

Users authenticate based on one or more of the following factors:

1. Something they know (a password)
2. Something they have (an ID)
3. Something they are or do (biometrics)

To be Two Factor Authentication (2FA), a user must use 2 out of 3 of the above. Multi-Factor Authentication (MFA) means a user can use 2 of the same type of factor listed above (i.e. MFA could mean that you just have to enter two passwords, or a password and a PIN).

Tokens can be used to create one-time passwords. A token is also known as a session password or a dynamic password vs. a static, reusable, or fixed password.

There are two types of tokens: Asynchronous Tokens

Asynchronous Token Process

And Synchronous Tokens (typically based on time):

Synchronous Token Process

So now let’s talk about biometrics – i.e. “What you are”. There are two types of Biometrics:

1. Physiological / static biometrics measure a unique, personal feature such as your fingerprints.
2. Behavioral / dynamic biometrics measures something you perform that’s unique to you such as your voice, hand signature, keystroke dynamics, walking gait, etc.

When comparing biometric systems, we typically look at Error Rates and again, there are two types:

1. Type I Error (aka False Reject rate error) is when a system rejects the correct/valid user.
2. Type II Error (aka False Acceptance rate error) is when a system accepts the incorrect/invalid person.

A trick to help you remember Type II is that it is FAR worse II let someone incorrect in!

The Crossover Error Rate is a percentage calculation measuring the accuracy of a biometric device. The smaller the percentage, the lower the false positive rate and the more accurate the device.

Biometric Crossover Error Rate

Federated Identity Management (FIM, not to be confused with File Integrity Monitoring) is the capability of managing a user’s identity across multiple, distinct, identity management systems. An example is using your Google login information to authenticate with Pintrest.

A Federated Login is a form of Single Sign On (SSO) dealing with internet services. Examples include OpenID, PayPal, Facebook, Google, etc.

Security Assertion Markup Language (SAML) is an XML based format for exchanging security information for SSO. It only provides a message format to authenticate a user and must be used with protocols that perform SSO. Again, examples include OpenID, Kerberos, Facebook, etc. SAML is a browser based component for SSO, not a system for SSO itself.

Oauth (Open Standard for Authentication) is a standard allowing 3rd party websites to gain access to resources without exchanging username and password if both sides support Oauth. The 3rd party website requests a token from the website that holds the user’s resources and if the user authorizes the transaction, a temporary access token is issued to the requesting website. An example of this is when Google maps requests access to your Uber account to book a ride.

As promised, here’s a summary of different types of Access Controls – note this is NOT an exhaustive list.

Control	Description
Mandatory Access Control (TCSEC Level B1)	Access is determined by an Access control policy, Classification labels (for objects), and Clearance labels (for subjects)
Discretionary (TCSEC C1)	Access is determined by the owner of the asset
Non-Discretionary	Access is maintained by the custodian / system
Rule Based	Access is based on policies
Role Based	Access is based on job functions
Content Dependent	Access is based on what needs to be performed on database records
Time Based	Access is based on schedules
Constrained User Interface	Deliberately restrict access based on conditions such as: Age, Location, Ability to Pay, etc.

CISSP Blog Post 21, Domain 4: Network Tunneling

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Network tunneling is where two networks are connected to one another over the public internet. This is accomplished through different protocols including:

Point-to-Point Protocol (PTPP) which can be used with different authentication options. Adding / removing frames is called “tunneling”.

Point-to-Point Protocol (PTPP) Tunneling

The Password Authentication Protocol (PAP) uses an ID and password to authenticate users, however passwords are sent in clear-text.

There are lots of versions of the Challenge-Handshake Authentication Protocol (CHAP), a common one is MD5-CHAP. CHAP provides for repeated / continuous authentication if desired to re-authenticate a client on a set time period (i.e. every 4 hours) to reduce risk of spoofing and session hijacking. In the standard version of CHAP, passwords are stored in clear text on the server, leaving it up to the implementation vendor on how to protect those passwords.

The CHAP Challenge number is typically a randomized session ID issued for tracking each user. The Nonce (short for “Number Used Once”) is typically a time-based session ID, only used once (timestamp) to reduce the risk of password replay attacks.

Challenge-Handshake Authentication Protocol (CHAP) Process

The Extensible Authentication Protocol (EAP) can be used with passwords, challenge and response (e.g. CHAP), biometrics, tokens, combining protocols (e.g. EAS+TLS), device authentication, etc.

Layer 2 Tunneling Protocol (L2TP) is a hybrid of L2F and PPTP. PPTP was developed by Microsoft, PPTP uses PPP frames but provides encryption. L2F was developed by Cisco and also uses PPP frames but does not have encryption. However, each tunnel can support multiple connections per user. L2F requires special hardware to use it.

Microsoft & Cisco worked together to combine L2F and PPTP into L2TP. L2TP uses PPP frames, supports multiple connections in a single tunnel, does not require special hardware, but also does NOT provide encryption.

IPSec is a protocol designed to protect IP traffic through use of an:

• Authentication Header (AH) which is designed to authenticate source IP addresses, and
• Encapsulating Security Payload (ESP) which provides encryption of both payload and header if desired.

Security Associations (SA’s) are one-way connections using either AH or ESP services. Each SA is uniquely identified using a:

• Security Parameter Index (a session ID for tracking connection)
• Destination IP address
• AH or ESP identifier A second SA must be defined for 2-way communication.

End-to-end encryption / Transport Mode is where only the payload portion of a packet is encrypted (if using encryption).

Two examples of this include: Transport Mode with ESP

Transport Mode with Encapsulating Security Payload (ESP)

And Link Encryption / Tunnel Mode – this is where the entire original packet is encrypted including the original header and payload. It is also known as a Gateway-to-Gateway VPN. A trick to help you remember it is <u>Linc</u>oln Tunnel -> Tunnel Mode uses <u>Link</u> Encryption

Here’s an example of Tunnel Mode with ESP:

Tunnel Mode with Encapsulating Security Payload (ESP)

To summarize:

Tunneling Protocol	Provides Encryption?
IPSec	Yes, ESP
SSH	Yes
L2F	No
L2TP	No, but yes if partnered with IPSec
PPTP	Yes
MPLS	No
TLS/SSL	Yes
PPP	No

To end our discussion about the OSI model, we’ll touch briefly on how Transport Layer Security (TLS) and Secure Sockets Layer (SSL) work:

TLS/SSL Connection Negotiation Process

CISSP Blog Post 19, Domain 4: Network: OSI Layers 5 (Session Link) & 6 (Presentation)

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Layer 5 protocols coordinate the orderly exchange of information. They include:

The Remote Procedure Call (RPC) Protocol which is utilized in client-server environments and Secure RPC which uses mutual authentication for client & server to authenticate one another.

Remote Procedure Call (RPC) Protocol Process

Layer 6 protocols are responsible for giving applications access to the network services, i.e. they help applications talk to the network. An example is Microsoft Outlook using Layer 6 protocols such as SMTP, POP3, or IMAP to handle email transmission on the network. Other examples include:

The Domain Name System (DNS) protocol which is a translation service to resolve Fully Qualified Domain Names (FQDN) to IP addresses. The way this works is:

1. Browser sends domain to ISP for lookup
2. ISP DNS goes through recursive search, first to Root DNS, which will return local .ORG DNS address
3. .ORG DNS will return IP address of DNS server of actual website Website
4. DNS server will return IP address of actual website to ISP DNS
5. ISP DNS then returns actual website IP address to Browser

Domain Name Service (DNS) Lookup Process

DNS Security (DNS-SEC) is a protocol designed to combat DNS cache poisoning using digital signatures to verify that DNS data is coming from authentic sources.

Network Address Translation (NAT) is the translation between public internet IP addresses and local (private) IP addresses. Private IP ranges include:

• 10.0.0.0 – 10.255.255.255
• 172.16.0.0 – 172.31.255.255
• 192.168.0.0 – 192.168.255.255

Network Address Translation (NAT)

CISSP Blog Post 18, Domain 4: Network OSI Layer 4, the Transport Layer

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

First off, Happy New Years! Hopefully your New Year’s resolution was to continue to study for the CISSP exam! 😉

Today we’re going to look at Layer 4 protocols which are responsible for end-to-end, host-to-host, or source-to-destination communications. Common protocols here include the:

Transmission Control Protocol (TCP) which is a connection-oriented protocol, i.e. it expects acknowledgments from the destination.

TCP Handshake

The sequence number increments by the size of bytes in the packet. To close transmission, device A send a final ACK transmission and then sends FIN with the last sequence it is up to:

TCP Communication Close

TCP Headers are sandwiched between the IP Header and the Data Payload. They typically include:

• Source & Destination Ports
• Sequence #’s
• Acknowledgement #’s
• Flags
• Checksums
• Etc.

Well known port numbers range from 0 to 1023. Examples are:

• HTTP = 80
• SMTP = 25
• FTP = 20 & 21
• Telnet = 23
• SSH = 22
• DNS = 53
• HTTPS = 443

Ports indicate the protocol being used. The sequence # and checksum are used to figure out if a packet needs to be resent.

User Datagram Protocol (UDP) is a connection-less protocol, i.e. it does not expect acknowledgements and does not have any error checking. It assumes best effort and there is no connection “state” for a firewall to observe / monitor. The UDP header only includes source and destination ports, checksum, etc.

Two common TCP exploits at the OSI Layer 4 level include the

TCP Sequence Number Attack

TCP Sequence Number Attack

SYN Flood Attack

SYN Flood Attack

CISSP Blog Post 17, Domain 4: Network: OSI Layers 2 (Data Link) & 3 (Network)

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Not going to spend a lot of time on OSI Layer 2, aka the Data Link Layer. At this layer, protocols are responsible for node-to-node or link-to-link communications between systems on the same network

Common protocols include:

• Address Resolution Protocol (ARP) which resolves IP addresses to MAC addresses. It is susceptible to ARP cache poisoning attacks.
• Multiprotocol Label Switching (MPLS) operates at both L2 and L3 of the OSI model. In summary, data enters the MPLS network through a Label Edge Router which passes traffic to Label Switch Routers until data gets to the final Edge Router and exits the MPLS network to the destination. MPLS routers add special encapsulating data labels to send data along pre-defined network paths. The Switch routers forward data along the directed path, where the last edge router strips off the labels before the data exists the network.

The OSI Layer 3, Network Layer protocols are responsible for network-to-network, router-to-router, or gateway-to-gateway communications and include:

The IP Protocol (IPv4) handles IP addressing. Without going into the difference between the two, the main reason IPv6 was developed is because IPv4 has literally run out of addressable addresses to assign to Internet devices.

• IPv4 = 32-bit addresses which provide 2^32 addresses
• IPv6 = 128-bit addresses which provides 2^128 addresses

The IP protocol handles data fragmentation and reassembly of packets if you’re sending over networks with different maximum transmission unit (MTU) sizes. The IP header element includes items such as:

• Total fragment length
• Fragment ID
• Different flags
• Fragment offset
• Time-to-live
• IP source address
• IP destination address
• Etc.

The Internet Control Message Protocol (ICMP) can be abused to conduct Man-in-the-Middle (MITM) and Denial of Service (DoS) attacks. It provides network diagnostics (such as ping, traceroute, etc.) and network error reporting. It also provides ICMP redirect functionality where a router can inform a sender of a better route to the final destination.

A couple of common Layer 3 attacks include the:

• Tear Drop Attack which crashes a system by exploiting the fragment offset field to overload the receiving system with malformed packets.
• SMURF Attack where an ICMP echo request is sent to the network broadcast address of a spoofed victim, causing all nodes to reply to the victim with echo reply. A similar attack called a FRAGGLE attack uses the UDP protocol to create a similar effect.

SMURF Attack Process

CISSP Blog Post 16, Domain 4: Network: WiFi

Credit: Post based on CISSP course presented by Dennis Lee, November 2018

Let’s dive a little deeper into the OSI Layer 1, Physical layer for WiFi:

There are a couple of different wireless transmission methods that include:

• Direct Sequence Spread Spectrum (DSSS) which is a wide frequency channel (band). An example is 802.11b
• Frequency Hopping Spread Spectrum (FHSS) uses multiple narrow frequency channels / bands in sequential order. An example technology is Bluetooth.
• Orthogonal Frequency Division Multiplexing (OFDM) uses multiple narrow frequency channels / bands simultaneously for faster throughput. The example here is 802.11n.

How about Wi-Fi Authentication Methods? These include:

• Open System Authentication is essentially no authentication – all that’s required is for the client device to transmit a service set identifier (aka SSID) for the access point to which it’s transmitting. This is a weak form of authentication because anyone can sniff and copy an SSID, even if the AP isn’t broadcasting it.
• MAC Address Filtering at the access point – this is also weak as someone can sniff and spoof a MAC address.
• The 802.1x protocol is the strongest authentication option. This is a port-based authentication protocol standard for both wired and wireless networks and has two components:
• The Extensible Authentication Protocol (EAP)
• And the Remote Authentication Dial-in User System aka RADIUS, which has a central server to control an access list.

RADIUS Authentication Process

Okay, you’ve authenticated your device to the network – so here’s a summary of how you can encrypt your data while in transit via Wi-Fi:

Wi-Fi Encryption Standard	Algorithm	Key Usage	Integrity Checking
Wired Equivalent Privacy (WEP)	RC4	Can choose either a 40-bit key or 104-bit key. The 40-bit key has a 24-bit IV so equivalent of a 64-bit key. The 104-bit key has a 24-bit IV, equivalent to a 128-bit key.	Checksums proves accidental changes did not occur
WiFi Protected Access (WPA)	RC4	128-bit temporal key + Client MAC (48-bit) + IV (48-bit) = Temporal Key Integrity Protocol (TKIP) key. This is different for each user session. The IV is different for each packet, it’s based on a data sequence number.	Uses HMAC with 2 phases of hashing (including key) proving both accidental and intentional tampering didn’t occur
IEEE 802.11i (WPA2)	AES	TKIP or Countermode of AES	CBD-MAC 2 stages of encryption proves both accidental and intentional tampering didn’t occur

Here’s a summary of all the IEEE Standard’s we’ve covered thus far:

IEEE	Description
802.1x	Port-based authentication protocol
802.11i WPA2	WiFi Encryption Standard
802.3	Ethernet
802.5	Token Ring

IEEE	Frequency Band	Data Rate	Transmission Method	Common Name
802.11b	2.4 GHz	11 Mbps	DSSS	These are all Wi-Fi Standards
802.11a	5 GHz	54 Mbps	OFDM
802.11g	2.4 GHz	54 Mbps	OFDM
802.11n	5 and 2.4 GHz	250 Mbps+	OFDM
802.11ac	5 GHz	430 Mbps+	OFDM
802.16	2 – 11 GHz or 10 – 66 GHz	Many	OFDM and Others	WiMax for wired metro network
802.15.1	2.4 GHz	Many	FHSS	Bluetooth