Secure Communication Channels & Protocols
CISSP CISSP · Communication and Network Security
Securing the channels, not just the network
Network architecture controls where traffic can go; secure communication controls protect the traffic itself. Domain 4's third objective covers the protocols and techniques that keep communications confidential, authenticated and intact as they cross untrusted networks. The exam's approach is practical: for each common communication need — web browsing, email, file transfer, voice, remote administration, name resolution — know the secure protocol, know the insecure predecessor it replaced, and know what breaks if you use the insecure one.
TLS: the foundation of secure web traffic
Transport Layer Security (TLS) is the protocol that secures HTTP (as HTTPS), email, and many other application protocols. TLS provides confidentiality (encryption), integrity (message authentication), and authentication (of the server, via its certificate; mutual TLS can authenticate the client too). The TLS handshake negotiates the protocol version and cipher suite, authenticates the server using its X.509 certificate, and establishes a shared session key — typically using Diffie-Hellman key exchange for perfect forward secrecy (PFS), which ensures that even if the server's private key is later compromised, past session keys cannot be derived and past traffic cannot be decrypted. TLS 1.3 is the current version; it removed support for obsolete algorithms, reduced the handshake to a single round trip, and made forward secrecy mandatory. SSL and TLS 1.0/1.1 are deprecated and must not be used. The exam's key points: TLS protects data in transit, not data at rest; a valid certificate is essential (a self-signed or expired certificate breaks the trust chain); and TLS does not protect against attacks on the server itself — an encrypted connection to a compromised server is still compromised.
Email security
Email is the most attacked communication channel, and the exam covers both its protection and its authentication. For confidentiality and integrity of message content, the two standards are S/MIME (Secure/Multipurpose Internet Mail Extensions, which uses X.509 certificates and PKI) and PGP/GPG (Pretty Good Privacy, which uses a web-of-trust model rather than a central CA). For authenticating the sender's domain and preventing spoofing, three DNS-based mechanisms work together: SPF (Sender Policy Framework — publishes which mail servers are authorised to send for a domain), DKIM (DomainKeys Identified Mail — the sending server digitally signs the message headers, proving the message came from the domain and was not altered), and DMARC (Domain-based Message Authentication, Reporting and Conformance — tells receivers what to do with messages that fail SPF or DKIM checks, and provides reporting). The exam tests the three together: SPF authorises senders, DKIM signs messages, DMARC enforces policy. STARTTLS opportunistically upgrades an SMTP connection to TLS, protecting mail in transit between servers — but it is opportunistic, so a downgrade attack can strip it; MTA-STS and DANE address that weakness. Email gateways add anti-spam, anti-malware and URL rewriting controls.
| Mechanism | What it does |
|---|---|
| SPF | Lists which servers may send mail for a domain (DNS TXT record) |
| DKIM | Digitally signs message headers so receivers can verify origin and integrity |
| DMARC | Policy for handling messages that fail SPF/DKIM; provides reporting |
| S/MIME | Encrypts and signs message content using PKI certificates |
| PGP/GPG | Encrypts and signs message content using a web of trust |
| STARTTLS | Opportunistically upgrades SMTP to TLS in transit |
DNS security
The Domain Name System translates human-readable names into IP addresses, and it is a frequent attack target because compromising DNS redirects users to attacker-controlled servers. DNSSEC (DNS Security Extensions) adds digital signatures to DNS records so that resolvers can verify the authenticity and integrity of responses — it prevents cache poisoning and spoofing, but it does not encrypt DNS queries, so it provides no confidentiality. DNS over HTTPS (DoH) and DNS over TLS (DoT) encrypt DNS queries, providing confidentiality but raising a management concern: encrypted DNS can bypass enterprise DNS-based filtering and monitoring, so organisations must decide how to handle it. DNS cache poisoning injects false records into a resolver's cache; DNS tunnelling exfiltrates data by encoding it in DNS queries (since DNS traffic is almost always allowed through firewalls); and domain hijacking takes over a domain's registration. The exam's key distinction: DNSSEC provides authentication and integrity, DoH/DoT provide confidentiality — they solve different problems and are complementary, not alternatives.
Secure protocols for common tasks
The exam expects you to know the secure replacement for each legacy protocol. SSH replaces Telnet and rlogin for remote administration, providing encrypted terminal access, file transfer (SCP/SFTP), and port forwarding. SFTP and FTPS replace FTP for file transfer (SFTP runs over SSH; FTPS is FTP over TLS). HTTPS replaces HTTP. LDAPS replaces LDAP. SNMPv3 replaces SNMPv1/v2, adding authentication and encryption — earlier SNMP versions send community strings (effectively passwords) in cleartext. NTP should be authenticated (NTS, Network Time Security) because accurate time underpins logging, certificates and Kerberos. The exam's pattern is consistent: if a scenario uses Telnet, FTP, HTTP, SNMPv1 or any other cleartext protocol for sensitive traffic, the correct answer is the encrypted replacement. The reason is always the same: cleartext protocols expose credentials and data to anyone who can observe the traffic.
Voice and multimedia security
Voice over IP (VoIP) and video conferencing bring their own security considerations. SIP (Session Initiation Protocol) sets up calls and should be protected with TLS; the media stream uses RTP (Real-time Transport Protocol), and its secure variant is SRTP (Secure RTP), which provides encryption and integrity for the audio/video itself. Common VoIP threats include eavesdropping (unencrypted calls can be intercepted), toll fraud (attackers abuse a PBX to make expensive calls), vishing (voice phishing), caller ID spoofing, and denial of service against call infrastructure. The exam's practical points: encrypt both the signalling (SIP over TLS) and the media (SRTP), segment voice traffic onto its own VLAN, and restrict PBX access to prevent toll fraud.
Remote access and cloud connectivity
Remote access has moved beyond the traditional VPN. Zero Trust Network Access (ZTNA) and the broader zero trust model assume no implicit trust based on network location: every access request is authenticated, authorised and encrypted, regardless of whether the user is inside or outside the corporate network. This contrasts with the traditional perimeter model, where being inside the network (or on the VPN) implies trust. The exam's zero trust principles are: verify explicitly (authenticate and authorise every request), use least privilege access, and assume breach (design as though the network is already compromised). Cloud connectivity adds its own options: direct connect / ExpressRoute style private links provide dedicated connections to cloud providers (more predictable and private than internet VPNs, but more expensive), and cloud access security brokers (CASBs) sit between users and cloud services to enforce policy, detect shadow IT, and protect data in cloud applications.
Exam mindset for secure communications
Protocol questions reward knowing the secure replacement: whenever a scenario uses a cleartext protocol, the answer is the encrypted version. For email, remember the SPF/DKIM/DMARC trio and their distinct roles. For DNS, remember that DNSSEC authenticates and DoH/DoT encrypts. And for any question about trust, zero trust is the modern framing: verify explicitly, least privilege, assume breach.
Exam traps to watch for
- Saying DNSSEC encrypts DNS queries — DNSSEC authenticates records; DoH and DoT provide confidentiality.
- Confusing SPF, DKIM and DMARC: SPF authorises sending servers, DKIM signs messages, DMARC sets policy for failures.
- Recommending SSL or TLS 1.0/1.1 — both are deprecated. TLS 1.2 is the minimum; TLS 1.3 is current.
- Thinking TLS protects data at rest or protects against a compromised server — TLS protects data in transit only.
Practice questions
Test yourself on this guide. Answers are checked instantly, with a full explanation for each question.
Q1.An organisation discovers that attackers are sending phishing emails that appear to come from its domain. Which combination of controls BEST prevents recipients from trusting these spoofed messages?
Explanation: SPF authorises which servers may send for the domain, DKIM cryptographically signs legitimate messages, and a DMARC policy of "reject" tells receiving servers to discard messages that fail both checks — together they make domain spoofing fail at the receiver. STARTTLS encrypts mail in transit but does nothing about spoofed senders. S/MIME protects message content for recipients who use it, not the domain's reputation with the wider internet. An inbound spam filter protects the organisation's own inboxes, not external recipients receiving spoofed mail in the organisation's name.
Exam objective: CISSP Domain 4: Understand secure communication channels (email authentication)
Q2.A web server’s private key is compromised and an attacker has recorded months of encrypted HTTPS traffic. Which TLS feature would have prevented the attacker from decrypting the recorded traffic?
Explanation: Perfect forward secrecy uses ephemeral key exchange (such as ephemeral Diffie-Hellman) so that each session key is independent of the server's long-term private key; compromising the private key later does not expose past sessions. A longer RSA key does not help if the attacker has the private key itself. Certificate validity period is irrelevant to recorded traffic. OCSP stapling concerns certificate revocation status, not session key derivation.
Exam objective: CISSP Domain 4: Understand secure communication channels (TLS, forward secrecy)
Q3.An attacker poisons a DNS resolver’s cache so that users requesting the company’s banking portal are directed to a malicious server. Which control BEST prevents this attack?
Explanation: DNSSEC adds cryptographic signatures to DNS records, allowing resolvers to verify that a response is authentic and unmodified — a poisoned record would fail validation. DoH encrypts queries for confidentiality but does not authenticate the records themselves. A longer TTL would make a poisoned entry persist longer, worsening the problem. Hosts files are unmanageable at scale and do not protect the resolver or other clients.
Exam objective: CISSP Domain 4: Understand secure communication channels (DNS security)
Q4.A network administrator manages routers using Telnet over the corporate network. What is the MOST significant risk, and what should replace Telnet?
Explanation: Telnet transmits everything — including administrator credentials and configuration commands — in cleartext, so anyone able to observe the traffic can capture full administrative access. SSH provides encrypted terminal access and is the standard replacement. HTTP is also cleartext and is not a terminal protocol. SNMP is a monitoring protocol, not a remote administration replacement, and SNMPv1/v2 are themselves cleartext. FTP is a file transfer protocol, not a terminal protocol.
Exam objective: CISSP Domain 4: Understand secure communication channels (secure protocols)
Q5.An organisation is adopting a zero trust architecture. Which principle is MOST central to the model?
Explanation: Zero trust removes the assumption that network location implies trust: every request is authenticated, authorised and encrypted based on identity, device state and context, with least-privilege access. Trusting internal traffic is the perimeter model that zero trust explicitly rejects. A single VPN concentrator is a perimeter chokepoint, not a zero trust control. Encrypting data at rest is a good control but is not the defining principle of zero trust.
Exam objective: CISSP Domain 4: Understand secure network architecture (zero trust)
Q6.A company’s VoIP system is experiencing toll fraud: attackers are using the PBX to place expensive international calls. Which control BEST addresses this specific threat?
Explanation: Toll fraud is unauthorised use of the phone system, so the direct countermeasures are restricting which destinations can be called (blocking international or premium numbers that the business does not need) and requiring strong authentication for PBX access so attackers cannot use it at all. SRTP encrypts call media but does not stop an attacker from placing calls. VLAN segmentation isolates voice traffic but does not control what the PBX allows. Caller ID does not prevent fraud and is easily spoofed.
Exam objective: CISSP Domain 4: Understand voice and multimedia security
Q7.An organisation uses SNMPv2 to monitor its network devices. An audit flags the SNMP community strings as a finding. Why?
Explanation: SNMPv1 and v2c send community strings — which act as passwords, with the "private" community typically granting read-write access — in cleartext, so they can be captured by anyone observing network traffic. The fix is SNMPv3, which adds authentication and encryption. SNMPv2 supports modern devices; the problem is security, not compatibility. Community strings are typically short and weak ("public"/"private" by default), not too complex. No separate management VLAN is required by the protocol.
Exam objective: CISSP Domain 4: Understand secure communication channels (SNMP)
Q8.A security architect is reviewing a web application that still supports TLS 1.0 for compatibility with old clients. What is the BEST recommendation?
Explanation: TLS 1.0 and 1.1 are formally deprecated (RFC 8996) and contain known vulnerabilities such as BEAST and weak cipher support; the correct action is to disable them and require TLS 1.2 or higher. Keeping them for compatibility exposes all users to downgrade attacks. SSL 3.0 is even older and is vulnerable to POODLE. Serving old clients over HTTP removes all protection.
Exam objective: CISSP Domain 4: Understand secure communication channels (TLS versions)