CyberKit

Network Case: answer key

Every question with its answer and the reasoning. Work the drill first, or the key only tells you what, not how.

Stages: Capture, Recall, Subnets.

Capture

  1. Open PCAP Triage, press "Load sample", and read the answer off the tool. The excerpt below is the same parser output if you would rather read it here. How many packets are in the capture?

    Packets: 37   Bytes: 3444
    First seen: 2026-03-01T12:47:09Z
    Last seen:  2026-03-01T12:53:50Z
    Protocol mix: ARP 1, UDP 7, TCP 29

    Answer: 37

    37 packets. Record the count and the time window before anything else: they bound every later claim ("no other traffic to that host in the capture") and they are what the next analyst checks your copy against.

    Read more: PCAP Triage

  2. Open PCAP Triage, press "Load sample", and read the answer off the tool. The excerpt below is the same parser output if you would rather read it here. One domain is queried repeatedly for TXT records, each time with a different random-looking label in front. What is that parent domain?

    DNS queries
    Name                         Type  Count
    ---------------------------  ----  -----
    cdn.example.org              A     2
    a1b2c3d4.tunnel.example.net  TXT   1
    e5f6a7b8.tunnel.example.net  TXT   1
    c9d0e1f2.tunnel.example.net  TXT   1
    a3b4c5d6.tunnel.example.net  TXT   1

    Answer: tunnel.example.net

    tunnel.example.net. 4 TXT queries, each with a unique, hex-looking leftmost label under the same parent. The constant parent is the attacker-controlled zone; the varying label is the payload.

    Read more: PCAP Triage

  3. Open PCAP Triage, press "Load sample", and read the answer off the tool. The excerpt below is the same parser output if you would rather read it here. What does that run of TXT lookups most likely represent?

    DNS queries
    Name                         Type  Count
    ---------------------------  ----  -----
    cdn.example.org              A     2
    a1b2c3d4.tunnel.example.net  TXT   1
    e5f6a7b8.tunnel.example.net  TXT   1
    c9d0e1f2.tunnel.example.net  TXT   1
    a3b4c5d6.tunnel.example.net  TXT   1

    Answer: DNS tunnelling: data carried in the query names and TXT answers

    Unique high-entropy labels under one parent, requested as TXT, from a workstation rather than a mail server. SPF/DKIM lookups go to the sending domain's _domainkey or apex, not to random subdomains; retries repeat the same name. No single query here is suspicious. The pattern is.

    Read more: PCAP Triage

  4. Open PCAP Triage, press "Load sample", and read the answer off the tool. The excerpt below is the same parser output if you would rather read it here. What host and URI did the one HTTP GET request ask for? Answer as host/uri.

    HTTP requests
    Method  Host               URI                 Count
    ------  -----------------  ------------------  -----
    GET     files.example.com  /updates/patch.bin  1

    Answer: files.example.com/updates/patch.bin

    files.example.com/updates/patch.bin. The Host header plus the request URI is the full resource. A .bin pulled over cleartext HTTP is either an update mechanism doing something unwise or a stager. The next step is the proxy logs and the file itself, not the packet.

    Read more: PCAP Triage

  5. Open PCAP Triage, press "Load sample", and read the answer off the tool. The excerpt below is the same parser output if you would rather read it here. The TLS session is encrypted. Which server name can you still read from its ClientHello?

    TLS server names
    SNI              Servers        Count
    ---------------  -------------  -----
    cdn.example.org  198.51.100.20  1

    Answer: cdn.example.org

    cdn.example.org. Server Name Indication rides in the cleartext ClientHello so the server can pick a certificate. It is client-asserted, so it says what the client asked for, not what it reached, and Encrypted Client Hello removes it entirely.

    Read more: PCAP Triage

  6. Open PCAP Triage, press "Load sample", and read the answer off the tool. The excerpt below is the same parser output if you would rather read it here. One external address receives a long run of small, near-identical TCP conversations. Which address and port? Answer as ip:port.

    Conversations
    Source            Destination        Proto  Packets  Bytes  First seen
    ----------------  -----------------  -----  -------  -----  --------------------
    192.0.2.10:51514  192.0.2.1:53       UDP    2        166    2026-03-01T12:47:09Z
    192.0.2.10:49212  198.51.100.20:443  TCP    2        310    2026-03-01T12:47:09Z
    192.0.2.10:49213  198.51.100.20:80   TCP    2        286    2026-03-01T12:47:09Z
    192.0.2.10:51600  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:10Z
    192.0.2.10:51601  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:10Z
    192.0.2.10:51602  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:11Z
    192.0.2.10:51603  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:12Z
    192.0.2.10:49300  203.0.113.47:8443  TCP    2        181    2026-03-01T12:47:14Z
    192.0.2.10:49301  203.0.113.47:8443  TCP    2        181    2026-03-01T12:47:44Z
    192.0.2.10:49302  203.0.113.47:8443  TCP    2        181    2026-03-01T12:48:14Z
    192.0.2.10:49303  203.0.113.47:8443  TCP    2        181    2026-03-01T12:48:44Z
    192.0.2.10:49304  203.0.113.47:8443  TCP    2        181    2026-03-01T12:49:14Z
    192.0.2.10:49305  203.0.113.47:8443  TCP    2        181    2026-03-01T12:49:44Z
    192.0.2.10:49306  203.0.113.47:8443  TCP    2        181    2026-03-01T12:50:14Z
    192.0.2.10:49307  203.0.113.47:8443  TCP    2        181    2026-03-01T12:50:44Z
    192.0.2.10:49308  203.0.113.47:8443  TCP    2        181    2026-03-01T12:51:14Z
    192.0.2.10:49309  203.0.113.47:8443  TCP    2        181    2026-03-01T12:51:44Z
    192.0.2.10:49310  203.0.113.47:8443  TCP    2        181    2026-03-01T12:52:14Z
    192.0.2.10:49311  203.0.113.47:8443  TCP    2        181    2026-03-01T12:52:44Z
    192.0.2.10:49400  198.51.100.20:445  TCP    1        58     2026-03-01T12:53:49Z
    2001:db8::1:53    2001:db8::a:51700  UDP    1        62     2026-03-01T12:53:50Z

    Answer: 203.0.113.47:8443

    203.0.113.47:8443. 12 short conversations, same size each time, asymmetric (a small POST out, a small reply back). Grouping the conversation table by destination is what makes it visible; in packet order it is interleaved with everything else.

    Read more: PCAP Triage

  7. Open PCAP Triage, press "Load sample", and read the answer off the tool. The excerpt below is the same parser output if you would rather read it here. Those conversations to 203.0.113.47:8443 start at a fixed interval. How many seconds apart are they?

    Conversations
    Source            Destination        Proto  Packets  Bytes  First seen
    ----------------  -----------------  -----  -------  -----  --------------------
    192.0.2.10:51514  192.0.2.1:53       UDP    2        166    2026-03-01T12:47:09Z
    192.0.2.10:49212  198.51.100.20:443  TCP    2        310    2026-03-01T12:47:09Z
    192.0.2.10:49213  198.51.100.20:80   TCP    2        286    2026-03-01T12:47:09Z
    192.0.2.10:51600  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:10Z
    192.0.2.10:51601  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:10Z
    192.0.2.10:51602  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:11Z
    192.0.2.10:51603  192.0.2.1:53       UDP    1        87     2026-03-01T12:47:12Z
    192.0.2.10:49300  203.0.113.47:8443  TCP    2        181    2026-03-01T12:47:14Z
    192.0.2.10:49301  203.0.113.47:8443  TCP    2        181    2026-03-01T12:47:44Z
    192.0.2.10:49302  203.0.113.47:8443  TCP    2        181    2026-03-01T12:48:14Z
    192.0.2.10:49303  203.0.113.47:8443  TCP    2        181    2026-03-01T12:48:44Z
    192.0.2.10:49304  203.0.113.47:8443  TCP    2        181    2026-03-01T12:49:14Z
    192.0.2.10:49305  203.0.113.47:8443  TCP    2        181    2026-03-01T12:49:44Z
    192.0.2.10:49306  203.0.113.47:8443  TCP    2        181    2026-03-01T12:50:14Z
    192.0.2.10:49307  203.0.113.47:8443  TCP    2        181    2026-03-01T12:50:44Z
    192.0.2.10:49308  203.0.113.47:8443  TCP    2        181    2026-03-01T12:51:14Z
    192.0.2.10:49309  203.0.113.47:8443  TCP    2        181    2026-03-01T12:51:44Z
    192.0.2.10:49310  203.0.113.47:8443  TCP    2        181    2026-03-01T12:52:14Z
    192.0.2.10:49311  203.0.113.47:8443  TCP    2        181    2026-03-01T12:52:44Z
    192.0.2.10:49400  198.51.100.20:445  TCP    1        58     2026-03-01T12:53:49Z
    2001:db8::1:53    2001:db8::a:51700  UDP    1        62     2026-03-01T12:53:50Z

    Answer: 30

    30 seconds, with no jitter at all. Real implants usually add jitter to defeat exactly this check, so a perfectly regular interval is either unsophisticated tooling or a legitimate heartbeat. The destination decides which.

    Read more: PCAP Triage

  8. Why can the beacon not be spotted by inspecting any single packet?

    Answer: Each packet is small and ordinary; the signal is the repetition and the timing across many conversations

    Beaconing is a property of the aggregate: count, size regularity and inter-arrival time. That is why triage starts from conversations and not from packets, and why the same analysis on Zeek conn.log is the standard hunt at scale.

    Read more: PCAP Triage

  9. Match each finding to the panel of the tool that shows it.

    Answer: The 30-second check-ins to 203.0.113.47:8443 → Conversations · Random-looking labels under tunnel.example.net → DNS queries · The download of /updates/patch.bin → HTTP requests · The name cdn.example.org on an encrypted session → TLS server names · Which pair of hosts moved the most bytes → Conversations

    The four panels map onto the four questions of first-pass network triage. Byte volume and timing live in Conversations; content-level indicators only exist where the protocol is readable (DNS, HTTP) or where the handshake leaks it (TLS SNI).

    Read more: PCAP Triage

  10. You are handed an unfamiliar capture. Order the first-pass triage steps.

    Answer: Note the packet count, time window and protocol mix Rank conversations by volume and look for regular timing Read DNS, HTTP and TLS names for indicators Pull the full packets for the few flows that matter

    Scope, then shape, then content, then detail. Opening packets first is the classic way to spend an hour on a capture and miss the beacon, because packet order interleaves every flow. Each step decides which small part of the capture the next step reads.

    Read more: PCAP Triage

Recall

  1. Port 22/TCP. What service is this?

    Answer: SSH (Secure Shell) / SFTP / SCP

    22/tcp: SSH (Secure Shell) / SFTP / SCP. An encrypted remote-login and command-execution protocol (RFC 4251/4253) that also carries file transfer as SFTP and SCP subsystems over the same authenticated, encrypted session, plus arbitrary TCP port forwarding/tunneling. It is the de facto standard for administering Linux/Unix hosts and network appliances, and is increasingly available on Windows via OpenSSH.

    Read more: Port 22 reference

  2. HTTP 200. What is the reason phrase?

    Answer: OK

    200 OK. The request succeeded. Any body carries the requested resource or the result of the action.

    Read more: HTTP Status Code Reference

  3. Which DNS record type does this? Maps a hostname to a 32-bit IPv4 address: the fundamental record type behind ordinary IPv4 name resolution.

    Answer: A

    A (TYPE 1): a host address Example: example.com. A 192.0.2.1

    Read more: DNS Record Type Reference

  4. Port 23/TCP. What service is this?

    Answer: Telnet

    23/tcp: Telnet. A plaintext remote-terminal protocol (RFC 854) that predates SSH by roughly two decades. It opens an interactive command-line session on the remote host with no encryption and, in its base form, no integrity protection at all.

    Read more: Port 23 reference

  5. HTTP 301. What is the reason phrase?

    Answer: Moved Permanently

    301 Moved Permanently. The resource has permanently moved to a new URI, and future requests should go straight to the URI given in the response's Location header.

    Read more: HTTP Status Code Reference

  6. Which DNS record type does this? Maps a hostname to a 128-bit IPv6 address: the IPv6 counterpart to the A record, needed because A's RDATA only has room for 32 bits.

    Answer: AAAA

    AAAA (TYPE 28): IP6 Address Example: example.com. AAAA 2001:db8::1

    Read more: DNS Record Type Reference

  7. Port 25/TCP. What service is this?

    Answer: SMTP (Simple Mail Transfer Protocol)

    25/tcp: SMTP (Simple Mail Transfer Protocol). SMTP is the protocol mail servers use to relay email between each other (MTA-to-MTA) and, historically, from a client to a server. Port 25/tcp is its IANA-registered default, governed by RFC 5321. Modern deployments largely restrict port 25 to server-to-server relay and inbound mail delivery, pushing client mail submission to dedicated ports (465/587) instead.

    Read more: Port 25 reference

  8. HTTP 302. What is the reason phrase?

    Answer: Found

    302 Found. The resource is temporarily available at a different URI, but the client should keep using the original request URI for future requests.

    Read more: HTTP Status Code Reference

  9. Which DNS record type does this? Points one name to another canonical name, so every other record for the alias is looked up through the target instead of being duplicated.

    Answer: CNAME

    CNAME (TYPE 5): the canonical name for an alias Example: www.example.com. CNAME example.com.

    Read more: DNS Record Type Reference

  10. Port 53/TCP/UDP. What service is this?

    Answer: DNS (Domain Name System)

    53/tcp/udp: DNS (Domain Name System). DNS resolves human-readable names to IP addresses, and vice versa, plus many other record types, and is defined across RFC 1034 (concepts) and RFC 1035 (implementation and wire format). It genuinely runs over both UDP and TCP on port 53: UDP handles the vast majority of ordinary queries for low overhead, while TCP is required whenever a response exceeds the practical UDP size limit, and always for zone transfers (AXFR, defined in RFC 5936) between authoritative name servers.

    Read more: Port 53 reference

  11. HTTP 304. What is the reason phrase?

    Answer: Not Modified

    304 Not Modified. A conditional GET request (one carrying an If-Modified-Since/If-None-Match header) found the resource unchanged, so the client's cached copy is still valid and doesn't need to be re-downloaded.

    Read more: HTTP Status Code Reference

  12. Which DNS record type does this? Identifies a mail server for the domain along with a preference value used to rank it against any other mail servers for the same domain: lower values are preferred.

    Answer: MX

    MX (TYPE 15): mail exchange Example: example.com. MX 10 mail.example.com.

    Read more: DNS Record Type Reference

  13. Port 80/TCP. What service is this?

    Answer: HTTP (Hypertext Transfer Protocol)

    80/tcp: HTTP (Hypertext Transfer Protocol). HTTP is the application-layer protocol underlying the World Wide Web: a stateless request/response protocol where a client requests a resource by method, path, and headers, and a server returns a status code, headers, and a body. Port 80/tcp is its IANA-registered default, governed by RFC 9110 (HTTP Semantics). Traffic on 80 is plaintext by default, with no confidentiality or integrity protection unless something above the transport adds it.

    Read more: Port 80 reference

  14. HTTP 400. What is the reason phrase?

    Answer: Bad Request

    400 Bad Request. The server can't or won't process the request because of something it perceives as a client-side problem (malformed syntax, invalid request framing, or similar) without a more specific 4xx code fitting better.

    Read more: HTTP Status Code Reference

  15. Which DNS record type does this? Delegates a zone (or subdomain) to a specific set of authoritative name servers.

    Answer: NS

    NS (TYPE 2): an authoritative name server Example: example.com. NS ns1.example.net.

    Read more: DNS Record Type Reference

  16. Port 88/TCP/UDP. What service is this?

    Answer: Kerberos

    88/tcp/udp: Kerberos. Kerberos is the network authentication protocol Active Directory and other Kerberos realms use to issue and validate tickets proving a principal's identity without repeatedly transmitting a password: a client obtains a Ticket-Granting Ticket (TGT) from the Key Distribution Center's Authentication Service, then uses it to request per-service Ticket-Granting Service (TGS) tickets. Port 88 is IANA-registered and defined by RFC 4120 (The Kerberos Network Authentication Service V5), and it genuinely runs over both UDP and TCP: RFC 4120 requires implementations to support both, with UDP preferred for small messages but TCP required whenever a message exceeds the practical UDP datagram size, which is routine in Active Directory once the Privilege Attribute Certificate is included.

    Read more: Port 88 reference

  17. HTTP 401. What is the reason phrase?

    Answer: Unauthorized

    401 Unauthorized. No valid credentials for the target resource. The WWW-Authenticate header tells the client how to authenticate.

    Read more: HTTP Status Code Reference

  18. Which DNS record type does this? Maps an address back to a hostname (the reverse of an A/AAAA lookup) conventionally served from the in-addr.arpa (IPv4) or ip6.arpa (IPv6) reverse-lookup zones.

    Answer: PTR

    PTR (TYPE 12): a domain name pointer Example: 1.2.0.192.in-addr.arpa. PTR example.com.

    Read more: DNS Record Type Reference

  19. Port 135/TCP. What service is this?

    Answer: MSRPC Endpoint Mapper (DCE/RPC)

    135/tcp: MSRPC Endpoint Mapper (DCE/RPC). Windows' RPC locator service: a client connects to port 135 and asks which dynamic high port (typically 49152-65535 on modern Windows) a specific RPC interface (a WMI provider, a DCOM activation request, a remote service-control call) is actually listening on, then opens a second connection to that port to do the real work. DCOM is itself built on top of this same MSRPC transport.

    Read more: Port 135 reference

  20. HTTP 403. What is the reason phrase?

    Answer: Forbidden

    403 Forbidden. The server understood the request and refuses to fulfill it, unlike 401, authenticating (again) won't help; the client simply lacks permission for this resource.

    Read more: HTTP Status Code Reference

  21. Which DNS record type does this? Marks the start of a zone of authority (one SOA record per zone), and carries its primary name server, admin mailbox, serial number, and the refresh/retry/expire timers secondary servers use to stay in sync.

    Answer: SOA

    SOA (TYPE 6): marks the start of a zone of authority Example: example.com. SOA ns1.example.net. hostmaster.example.com. 2026072200 7200 3600 1209600 3600

    Read more: DNS Record Type Reference

  22. Port 139/TCP. What service is this?

    Answer: NetBIOS Session Service

    139/tcp: NetBIOS Session Service. NetBIOS Session Service provides the connection-oriented session layer NetBIOS applications use to exchange data. Historically this was how SMB file and print sharing operated, carried inside a NetBIOS session on 139, before Windows 2000 introduced direct-hosted SMB over TCP port 445 without a NetBIOS session wrapper. It is defined by RFC 1001/1002 and, per the IANA registry, formally listed on both TCP and UDP, though the session service itself is inherently connection-oriented and runs over TCP.

    Read more: Port 139 reference

  23. HTTP 404. What is the reason phrase?

    Answer: Not Found

    404 Not Found. The origin server has no current representation for the target resource, and deliberately isn't saying whether that's temporary or permanent.

    Read more: HTTP Status Code Reference

  24. Which DNS record type does this? Publishes the hostname, port, and relative priority/weight for a specific service, addressed via an underscore-prefixed "_service._proto.name" label instead of a plain hostname.

    Answer: SRV

    SRV (TYPE 33): Server Selection Example: _sip._tcp.example.com. SRV 10 60 5060 sipserver.example.com.

    Read more: DNS Record Type Reference

  25. Port 389/TCP/UDP. What service is this?

    Answer: LDAP (Lightweight Directory Access Protocol)

    389/tcp/udp: LDAP (Lightweight Directory Access Protocol). LDAP is the protocol clients use to query and modify a directory service, in a Windows environment, Active Directory's own interface for users, groups, computers, group policy, and schema. Port 389/tcp is its IANA-registered default and is defined by RFC 4511 (part of the RFC 4510 series). It also has a genuine UDP role, connectionless LDAP (CLDAP), which Active Directory domain controllers use to answer lightweight LDAP ping queries from the Windows DC Locator service before a client opens a full TCP LDAP session. Standard LDAP traffic on 389 is unencrypted unless upgraded via STARTTLS.

    Read more: Port 389 reference

  26. HTTP 407. What is the reason phrase?

    Answer: Proxy Authentication Required

    407 Proxy Authentication Required. The same idea as 401, but the client needs to authenticate to a proxy in the request path before that proxy will forward the request onward.

    Read more: HTTP Status Code Reference

  27. Which DNS record type does this? Holds arbitrary text data attached to a name. The RFC leaves the meaning of that text up to whatever's using it, and in practice that's grown to include domain-ownership verification and email-authentication policies such as SPF.

    Answer: TXT

    TXT (TYPE 16): text strings Example: example.com. TXT "v=spf1 -all"

    Read more: DNS Record Type Reference

  28. Port 443/TCP. What service is this?

    Answer: HTTPS (HTTP over TLS/SSL)

    443/tcp: HTTPS (HTTP over TLS/SSL). HTTPS wraps HTTP inside a TLS session, so the TCP handshake is followed by a TLS handshake (certificate exchange, cipher negotiation) before any HTTP request or response is exchanged. Port 443/tcp is IANA's registered default and is referenced from RFC 9110 as the default port for the https URI scheme. It now carries the overwhelming majority of web traffic, including HTTP/2 and, via QUIC over UDP on the same port number, HTTP/3.

    Read more: Port 443 reference

  29. HTTP 429. What is the reason phrase?

    Answer: Too Many Requests

    429 Too Many Requests. Too many requests in too little time. The standard rate-limiting code.

    Read more: HTTP Status Code Reference

  30. Which DNS record type does this? Restricts which certificate authorities are allowed to issue TLS certificates for the domain, and can name where a CA should report policy violations.

    Answer: CAA

    CAA (TYPE 257): Certification Authority Restriction Example: example.com. CAA 0 issue "letsencrypt.org"

    Read more: DNS Record Type Reference

  31. Port 445/TCP. What service is this?

    Answer: SMB (Server Message Block) / CIFS

    445/tcp: SMB (Server Message Block) / CIFS. Windows' (and Samba's) native file, printer, and named-pipe sharing protocol. Port 445 is "direct-hosted" SMB running straight over TCP without the legacy NetBIOS session service (ports 137-139) that older SMB required. Beyond plain file shares, SMB named pipes also carry a range of other Windows RPC-based administrative protocols such as service control, remote registry, and SAM access.

    Read more: Port 445 reference

  32. HTTP 500. What is the reason phrase?

    Answer: Internal Server Error

    500 Internal Server Error. The server hit an unexpected condition and couldn't fulfill the request. It's the catch-all when no more specific 5xx fits.

    Read more: HTTP Status Code Reference

  33. Port 1433/TCP. What service is this?

    Answer: Microsoft SQL Server

    1433/tcp: Microsoft SQL Server. The default port for Microsoft SQL Server's default database engine instance, carrying the Tabular Data Stream (TDS) protocol that clients use to authenticate and run queries. IANA registers the service name ms-sql-s to port 1433 for both TCP and UDP, but in real deployments the database engine itself listens on TCP; the companion UDP service on port 1434 is the separate SQL Server Browser/Resolution service that helps clients locate named instances using dynamic ports, not the query channel itself.

    Read more: Port 1433 reference

  34. HTTP 502. What is the reason phrase?

    Answer: Bad Gateway

    502 Bad Gateway. The server, while acting as a gateway or proxy, received an invalid response from an inbound server it was trying to fulfill the request through.

    Read more: HTTP Status Code Reference

  35. Port 3268/TCP. What service is this?

    Answer: Global Catalog LDAP (Microsoft Active Directory)

    3268/tcp: Global Catalog LDAP (Microsoft Active Directory). Port 3268/tcp is Microsoft's extension of LDAP for Active Directory's Global Catalog: a partial, read-only, forest-wide replica of every domain's directory data, hosted on domain controllers designated as Global Catalog servers. Querying 3268 instead of 389 lets a client search across every domain in the forest in a single query, at the cost of only a subset of each object's attributes being available. It is registered with IANA as msft-gc but is not defined by any IETF RFC; it is Active Directory-specific.

    Read more: Port 3268 reference

  36. HTTP 503. What is the reason phrase?

    Answer: Service Unavailable

    503 Service Unavailable. The server can't handle the request right now, usually from temporary overload or maintenance. Often paired with a Retry-After header.

    Read more: HTTP Status Code Reference

  37. Port 3389/TCP/UDP. What service is this?

    Answer: RDP (Remote Desktop Protocol)

    3389/tcp/udp: RDP (Remote Desktop Protocol). Microsoft's remote graphical desktop protocol (MS-RDPBCGR), historically a pure TCP protocol but, since RDP 8.0, able to negotiate an additional UDP transport (MS-RDPEUDP) for the same session, used to improve responsiveness for graphics, input, and audio when available, with TCP always present as the base/fallback channel. IANA registers ms-wbt-server on both TCP and UDP port 3389, and Microsoft's own documentation lists "TCP and UDP 3389" as the standard RDP port.

    Read more: Port 3389 reference

  38. Port 5985/TCP. What service is this?

    Answer: WinRM (Windows Remote Management): HTTP

    5985/tcp: WinRM (Windows Remote Management): HTTP. Microsoft's implementation of the WS-Management (WS-Man) standard; it is the transport underneath PowerShell Remoting (Enter-PSSession, Invoke-Command) and WMI-over-WinRM. Port 5985 is the default plaintext-HTTP listener. The SOAP payload itself is typically still protected at the message level by Kerberos/NTLM in a default domain configuration, but the outer HTTP channel is not TLS.

    Read more: Port 5985 reference

  39. Port 5986/TCP. What service is this?

    Answer: WinRM (Windows Remote Management): HTTPS

    5986/tcp: WinRM (Windows Remote Management): HTTPS. The same WS-Management/PowerShell Remoting service as port 5985, bound instead to a TLS-wrapped HTTPS listener backed by a certificate. It is the encrypted counterpart, commonly required when remoting across an untrusted network segment or to a non-domain-joined/workgroup host where Kerberos isn't available.

    Read more: Port 5986 reference

Subnets

  1. A new lab segment needs a single CIDR block that exactly covers the 16 addresses from 10.30.6.0 through 10.30.6.15, no more and no fewer. Build the block.

    Answer: 10.30.6.0/28

    16 addresses is 2⁴, so the prefix is /28 (32 − 4). 10.30.6.0/28 has network address 10.30.6.0 and broadcast address 10.30.6.15, exactly the requested range, with 10.30.6.16 and 10.30.5.255 falling just outside it in the neighboring blocks.

  2. 100.64.55.10 falls inside 100.64.0.0/10 (Shared Address Space). What category does this block belong to?

    Answer: Shared address space (CGNAT)

    Reserved for carrier-grade NAT (CGNAT) between an ISP's customer-facing network and its own NAT devices: routable only within that provider's network, distinct from RFC 1918. Seeing this in a log usually means the traffic traversed an ISP's CGNAT layer, not a private LAN.

    Read more: RFC 6598

  3. Team A's workstation pool must include 172.20.14.10 and 172.20.14.200. Team B owns the /24 blocks immediately before and after it. Build a single CIDR block that covers both of Team A's addresses without reaching into Team B's space.

    Answer: 172.20.14.0/24

    A /25 only reaches 172.20.14.0-.127, too small to cover .200. Bumping up to /24 gives 172.20.14.0-172.20.14.255, which covers both required addresses while still stopping cleanly at the neighboring /24s on either side.

  4. A firewall rule must match traffic from 10.50.8.64 and 10.50.8.95 only, not from 10.50.8.63, 10.50.8.96, or the unrelated address 10.50.9.1. Build the CIDR block.

    Answer: 10.50.8.64/27

    64 through 95 inclusive is 32 addresses: 2⁵, a /27. 10.50.8.64/27 aligns cleanly (64 is a multiple of 32), giving network 10.50.8.64 and broadcast 10.50.8.95, with .63 and .96 falling just outside in the neighboring /27s and .9.1 sitting in an entirely different /24.

  5. 169.254.10.5 falls inside 169.254.0.0/16 (Link Local (APIPA)). What category does this block belong to?

    Answer: Link-local

    Auto-assigned by Windows/most OSes when DHCP fails: seeing this on a host is a strong signal it never reached a DHCP server, not evidence of a specific attack on its own.

    Read more: RFC 3927

  6. A WAN link uses an RFC 3021 point-to-point IPv4 addressing scheme with just two host addresses and no separate network/broadcast address: 10.255.255.0 and 10.255.255.1. Build the smallest CIDR block containing both.

    Answer: 10.255.255.0/31

    RFC 3021 repurposes /31 for exactly this case: a 2-address block with no wasted network/broadcast address, both addresses usable as hosts. 10.255.255.0/31 covers precisely 10.255.255.0 and .1, with the neighboring addresses on either side falling outside it.

  7. An incident report references two hosts inside the same site allocation: 10.4.1.5 and 10.7.254.20. Build the smallest single CIDR block that contains both addresses.

    Answer: 10.4.0.0/14

    10.4.1.5 and 10.7.254.20 are far enough apart (third octet 4 vs. 7) that no /24, or even /16, covers both while staying aligned. The smallest block that does is 10.4.0.0/14, spanning 10.4.0.0 through 10.7.255.255. That's a real supernetting move: the covering block is much bigger than either address individually needs, but it's the smallest one that's actually valid.

  8. 192.0.2.55 falls inside 192.0.2.0/24 (Documentation (TEST-NET-1)). What category does this block belong to?

    Answer: Documentation

    Reserved purely for documentation/examples: never routed on the real internet. This site's own fabricated Email Header Analyzer sample uses this range for exactly that reason.

    Read more: RFC 5737

  9. An analyst wants a single rule that flags any traffic from IANA's benchmarking address space (RFC 2544), reserved for testing network-interconnect throughput so a benchmark never floods a real internet address. A benchmarking host lives at 198.18.0.1, another at 198.19.255.254. Build the CIDR block that covers the whole benchmarking range.

    Answer: 198.18.0.0/15

    IANA reserves 198.18.0.0/15 for benchmarking (RFC 2544). A /15 spans two consecutive /16s (198.18.0.0-198.19.255.255), exactly the range needed to cover both sample hosts while stopping at the surrounding address space.

    Read more: RFC 2544

  10. A report cites hosts at 203.0.113.10 and 203.0.114.250, addresses in two adjacent /24s from this site's own documentation-range examples. Build the smallest single CIDR block that contains both addresses.

    Answer: 203.0.112.0/22

    203.0.113.10 and 203.0.114.250 sit in two different /24s, so the smallest block that contains both is 203.0.112.0/22, spanning 203.0.112.0 through 203.0.115.255. Notice that block also swallows 203.0.112.0/24 and 203.0.115.0/24, neither of which was asked for. That is inherent to supernetting, not a mistake: there's no smaller single CIDR block that covers two addresses this far apart.

    Read more: Try it in the CIDR & VLAN Calculator's Supernet mode

Back to the drill