IP Services: NAT, DHCP, DNS, NTP, SNMP, Syslog & QoS
CCNA 200-301 · IP Services
IP Services is the smallest domain on the exam at 10%, but it covers nine objectives — the widest surface of any domain. The good news is that the questions are mostly about purpose and basics: what each service does, which port it uses, and the one or two commands that configure it. Nobody is asked to design a QoS policy or an SNMP architecture; they are asked to recognise the right service for a scenario and to read basic output. Learn each service's purpose, port, and key command, and this domain becomes easy points.
NAT: connecting private networks to the internet
NAT (Network Address Translation) lets devices with private RFC 1918 addresses reach the public internet by translating their private address to a public one at the edge router. The vocabulary matters: the inside local address is the device's private address, the inside global address is the public address it appears as on the internet, the outside global address is the public address of the internet destination, and the outside local address is how that destination appears inside the network (usually the same as the outside global). Static NAT maps one private address to one public address permanently, typically for a server that must be reachable from the internet. Dynamic NAT maps private addresses to a pool of public addresses on demand. PAT (Port Address Translation), also called NAT overload, maps many private addresses to a single public address by giving each connection a unique source port number — this is what home routers and small offices use, and it is why thousands of devices can share one public IP address.
Configuration marks the interfaces first — ip nat inside on the LAN-facing interface and ip nat outside on the internet-facing interface — then defines the translation: ip nat inside source static 192.168.1.10 203.0.113.10 for static NAT, or an access list plus ip nat inside source list 1 interface GigabitEthernet0/1 overload for PAT. Verify with show ip nat translations, which shows the current mappings, and show ip nat statistics.
DHCP, DNS, and NTP
DHCP automatically assigns IP addresses, subnet masks, default gateways, and DNS servers to hosts. The exchange has four steps, remembered as DORA: the client broadcasts a Discover, the server replies with an Offer, the client sends a Request, and the server confirms with an Acknowledgement. DHCP uses UDP ports 67 (server) and 68 (client). Because DHCP messages are broadcasts, they do not cross routers — so when the DHCP server sits on a different subnet from the clients, the router interface facing the clients needs a DHCP relay, configured with ip helper-address 10.1.1.5, which forwards the broadcast as a unicast to the server. A router can also act as a DHCP client itself (ip address dhcp on an interface) or as a small DHCP server (ip dhcp pool, with network, default-router, and dns-server commands, and ip dhcp excluded-address to reserve addresses).
DNS resolves human-readable names to IP addresses, using UDP port 53 for queries. Without DNS, every user would need to memorise IP addresses. On a Cisco device, ip name-server 8.8.8.8 sets the DNS server, and ip domain-lookup enables name resolution (many engineers disable it on lab routers with no ip domain-lookup to avoid long waits when mistyping a command). NTP (Network Time Protocol) synchronises clocks across the network over UDP port 123. Accurate time matters more than it seems: log timestamps, certificate validity, and troubleshooting all depend on every device agreeing on the time. NTP organises servers into stratum levels — stratum 0 is a reference clock such as an atomic clock or GPS receiver, stratum 1 servers sync directly to it, and each level further down increments the stratum number. A device acts as an NTP client with ntp server 192.168.1.1 and as a server with ntp master.
SNMP, syslog, and device management
SNMP (Simple Network Management Protocol) lets a management system monitor and manage network devices: it polls devices for information (interface status, CPU use, error counts) and receives traps — unsolicited alerts sent when something happens. SNMP uses UDP port 161 for manager-to-agent communication and UDP 162 for traps. Version 1 and version 2c authenticate with community strings sent in plaintext — effectively a password anyone on the wire can read — while SNMPv3 adds proper authentication and encryption and is the only version that should be used on a production network.
Syslog collects log messages from devices in a central place, so an engineer can search one server's logs instead of logging into every device. Syslog uses UDP port 514. Messages have a severity level from 0 to 7, and the exam tests this scale: 0 Emergency (system unusable), 1 Alert (immediate action needed), 2 Critical, 3 Error, 4 Warning, 5 Notification, 6 Informational, 7 Debugging. The lower the number, the more severe the message — and when you configure logging trap 4, you are telling the device to send messages of severity 4 and below (more severe), not 4 and above. Messages also carry a facility code identifying the source subsystem. Configure a syslog server with logging host 10.1.1.10.
Remote device management itself is part of this domain too: SSH (TCP 22) provides encrypted remote command-line access and should always be preferred over Telnet (TCP 23), which sends everything — including passwords — in plaintext. Setting up SSH requires a hostname, a domain name (ip domain-name example.com), an RSA key pair (crypto key generate rsa), a local username, and transport input ssh on the VTY lines. Console access is the physical fallback, and HTTP/HTTPS provide web-based management on devices that support it. Authentication for remote access can be local or centralised through TACACS+ or RADIUS servers, covered in the security guide.
QoS and file transfer
QoS (Quality of Service) manages the forwarding treatment of different traffic types when bandwidth is limited. The per-hop behaviour (PHB) at each router follows a pipeline the exam names explicitly: classification (identifying the traffic type), marking (tagging it with a priority value such as a DSCP value), queuing (deciding the order packets leave an interface), congestion management (deciding which packets to drop when a queue fills), policing (dropping traffic that exceeds a configured rate), and shaping (buffering excess traffic and sending it later instead of dropping it). The policing-versus-shaping distinction is the exam's favourite QoS question: policing drops, shaping delays. Voice traffic gets the highest priority marking (DSCP EF, value 46) because it cannot tolerate delay or jitter.
Finally, TFTP and FTP move files to and from network devices — most commonly IOS images and configuration files. TFTP uses UDP port 69, has no authentication, and is simple but insecure; it is the traditional choice for quick transfers inside a trusted network. FTP uses TCP ports 20 and 21, supports authentication, and is more reliable. The exam's point is the comparison: TFTP is UDP-based and unauthenticated, FTP is TCP-based and authenticated.
Exam traps to watch for
- Mixing up inside local and inside global: inside local is the private address, inside global is the public address it is translated to.
- Forgetting ip helper-address. DHCP broadcasts do not cross routers, so a DHCP server on another subnet needs a relay.
- Reading syslog severity backwards: 0 is the most severe (Emergency) and 7 is the least (Debugging). logging trap 4 sends levels 0-4.
- Confusing policing and shaping: policing drops excess traffic, shaping buffers it and sends it later.
Practice questions
Test yourself on this guide. Answers are checked instantly, with a full explanation for each question.
Q1.A host with private address 192.168.1.10 accesses the internet through a NAT router, which translates it to public address 203.0.113.5. What is 192.168.1.10 called in NAT terminology?
Explanation: The inside local address is the device's private address as it appears inside the local network — 192.168.1.10. Option A is wrong because the inside global address is the public address the device is translated to (203.0.113.5). Option C is wrong because the outside global address is the public address of the internet destination. Option D is wrong because the outside local address is how the destination appears inside the network.
Exam objective: 200-301 4.1: Configure and verify inside source NAT using static and pools
Q2.Which NAT type allows many private hosts to share a single public IP address by using unique source port numbers?
Explanation: PAT (overload) maps many inside local addresses to one inside global address, distinguishing the connections by source port number — this is how a home or small-office router shares one public address among many devices. Option A is wrong because static NAT is a permanent one-to-one mapping. Option B is wrong because dynamic NAT maps private addresses to a pool of public addresses, one public address per active translation. Option D is wrong because NAT64 translates between IPv6 and IPv4 networks.
Exam objective: 200-301 4.1: Configure and verify inside source NAT using static and pools
Q3.A DHCP server is on a different subnet from the clients it serves. What must be configured on the router interface facing the clients so they can obtain addresses?
Explanation: DHCP Discover messages are broadcasts, which routers do not forward. ip helper-address on the client-facing interface turns the router into a DHCP relay, forwarding the broadcasts as unicast packets to the DHCP server on the other subnet. Option A is wrong because ip dhcp pool makes the router itself a DHCP server, which is not what is needed when a server already exists elsewhere. Option C is wrong because NAT translates addresses and has nothing to do with DHCP relaying. Option D is wrong because NTP synchronises clocks, not DHCP messages.
Exam objective: 200-301 4.6: Configure and verify DHCP client and relay
Q4.What is the correct order of the four DHCP messages?
Explanation: The DORA sequence is: the client broadcasts a Discover, the server replies with an Offer, the client sends a Request for the offered address, and the server confirms with an Acknowledgement. Option A starts with a Request, which cannot come before the client has discovered a server. Option B has the client request before any offer has been made. Option C has the server offer before the client has discovered anything.
Exam objective: 200-301 4.3: Explain the role of DHCP and DNS within the network
Q5.Which NTP stratum level is closest to the reference clock, and what is stratum 0?
Explanation: Stratum 0 is the reference clock itself — an atomic clock, GPS receiver, or radio clock. Stratum 1 servers sync directly to a stratum 0 device, and each level further from the source increments the stratum number, so lower stratum numbers are more accurate. Option A reverses the scale. Option C swaps stratum 0 and stratum 1. Option D is wrong because accuracy decreases as stratum number increases.
Exam objective: 200-301 4.2: Configure and verify NTP operating in a client and server mode
Q6.A network engineer configures logging trap 4 on a router. Which syslog severity levels will be sent to the syslog server?
Explanation: logging trap sets the least severe level to send, and all more severe levels are included: trap 4 sends levels 0 (Emergency), 1 (Alert), 2 (Critical), 3 (Error), and 4 (Warning). Option A is backwards — it treats higher numbers as more severe, when the scale runs the other way. Option B is wrong because the trap level is a threshold, not a single level. Option D is wrong because it only includes the two most severe levels.
Exam objective: 200-301 4.5: Describe the use of syslog features, including facilities and severity levels
Q7.Which SNMP version should be used on a production network, and why?
Explanation: SNMPv3 adds proper authentication and encryption, while v1 and v2c send community strings in plaintext where anyone on the network can read them — so v3 is the only version suitable for production use. Option A is wrong because simplicity does not outweigh the security risk of plaintext credentials. Option B is wrong because 64-bit counters do not fix v2c's plaintext authentication. Option D is wrong because all SNMP versions use UDP, not TCP.
Exam objective: 200-301 4.4: Explain the function of SNMP in network operations
Q8.What is the difference between QoS policing and QoS shaping?
Explanation: Policing enforces a rate limit by dropping packets that exceed it, while shaping buffers the excess packets and transmits them when bandwidth is available, smoothing the traffic flow instead of discarding it. Option B reverses the two. Option C is wrong because both policing and shaping can apply to any traffic class. Option D is wrong because the two mechanisms have opposite effects on excess traffic.
Exam objective: 200-301 4.7: Explain the forwarding per-hop behavior (PHB) for QoS — policing and shaping
Q9.Which steps are required before a Cisco device can accept SSH connections?
Explanation: SSH needs a hostname and a domain name (the two are combined to name the RSA key), and an RSA key pair generated with crypto key generate rsa — without a key, the device cannot offer SSH at all. Option A configures DHCP services, which are unrelated to SSH. Option C configures syslog. Option D configures NAT.
Exam objective: 200-301 4.8: Configure network devices for remote access using SSH
Q10.Which statement correctly compares TFTP and FTP?
Explanation: TFTP is a simple UDP-based protocol (port 69) with no authentication, traditionally used for transferring IOS images and configs inside a trusted network; FTP is TCP-based (ports 20/21) and supports username/password authentication. Option A reverses the two protocols. Option C is wrong because TFTP uses UDP 69, not TCP 21. Option D is wrong because neither protocol transfers email or web pages.
Exam objective: 200-301 4.9: Describe the capabilities and functions of TFTP/FTP in the network