NAT practice questions with answers
Fifteen free NAT practice questions with explained answers, written for anyone preparing for the Cisco CCNA. Network Address Translation (NAT) rewrites IP addresses as packets cross a router, so private inside hosts can reach public networks; the CCNA tests inside source NAT with static entries, pools and PAT (overload). The topic appears in the IP Services domain of CCNA 200-301. All questions are original and use Cisco IOS behavior.
How to use these questions
Answer each question before opening the explanation, and commit to one letter. Underline the constraint word in the stem, such as “many hosts”, “permanent” or “without overload”, because it usually picks the NAT type. Read the explanation even when the answer was right, since each one also says why the other three options fail. The concepts behind the questions are covered in the NAT types topic guide.
Warm-up: terms and NAT types
Question 1. A router performs inside source NAT. The inside host 10.1.1.10 is translated to 203.0.113.5 when it connects to the Internet server 198.51.100.7. Which term describes the address 203.0.113.5?
- A. Inside local
- B. Inside global
- C. Outside local
- D. Outside global
Show answer
Answer: B
The inside global address is how an inside host appears to the outside world, here 203.0.113.5. The inside local address is the host real address, 10.1.1.10. The server address 198.51.100.7 is an outside address (outside global, and outside local too unless outside NAT is used). A router interface address is not a NAT term at all.
Question 2. A web server at 192.168.10.20 must always be reachable from the Internet at the same public address, 203.0.113.20, and the mapping must exist permanently, even when no traffic is flowing. Which NAT type fits?
- A. PAT (overload) on the WAN interface
- B. Dynamic NAT with a pool of public addresses
- C. Dynamic NAT for outside source addresses
- D. Static NAT
Show answer
Answer: D
Static NAT creates a permanent one-to-one mapping that exists whether or not traffic flows, so outside hosts can initiate connections to the server. Dynamic NAT builds entries on demand from a pool when inside hosts send traffic, so nothing exists for an outside host to connect to. PAT also creates entries only from inside traffic and shares addresses by port. Outside source NAT translates addresses of outside hosts, not the inside server.
Question 3. An office has 200 PCs and one public IPv4 address that the ISP assigns to the router WAN interface GigabitEthernet0/0/1 by DHCP. Which command lets all PCs share that address? Assume ACL 10 permits the inside network.
- A. ip nat inside source list 10 interface GigabitEthernet0/0/1 overload
- B. ip nat inside source list 10 interface GigabitEthernet0/0/1
- C. ip nat outside source list 10 interface GigabitEthernet0/0/1 overload
- D. ip nat inside source static 10.0.0.0 interface GigabitEthernet0/0/1
Show answer
Answer: A
ip nat inside source list 10 interface GigabitEthernet0/0/1 overload translates addresses matched by ACL 10 to the interface address and tells the router to use unique port numbers (PAT), so one address serves many hosts. Without overload, the interface address could be used by one inside host at a time. ip nat outside source translates outside hosts, which is not the goal. A static entry maps one inside host only and would not serve 200 PCs.
Question 4. A technician configures ACL 1 for the inside network and the command ip nat inside source list 1 interface GigabitEthernet0/0/1 overload. Hosts generate Internet traffic, but show ip nat translations stays empty. The running configuration shows no ip nat commands under any interface. What is missing?
- A. The overload keyword
- B. A deny statement in ACL 1
- C. ip nat inside and ip nat outside on the LAN and WAN interfaces
- D. A loopback interface with the public address
Show answer
Answer: C
The router must know which interfaces face the inside and the outside. ip nat inside on the LAN interface and ip nat outside on the WAN interface mark the direction in which translation happens; without them the global command never matches any traffic. Overload is already present. Changing the ACL to deny would stop hosts from being translated at all. A loopback interface plays no part in this translation.
Question 5. Dynamic NAT is configured without overload: ip nat inside source list 1 pool PUB, where pool PUB has three addresses. Three inside hosts already hold active translations. A fourth inside host sends its first packet to the Internet. What happens?
- A. It shares one of the pool addresses by using different port numbers
- B. The packet is dropped until a translation entry is freed
- C. It is forwarded with its private source address
- D. The router borrows the WAN interface address for it
Show answer
Answer: B
Without overload, each translation uses a whole pool address, so with three addresses in use the pool is exhausted. The packet from the fourth host is dropped (the router returns an ICMP host unreachable) until an existing entry times out or is cleared. NAT does not reuse an address by adding port numbers unless overload is configured. Forwarding the packet untranslated and borrowing the interface address are not behaviors of a pool without overload.
Scenarios: reading and building configurations
Question 6. The output of show ip nat translations on a router contains this entry (columns: Pro, Inside global, Inside local, Outside local, Outside global): tcp 203.0.113.5:1024 10.1.1.10:51000 198.51.100.7:80 198.51.100.7:80. Which statement is correct?
- A. Host 10.1.1.10 was translated to 203.0.113.5 and its source port 51000 was changed to 1024
- B. 198.51.100.7 is the inside global address
- C. This is a static one-to-one translation
- D. 203.0.113.5:1024 is the address of the server
Show answer
Answer: A
The second column is the inside global address and port, the third the inside local. Host 10.1.1.10 used source port 51000, and the router rewrote it to port 1024 on 203.0.113.5, which is PAT behavior. 198.51.100.7 is in the outside columns, not an inside global address. A static one-to-one entry would show no port numbers. 203.0.113.5:1024 belongs to the translated inside host, not to the server.
Question 7. A router uses ip nat inside source list 1 interface GigabitEthernet0/0/1 overload with access-list 1 permit 10.1.1.0 0.0.0.255. A new host at 10.1.2.5 on another inside subnet, also routed through this router, sends a packet to the Internet. What happens to the packet?
- A. The router drops it and sends an ICMP unreachable to the host
- B. It is translated to the WAN interface address because the interface is configured for overload
- C. The router refuses to forward it because of the implicit deny in the ACL
- D. It is forwarded untranslated with the source address 10.1.2.5
Show answer
Answer: D
In NAT the ACL only selects which inside local addresses are translated; it is not a filter. 10.1.2.5 does not match the permit statement, so the NAT process ignores it and the packet is routed normally with its private source address, where the ISP will typically discard it. NAT does not send an ICMP error for unmatched traffic. The interface address is used only for hosts that match the ACL. The router has no reason to refuse to forward the packet.
Question 8. The internal server 10.1.1.20 must be reachable from the Internet as 203.0.113.20. Which command creates the mapping?
- A. ip nat outside source static 10.1.1.20 203.0.113.20
- B. ip nat inside source static 203.0.113.20 10.1.1.20
- C. ip nat inside source static 10.1.1.20 203.0.113.20
- D. ip nat inside source list 10.1.1.20 pool 203.0.113.20
Show answer
Answer: C
The syntax is ip nat inside source static followed by the inside local address and then the inside global address, so ip nat inside source static 10.1.1.20 203.0.113.20 is correct. The version with the two addresses reversed describes a different mapping. ip nat outside source static is for translating outside hosts. Specifying an address where an ACL number or name belongs is invalid syntax for a list-based command.
Question 9. A branch router has the command ip nat inside source list 1 interface GigabitEthernet0/0/1 (no other NAT commands). The first PC browses normally, but a second PC cannot reach the Internet until the first has been idle for a long time. What fixes the problem?
- A. Remove access-list 1
- B. Add the overload keyword to the command
- C. Add ip nat outside on the LAN interface
- D. Create a second NAT pool for the second PC
Show answer
Answer: B
Without overload this is dynamic NAT that maps one inside host to the single interface address, so only one host can hold a translation at a time. Adding overload enables PAT, which lets many hosts share the address by using different port numbers. Removing the ACL would stop translation. ip nat outside is already needed and present, since the first PC works. A second pool cannot help because the interface provides only one address.
Question 10. An administrator must remove all dynamic NAT entries immediately, while keeping the NAT configuration and any static entries. Which command does this?
- A. clear ip nat translation *
- B. clear ip nat statistics
- C. no ip nat inside on the LAN interface
- D. clear ip route *
Show answer
Answer: A
clear ip nat translation * removes all dynamic translation entries; static entries created with ip nat inside source static stay. clear ip nat statistics only resets the counters and leaves the table alone. Removing ip nat inside from the interface changes the configuration and stops translation. clear ip route * affects the routing table, not the NAT table.
Edge cases and traps
Question 11. Two inside hosts, 10.1.1.10 and 10.1.1.11, both open connections to the same server 198.51.100.7:443 using the same source port 50000. The router performs PAT to a single address. How does it keep the two sessions apart?
- A. It drops the second session until the first one ends
- B. It uses the source MAC addresses of the hosts
- C. It changes the source port of one flow to an unused port on the shared address
- D. It automatically assigns a second public address to the second host
Show answer
Answer: C
PAT keeps one translation entry per flow. For the second flow it rewrites the source port to an unused port on the inside global address, so the return traffic can be mapped back to the correct inside host. It does not drop the second session. NAT works on IP addresses and Layer 4 ports and does not use MAC addresses for this. It does not take a second public address, because overload shares the one it has.
Question 12. Two inside hosts each ping the same Internet host at the same time through a router doing PAT to one address. ICMP echo has no TCP or UDP port numbers. What does the router use to distinguish the replies?
- A. The destination port 7
- B. The source MAC address
- C. The TTL field
- D. The ICMP identifier (query ID)
Show answer
Answer: D
For ICMP echo, Cisco IOS translates the ICMP query identifier instead of a port number, so each echo reply can be matched to its translation entry. Port 7 is the echo service port for the old UDP and TCP echo protocols and is not part of ICMP. MAC addresses are rewritten at every router hop and are not used for translation. The TTL value is decremented at each hop and does not identify a flow.
Question 13. Hosts on 10.1.1.0/24 reach the Internet through a router that performs NAT. An ACL is applied inbound on the inside LAN interface. Which source address do the ACL entries match for traffic from these hosts?
- A. The inside local address
- B. The inside global address
- C. The outside local address
- D. The outside global address
Show answer
Answer: A
Traffic from the inside enters the router on the inside interface before NAT is applied, so an inbound ACL there sees the original inside local address, such as 10.1.1.10. The inside global address is created only when the packet is translated on its way out, after the inbound ACL check. The outside local and outside global addresses describe the destination server, not the source.
Question 14. A router shows this entry in show ip nat translations: Pro and port columns display --- 203.0.113.20 10.1.1.20 --- ---. What can be concluded?
- A. It is a PAT entry for a TCP session
- B. It is an outside source NAT entry
- C. It was created by the overload keyword
- D. It is a one-to-one translation, not PAT
Show answer
Answer: D
An entry without protocol and port numbers is a one-to-one address translation, created either by a static command or by dynamic NAT without overload. PAT entries always carry protocol and port values, so this is not PAT or an overload entry. Nothing in it indicates outside source NAT, because the outside columns hold no translation. The entry cannot be called a TCP entry, since the protocol field is empty.
Question 15. Static NAT maps the inside server 10.1.1.20 to 203.0.113.20. An ACL is applied inbound on the outside WAN interface to allow only HTTPS to the server. For packets arriving from the Internet, which destination address must the permit entry match?
- A. 10.1.1.20, the inside local address
- B. 10.1.1.0/24, the inside subnet
- C. 203.0.113.20, the inside global address
- D. The WAN interface address of the router
Show answer
Answer: C
Traffic from the outside is checked by the inbound ACL on the outside interface before the outside-to-inside translation, so the packet still carries the inside global destination address 203.0.113.20 and the ACL must permit that address. Matching 10.1.1.20 would fail, because the packet has not been translated yet. The server real subnet and the router interface address are not the destination of these packets.
Where to go next
Review the concepts in the NAT types topic guide. For related drills, try the ACL practice questions and the subnetting practice problems. When these feel easy, take the free CCNA 20-question practice test and read the multiple-choice exam tips on handling constraint words under time pressure.