Home › Blog › Network Engineer Interview Questions

Network Engineer Interview Questions: 30 Questions With Sample Answers

Fundamentals, subnetting, routing, switching, security and troubleshooting questions, with what interviewers listen for in each answer.

Network engineer interview questions test whether you understand how traffic actually moves (the OSI and TCP/IP models, addressing, routing and switching), whether you can subnet quickly, and whether you can find a fault methodically when something breaks. The three questions you are most likely to get are "Walk me through the OSI model", "What happens when you type a URL into a browser?" and "A user can't reach a server. How do you troubleshoot it?" This guide gives you 30 questions grouped by topic, what a strong answer shows for each, five sample answers we wrote, and protocol facts checked against the RFCs and Cisco's own pages as of October 2026.

How the network engineer interview usually runs

Stages vary by employer. An internet provider or data-center team will go deep on routing protocols; an enterprise IT team may focus on campus switching, wireless and firewalls; a cloud team will ask about virtual networks and automation. Ask the recruiter what the network looks like and which vendors it runs.

StageWhat it usually coversHow to prepare
Recruiter screenVendors, scale, certifications, on-callName the gear and protocols you ran, and the network size
Technical screenOSI, TCP/UDP, subnetting, VLANs, basic routingSubnet on paper quickly and explain each layer
Deep technicalOSPF, BGP, spanning tree, redundancy, securityKnow how each protocol chooses a path and fails over
Troubleshooting or labA broken topology or a scenarioWork bottom-up and say what each command proves
DesignA branch office, a campus or a data-center networkLead with requirements, redundancy and addressing
BehavioralOutages, change windows, working with other teamsA story about a change that went wrong and the fix

Cisco's associate-level certification is the CCNA, exam 200-301 (Implementing and Administering Cisco Solutions). As of October 2026, Cisco lists it as a 120-minute exam covering network fundamentals, network access, IP connectivity, IP services, security fundamentals, and automation and programmability (Cisco CCNA page). Its topic list is a reasonable syllabus for these interviews even if you don't take the exam.

Fundamentals and addressing questions

  1. Walk me through the OSI model. A strong answer names all seven layers and gives a real protocol or device at each, rather than just reciting names.
  2. What is the difference between TCP and UDP? Shows connection setup, ordering and retransmission versus low overhead, with examples like DNS and voice over UDP.
  3. Explain the TCP three-way handshake. Shows SYN, SYN-ACK, ACK, and that each side's initial sequence number is exchanged. TCP is now specified in RFC 9293, which obsoletes RFC 793 (RFC 9293).
  4. What does ARP do? Shows mapping an IPv4 address to a MAC address on the local segment, and what a stale or spoofed ARP entry causes.
  5. How does DHCP assign an address? Shows Discover, Offer, Request, Acknowledge, and the need for a relay when the server is on another subnet.
  6. What are the private IPv4 ranges? Shows 10.0.0.0/8, 172.16.0.0/12 and 192.168.0.0/16 from RFC 1918 (RFC 1918).
  7. How many usable hosts are in a /26? Shows 64 addresses minus network and broadcast: 62 usable.
  8. What is NAT, and what is PAT? Shows address translation at the edge, and many inside hosts sharing one public address using ports.
  9. How does IPv6 differ from IPv4 for a network engineer? Shows 128-bit addresses, no broadcast, neighbor discovery instead of ARP, and SLAAC alongside DHCPv6.

Routing and switching questions

  1. What is a VLAN, and what is a trunk? Shows separate broadcast domains on shared switches, and 802.1Q tagging on links that carry several VLANs.
  2. Why does spanning tree exist? Shows that redundant layer 2 links create loops and broadcast storms, and STP blocks ports to keep one loop-free path.
  3. How do devices in different VLANs talk to each other? Shows inter-VLAN routing on a layer 3 switch or a router-on-a-stick.
  4. Static versus dynamic routing: when do you use each? Shows static for small or stub networks and defaults, dynamic for scale and automatic failover.
  5. How does OSPF choose a path? Shows a link-state protocol where each router builds a map of the area and runs a shortest-path calculation on link cost.
  6. Why does OSPF use areas? Shows smaller link-state databases, less churn from flapping links, and summarization at area borders.
  7. How is BGP different from OSPF? Shows BGP as a path-vector protocol between autonomous systems, chosen by policy and attributes, not just shortest path. BGP runs over TCP port 179 (RFC 4271).
  8. Two routing protocols learn the same route. Which wins? Shows administrative distance (on Cisco devices by default: connected 0, static 1, eBGP 20, internal EIGRP 90, OSPF 110, RIP 120), and longest prefix match before that.
  9. How do you give hosts a gateway that survives a router failure? Shows a first-hop redundancy protocol such as HSRP or VRRP with a shared virtual IP.

Security, services and design questions

  1. Stateful firewall versus ACL? Shows that a stateful firewall tracks connections and allows return traffic automatically, while a basic ACL checks each packet on its own.
  2. How would you segment a network for security? Shows zones (users, servers, management, guests), firewalling between them and least access between zones.
  3. Site-to-site VPN versus remote-access VPN? Shows connecting networks versus connecting individual users, and IPsec or TLS-based options.
  4. How does DNS resolution work? Shows stub resolver, recursive resolver, root, TLD and authoritative servers, caching and TTLs, and DNS on port 53 over UDP and TCP.
  5. What is QoS, and when do you need it? Shows classifying and prioritizing delay-sensitive traffic like voice on congested links.
  6. Design the network for a 200-person branch office. Shows requirements first, redundant uplinks, VLAN plan, addressing scheme, wireless, security and how it connects to headquarters or the cloud.
  7. How have you automated network changes? Shows scripts or tools that push config from templates, with review and backups, instead of typing changes into each device.

Troubleshooting and behavioral questions

  1. A user can't reach a server. How do you troubleshoot? Shows a layer-by-layer method from physical link and IP config up to DNS and the application.
  2. The network is slow for one site. Where do you start? Shows checking interface errors, utilization, duplex mismatches, packet loss and recent changes.
  3. Tell me about an outage you caused or fixed. Shows ownership, how you found the cause, and what you changed in your change process.
  4. How do you plan a change in a maintenance window? Shows a written plan, pre-checks, a tested rollback, a time limit and post-checks.
  5. Tell me about a disagreement with another team over a network issue. Shows using packet captures and data to settle "it's the network" debates without blame.

Prepare these stories with the free STAR story builder.

Sample answers to common network engineer interview questions

These are examples we wrote. Replace the details with your own network and numbers.

"Walk me through the OSI model."

Example (entry level): "Layer 1 is physical, cables and signals. Layer 2 is data link, where switches forward frames by MAC address and VLANs live. Layer 3 is network, IP addressing and routing. Layer 4 is transport, TCP and UDP with ports. Layers 5 and 6, session and presentation, handle things like session setup and encryption formats, and in practice they're often folded into the application. Layer 7 is the application, like HTTP or DNS. When I troubleshoot, I use it as a checklist from the bottom up: link light, then IP and gateway, then whether the port is reachable, then the application."

"A user can't reach a server. How do you troubleshoot?"

Example (mid-level): "First I'd scope it: one user or many, one server or everything. Then I'd go bottom-up from the user's side. Is the link up and does the machine have a valid IP, mask and gateway? Can it ping the gateway? Does the server's name resolve to the right address? Can it reach the server's IP, and if not, where does a traceroute stop? If the IP is reachable, I'd test the actual port, because a firewall rule or the service itself is the usual culprit at that point. I'd also ask what changed recently, because a change is the most common cause."

"How many hosts and subnets do you get if you split 192.168.10.0/24 into /26s?"

Example (subnetting): "A /26 borrows two bits from the host part, so you get four subnets of 64 addresses each: .0, .64, .128 and .192. Each has 62 usable hosts, because the first address is the network and the last is broadcast. So the first subnet runs from .1 to .62 for hosts, with .63 as broadcast, and the next network starts at .64."

"How is BGP different from OSPF?"

Example (experienced hire): "OSPF is an interior gateway protocol. Every router in an area shares link-state information, builds the same map, and picks the lowest-cost path, and it converges fast inside one organization. BGP is the exterior protocol between autonomous systems. It runs over TCP and makes decisions with path attributes and policy, like local preference and AS path length, because between organizations you care about business rules, not just the shortest path. In my last role we ran OSPF inside the data center and eBGP to two upstream providers, with local preference to prefer the cheaper link."

"Tell me about an outage you caused or fixed."

Example (STAR): "During a planned change I added a new VLAN to a trunk but typed the allowed-VLAN command without the add keyword, which replaced the list. Two floors lost connectivity. Because I was watching monitoring as part of the post-checks, I saw it within a couple of minutes, restored the saved config and everyone was back in under ten minutes. I told my manager right away and wrote it up. We added a config diff review to every change plan, and I started pasting changes from a reviewed template instead of typing them live."

What interviewers listen for, by skill

SkillSample questionWhat they are listening for
FundamentalsWalk through the OSI modelReal protocols and devices at each layer
AddressingSplit a /24 into /26sFast, correct math and the usable-host rule
SwitchingWhy does spanning tree exist?Loops, broadcast storms, blocked ports
RoutingBGP versus OSPFLink-state versus path-vector, policy versus cost
SecuritySegment this networkZones, firewalling, least access
TroubleshootingA user can't reach a serverA bottom-up method and "what changed?"
Change disciplinePlan a maintenance windowPre-checks, rollback, post-checks

Questions to ask the interviewer

More ideas are in our list of questions to ask the interviewer.

How to practice network engineer interview questions

  1. Subnet ten random prefixes on paper a day until you can do a /27 or /29 in your head.
  2. Explain the OSI model, the TCP handshake and DNS resolution out loud in under a minute each.
  3. Build a small lab with OSPF and two VLANs in a free simulator, break it, and fix it while narrating.
  4. Prepare one outage story and one change-planning story with real details.

CoPilot Interview's Mock Interview builds questions from your résumé and the job description, reads them aloud and scores each spoken answer out of 10 with what to fix and a model answer. You can start with the free browser mock interview. If the role touches cloud networking, also review cloud engineer interview questions and AWS interview questions; for Linux-side debugging, see Linux interview questions, and for security-heavy roles, security engineer interview questions.

Say your troubleshooting steps out loud

Network rounds reward a clear method. The free browser mock interview lets you answer by voice, get a score, and see what to tighten.

Try a free AI mock interview

FAQ

What are the most common network engineer interview questions?

The most common are the OSI model, TCP versus UDP, the TCP three-way handshake, subnetting a prefix, VLANs and trunks, OSPF versus BGP, how DNS works, and how you would troubleshoot a user who cannot reach a server.

How do I prepare for a network engineer interview?

Review fundamentals and subnetting until they are fast, learn how OSPF, BGP and spanning tree choose paths and fail over, practice a bottom-up troubleshooting method out loud, and prepare stories about outages and planned changes.

Is CCNA required for a network engineer job?

It depends on the employer. Many entry-level postings ask for it, and its exam topics are a solid interview syllabus. As of October 2026 the CCNA exam is 200-301, a 120-minute exam covering fundamentals, access, IP connectivity, IP services, security and automation.

How much subnetting is in a network engineer interview?

Expect at least one subnetting question, such as how many usable hosts are in a /26 (62) or how to split a /24 into smaller subnets. Interviewers want quick, correct math without a calculator.

What troubleshooting method should I describe?

Describe a bottom-up method: scope the problem, check the physical link and IP settings, test the gateway, check DNS, test reachability and the specific port, and ask what changed recently. Say what each step proves.

How can I practice network engineer interview questions for free?

Use a free network simulator to build and break small labs, and answer questions out loud while recording yourself. CoPilot Interview's free browser mock interview scores your spoken answers.