Interview questionsUpdated 12 Oct 2026 · Team InterviewPlan

Network Engineer Interview Questions and Answers

Network engineer interviews test fundamentals, routing and switching depth and calm troubleshooting during outages. In 2026, panels also ask about SD-WAN, cloud networking and automation. These questions reflect what network engineers are asked at telecoms, managed service providers, enterprises and data centre teams.

What interviewers test in a network engineer interview

Interviews often begin with fundamentals and subnetting, then move into routing protocols, security and a troubleshooting scenario. Some include a lab task or show configuration snippets and ask what is wrong.

Interviewers look for structured troubleshooting, change discipline and the ability to explain complex topics simply to non-network colleagues.

  • Fundamentals: OSI, TCP and UDP, ARP, DHCP and DNS
  • Switching: VLANs, trunks, STP, EtherChannel and port security
  • Routing: static, OSPF, BGP, route redistribution and policies
  • Security: ACLs, firewalls, NAT, VPNs and segmentation
  • Modern networking: SD-WAN, cloud VPCs and VNets, and automation

Technical topics to revise

Use Cisco Packet Tracer, GNS3 or EVE-NG to build a lab with VLANs, OSPF areas, a BGP peering and a site-to-site VPN. Break it and fix it, writing down the show commands you used.

  • Subnetting and CIDR calculations without a calculator
  • STP root bridge election, PortFast and BPDU Guard
  • OSPF neighbour states, areas, LSA types and DR and BDR election
  • BGP path selection attributes: weight, local preference, AS path and MED
  • HSRP, VRRP and first-hop redundancy
  • IPsec phases, IKEv2 and common tunnel failures
  • Palo Alto, Fortinet or Cisco ASA firewall policies
  • Python with Netmiko or Ansible for configuration automation

How to structure your answers

For troubleshooting, follow the OSI model bottom up or a divide-and-conquer approach, naming the exact commands you would run, such as show interface, show ip route or a packet capture.

For project stories, describe the network scale, the change, how you planned and rolled it back if needed, and the outcome, like reduced downtime or improved bandwidth.

How InterviewPlan AI helps you prepare

Network engineering mock interviews ask fundamentals, routing and troubleshooting scenarios and give feedback on structure. Resume AI helps express projects in outcomes, such as uptime and migration results.

Interview Copilot provides real-time speaker-labelled transcription and resume-grounded drafts on Zoom, Teams, Meet, Webex and phone interviews. The free plan includes 5 live Copilot minutes and 100 AI credits, no card required.

Questions and model answers

9 questions you are likely to hear.

  1. A user cannot access a web application. How do you troubleshoot?

    Why they ask: This open scenario tests a structured troubleshooting method.

    Sample answer: I start by scoping: is it one user, one site or everyone, and is it one app or all apps. Then I go layer by layer. I check physical link and IP configuration with ipconfig, then ping the gateway, then test DNS with nslookup. If DNS resolves, I test connectivity to the server IP and the port with tools like Test-NetConnection or telnet. If that fails, I check firewall logs and ACLs along the path and use traceroute to find where traffic stops. In one case, the issue was a DNS record pointing to an old load balancer IP after a migration.

  2. How many usable hosts are in a /27 subnet, and what are the ranges for 192.168.10.0/27?

    Why they ask: Subnetting is a basic but essential skill tested in almost every interview.

    Sample answer: A /27 leaves 5 host bits, so there are 32 addresses per subnet and 30 usable hosts after removing the network and broadcast addresses. For 192.168.10.0/27, the first subnet is 192.168.10.0 with usable hosts from .1 to .30 and broadcast .31. The next subnets start at .32, .64, .96 and so on, in blocks of 32. The subnet mask is 255.255.255.224. I use this kind of sizing when planning VLANs for small departments, leaving room for growth instead of choosing the tightest fit.

  3. How does OSPF form neighbour relationships, and what stops it?

    Why they ask: OSPF is the most common interior routing protocol and adjacency issues are frequent.

    Sample answer: OSPF routers send hello packets on enabled interfaces. To become neighbours, they must match area ID, subnet and mask, hello and dead timers, authentication settings and stub area flags, and have unique router IDs. They move through states from Down and Init to 2-Way, then ExStart, Exchange and Loading to Full. On broadcast networks, routers form full adjacency only with the DR and BDR. Common blockers are mismatched timers, MTU mismatches that leave neighbours stuck in ExStart, and ACLs blocking multicast. I once fixed a branch stuck in ExStart caused by an MTU mismatch on a new WAN circuit.

  4. Explain BGP path selection.

    Why they ask: BGP knowledge is essential for WAN, ISP and data centre roles.

    Sample answer: BGP chooses the best path using attributes in order. On Cisco, it first prefers the highest weight, which is local to the router, then the highest local preference across the AS, then locally originated routes, then the shortest AS path, then the lowest origin type, then the lowest MED from the same neighbouring AS, then eBGP over iBGP, then the lowest IGP metric to the next hop, and finally tie-breakers like router ID. In a dual-ISP design, I used local preference to make one ISP primary for outbound traffic and AS path prepending to influence inbound traffic.

  5. What is the difference between a VLAN and a subnet, and how does inter-VLAN routing work?

    Why they ask: Switching and routing integration is a common fundamentals topic.

    Sample answer: A VLAN is a Layer 2 broadcast domain created on switches, while a subnet is a Layer 3 IP range. Usually each VLAN maps to one subnet. Devices in different VLANs cannot talk without routing. Inter-VLAN routing is done either with a router on a stick, using subinterfaces on a trunk, or more commonly with SVIs on a Layer 3 switch, where each VLAN has a virtual interface acting as the gateway. In an office redesign, I moved routing from a router on a stick to SVIs on core switches, which removed a bottleneck on the single trunk link.

  6. A site-to-site VPN tunnel is down. How do you troubleshoot it?

    Why they ask: VPN issues are a frequent operational task.

    Sample answer: I check whether phase 1 is up, then phase 2. For phase 1, I verify that peers can reach each other on UDP 500 and 4500, and that IKE version, encryption, hash, DH group, lifetime and pre-shared key or certificates match on both sides. For phase 2, I check that the proxy IDs or encryption domains match exactly, along with transform sets and PFS settings. I also check NAT exemption rules and routes for the remote subnets. Logs from both firewalls usually point to the mismatch. A common case I fixed was a partner changing their subnet without updating our side.

  7. How does SD-WAN differ from traditional WAN?

    Why they ask: SD-WAN is now common, and interviewers check modern networking knowledge.

    Sample answer: Traditional WAN relies mostly on MPLS circuits with routing configured device by device. SD-WAN abstracts multiple transports like MPLS, broadband and LTE under a centrally managed controller. It steers traffic per application based on real-time measurements of latency, jitter and loss, and uses encrypted overlays between sites. This improves performance for cloud apps and lowers cost by using cheaper internet links. I was part of a rollout for about 50 branches, where we kept MPLS for voice initially and moved Microsoft 365 traffic to local internet breakout, which improved user experience noticeably.

  8. How do you use automation in network operations?

    Why they ask: Automation skills are increasingly expected.

    Sample answer: I use Python with Netmiko or NAPALM for tasks like collecting show outputs and backing up configs across hundreds of devices, and Ansible playbooks for standard configuration changes, such as updating NTP or SNMP settings. Templates with Jinja2 keep configurations consistent. Changes go through Git for review and version history. In my last role, an Ansible playbook replaced a manual process of updating ACLs on around 120 branch routers, which used to take two engineers a full night, and reduced configuration errors.

  9. Tell me about a major network outage you handled.

    Why they ask: This reveals calm under pressure, process and communication.

    Sample answer: A data centre lost connectivity to several branches during business hours. I joined the bridge, confirmed the scope and checked the core routers. BGP sessions with one MPLS provider had dropped and traffic was not failing over, because a route map on the backup link filtered the branch prefixes after a recent change. I reverted the route map through the change rollback plan, and connectivity restored within about 30 minutes. In the review, we added pre-change testing of failover paths and updated the change template to require a backup path check.

FAQ

Questions people ask about network engineer.

Is CCNA enough for network engineer interviews?

CCNA is a strong foundation for junior roles. Mid and senior roles often expect CCNP-level knowledge in routing, switching or security, plus vendor experience with Palo Alto, Fortinet or Juniper. Lab practice matters as much as certifications.

Do network engineers need cloud knowledge in 2026?

Increasingly, yes. Know AWS VPC and Azure VNet basics, connectivity with VPN and Direct Connect or ExpressRoute, security groups and routing. Many enterprise networks now extend into the cloud, so hybrid routing questions are common.

Should network engineers learn Python?

Yes. Basic Python with libraries like Netmiko, plus Ansible, helps automate repetitive tasks and is now listed in many job descriptions. Interviewers may ask what you have automated and how you test changes.

Can InterviewPlan AI help with phone interviews?

Yes. Interview Copilot works on phone calls as well as Zoom, Teams, Meet and Webex, giving real-time transcription and resume-based drafts. Follow the employer's interview rules. The free plan includes 5 live minutes.

P.S.

Reading helps. Practising out loud helps more.

Start free with 5 live minutes and 100 AI credits for Resume AI, Job Search, Auto-Apply, mock interviews and Behavioral Coach.