SOC Analyst Interview Questions: Protocols, Wireshark, Linux and Investigation Reports
The second half of Module 1 moves from theory to investigation: the OSI model, TCP and UDP, the ports and protocols you see in logs, monitoring and the DMZ, Linux and SSH, Wireshark, and how to write up what you found. These are the questions interviewers ask on it.
OSI, TCP & UDP
Explain the OSI model. Which layers matter most to a SOC analyst?
OSI = Open Systems Interconnection, 7 layers (bottom to top): Physical, Data Link, Network, Transport, Session, Presentation, Application. A SOC focuses on Layer 3 (IP addresses — who and where is the actor), Layer 4 (TCP/UDP ports — open ports are entry points) and Layer 7 (application — HTTP, DNS, SMTP, where malicious content is injected).
TCP vs UDP — what is the difference? Give examples of each.
TCP is connection-oriented: a three-way handshake first, ordered and reliable delivery, lost data is re-sent — slower. Used for web pages, email, file transfer, anything sensitive. Like registered post with a delivery receipt.
Explain the TCP three-way handshake. How does it show up in an attack?
Client sends SYN → server replies SYN-ACK → client sends ACK. Only then does data flow. Port scan (reconnaissance): one source sends many SYNs to many ports and almost none complete the handshake — it is checking which ports are open.
What is the difference between a MAC address and an IP address?
MAC (Media Access Control): the hardware address burnt into a network card — 6 bytes (48 bits) in hexadecimal. Used by switches to deliver frames inside the local network (Layer 2). It rarely changes; the first half identifies the vendor.
Ports & protocols
What is a port? What is the difference between a port and a protocol? What is a socket?
A port is a logical number (not a physical socket) that tells the computer which application or service should receive the traffic. There are 65,535 ports. Port = where the traffic goes. Protocol = how the communication works (the rules).
Tell me the important port numbers and why the SOC watches each.
20/21 FTP (20 data, 21 commands) — unencrypted; large uploads to unknown servers = exfiltration. 22 SSH — admin-only remote access; thousands of attempts = brute force. 23 Telnet — plain text, should not exist. 25 SMTP — email, the main phishing entry. 53 DNS — tunnelling. 80 HTTP — clear text, web attacks. 443 HTTPS — encrypted, attackers hide C2 in it. 445 SMB — file sharing, EternalBlue/WannaCry. 3389 RDP — remote desktop, brute force and lateral movement. 8080/8443 — alternate HTTP/HTTPS.
Why is Telnet dangerous, and what replaced it?
Telnet (port 23) sends everything — including usernames and passwords — in plain text. Anyone capturing the traffic can read it. SSH (port 22) replaced it: an encrypted remote shell and secure file copy. Seeing Telnet in a modern network is itself a finding.
FTP vs SMB — what is the difference?
FTP (20/21) moves a file between client and server — you download it first, then work on it. Plain FTP is unencrypted, so credentials can be read. SMB (445) shares files, folders and printers between Windows machines — you can open and edit the file directly on the server. SMB exposed to the Internet is a classic attack path.
Explain EternalBlue in two minutes.
EternalBlue (CVE-2017-0144, Microsoft bulletin MS17-010) is a remote code execution exploit against Windows SMBv1 on port 445, using a buffer-overflow flaw. It was developed by a US intelligence group, leaked in 2017 by the hacker group Shadow Brokers, and then used by the WannaCry ransomware to spread worm-like across networks.
What is C2 traffic, and why do attackers use port 443 for it?
Command and Control (C2): an infected machine regularly "calls home" (beaconing) to the attacker’s server to receive commands or send data. Port 443 (HTTPS) is encrypted and used by almost everything, so malicious traffic blends into normal web traffic.
Network monitoring & DMZ
What is a DMZ, and why put two firewalls around it?
A DMZ (demilitarised zone) is a separate network between the Internet and the internal LAN for public-facing servers (website, mail). Outside users reach the website without touching the internal network. Firewall 1 filters Internet traffic into the DMZ (e.g. only HTTPS). Firewall 2 sits between the DMZ and the LAN — if the public server is hacked, the attacker still can’t reach internal data.
Write the firewall rules you would use for a web server in a DMZ.
Allow Internet → DMZ web server on 443/HTTPS. Block (or redirect to HTTPS) port 80, because HTTP is unencrypted. Allow SSH (22) only from approved admin IPs. Block DMZ → internal LAN on any port. Default: deny everything not listed.
When a user opens a website, which logs are created along the way?
Endpoint (EDR logs: process and connection) → DNS server (query log) → router/gateway → firewall, IDS/IPS and proxy (connection and web logs) → ISP → web server (request logs). The firewall is the SOC’s first sensor: all Internet traffic passes through it and is checked against rules. The Internet boundary is where internal meets external — firewalls, gateways, IDS/IPS and proxies sit there.
Read these three logs and tell me what happened.
Firewall: an external IP tried SSH on an internal server. DNS: the same server sent large TXT queries to random subdomains (DNS tunnelling). NetFlow: the server sent about 524 MB to the same external IP over 30 minutes (exfiltration).
What is a VPN? What is a site-to-site VPN?
A VPN (Virtual Private Network) builds an encrypted tunnel across the public Internet between an endpoint and the company network. Encryption turns readable information into data unauthorised people can’t understand. Site-to-site VPN: a permanent tunnel between two offices (e.g. Hyderabad ↔ Bangalore).
What is EDR, and what does it show that a firewall cannot?
EDR (Endpoint Detection & Response) runs on laptops, desktops and servers and records process execution, command lines, network connections and suspicious behaviour — and can isolate the machine. A firewall sees traffic between IPs; EDR sees which program on the machine did it and what that program did next.
Cloud, Wi-Fi & attack chain
Explain IaaS, PaaS and SaaS. Which cloud logs would you check?
IaaS: you rent VMs, storage and networking and manage the most (AWS EC2, Google Compute Engine). PaaS: a platform to build and deploy apps; the provider manages the platform (Azure App Service). SaaS: ready-to-use software; the provider manages everything (Microsoft 365, Google Workspace, Salesforce).
Which Wi-Fi standards and wireless attacks do you know?
Standards: WEP (outdated, insecure), WPA, WPA2 (widely deployed), WPA3 (newest, strongest). Attacks: fake public Wi-Fi (free hotspot that intercepts traffic); evil twin (same name as the company’s real network); rogue access point (unauthorised hardware plugged into the company network); deauthentication (forces devices off the real Wi-Fi so they reconnect to the attacker).
Walk me through a full attack, and where each step can be detected.
1 Recon (find emails/targets) → 2 Phishing email with attachment → 3 Malware runs when the user opens it → 4 C2 over 443, hidden in HTTPS → 5 Lateral movement to servers over SMB 445 → 6 Exfiltration to the attacker’s server.
What is Sysmon Event ID 3, and how do you use it?
Sysmon (a Microsoft Sysinternals tool) adds detailed Windows telemetry. Event ID 3 = a network connection made by a process: which process, source/destination IP and port. If an endpoint process connects to an unknown IP on 443: check the process origin (path, parent), the IP’s reputation and location, and whether it connected before.
How can an attacker learn a victim’s IP, and what are signs a device is compromised?
Tricks: getting the user to click a link, opening an Excel file with hidden PowerShell that runs ipconfig in the background, or granting permissions to a malicious app. Packet sniffing (capturing traffic, e.g. with Wireshark) on a shared network also reveals it.
Linux & SSH
Explain the Linux architecture: kernel and shell. Why does whoami matter in an investigation?
Linux is an open-source OS built around the Linux kernel (Ubuntu is one distribution). Layers: hardware → kernel (the "heart": manages CPU, memory, devices; programs need its permission to use the network) → shell (bash: where you type commands) → applications. /home holds users’ files.
Scenario: 300 failed SSH logins and then a successful root login. What do you do?
Investigate: search /var/log/auth.log for "Failed password" and "Accepted", note attacker IPs and times; check who is logged in now and active connections; review login history and what the attacker did first (commands, changed files, new users or SSH keys).
Wireshark
What is Wireshark, and how does a SOC use it?
Wireshark is a free, open-source network protocol analyzer: it captures the traffic on a network interface and lets you open every packet. Think of it as a CCTV camera for network traffic. Network engineers use it to check traffic flows as designed. In a SOC, L1 uses it to look at the traffic behind an alert; L3 uses it much more deeply in investigations.
How do you capture and save traffic in Wireshark?
Pick the interface that has traffic (Wi-Fi or Ethernet — the small live graph shows activity) and double-click it. Blue shark fin = start, red square = stop, green arrow = restart. File → Save As → a clear name (e.g. W1.pcapng). The .pcapng file is your evidence and goes to L2/L3. When restarting, Wireshark asks to save the previous capture first.
Which Wireshark display filters do you use most?
A busy network produces thousands of packets a minute; display filters show only what matters. Filtering for tcp still shows TLS/HTTP, because they run on top of TCP.
Open a packet: what do the Frame, Ethernet II, IP and TCP/UDP sections tell you?
Three panes: packet list (one line per packet), packet details (layers), packet bytes (hex/ASCII). Frame (Layer 1): frame number, bytes on wire vs captured (match = nothing lost), arrival and epoch time. Ethernet II (Layer 2): source/destination MAC and vendor, type (0x0800 IPv4, 0x86DD IPv6). IP (Layer 3): source/destination IP, TTL or hop limit, protocol (6 TCP, 17 UDP), checksum. TCP/UDP (Layer 4): ports, length, flags, checksum. Then the application data.
What are TTL and hop limit? What do "don’t fragment" and "checksum unverified" mean?
TTL (IPv4) / hop limit (IPv6): how many routers the packet may pass, 0–255. Each router subtracts 1; at 0 the packet is dropped. Flags "don’t fragment" with offset 0: the packet was not split into fragments. Checksum "[unverified]": Wireshark did not validate it — it does not mean the data was changed.
What is an ephemeral port? Source port 58941 → destination 3478 over UDP: what is happening?
An ephemeral port is a temporary port the client’s OS picks for one connection, usually 49152–65535; it changes every session. A service port (53, 443, 3478…) stays fixed and tells you the application. 3478 is STUN/TURN, used by live audio/video calls such as Google Meet. So 58941 → 3478 over UDP = a laptop in a live call.
In a TCP packet, what do sequence number, acknowledgment number and the PSH/ACK flags mean?
Sequence number: where this data belongs in the conversation. Next sequence number: where the next segment should start. Acknowledgment number: the next byte the sender expects from the other side. ACK acknowledges received data; PSH tells the receiver to pass the data straight to the application. Window size: how much data the receiver can take before acknowledging. Stream index: Wireshark’s ID for the conversation.
How are TCP, TLS and HTTPS related? Why can’t you read HTTPS data in Wireshark?
TCP carries the data, TLS encrypts it, and HTTPS (port 443) is web traffic protected by TLS. In the packet bytes the application data is unreadable. Without encryption (HTTP on 80, Telnet, FTP) anyone capturing the traffic can read it — that is exactly why TLS exists. You can still see IPs, ports, timing and the server name.
Investigation & reports
How do you write an investigation report as an L1 analyst?
Same sections every time: Heading, Investigation ID, Objective (one line), Evidence (packet number, timestamp, source/destination IP and port, protocol, length, duration), Analysis (what the fields show — direction, protocol, application), Conclusion (normal, suspicious, or needs more investigation), Justification (why).
Can you decide an alert is malicious from one packet?
Rarely. One packet shows one moment. You need the pattern (how many, how often), the duration, and context — logs from the SIEM, firewall and endpoint for the same time, and whether the activity was expected (a scheduled call, a change ticket).
Why are timestamps so important in an investigation?
Matching times is how you connect a packet to a SIEM alert, a firewall log and an endpoint event. Filter by timestamp to isolate the window where the suspicious activity happened. Wireshark shows arrival time (your time zone) and epoch time (seconds since 1 Jan 1970 UTC) — epoch makes matching logs from different systems easy. Always note the time zone.
You see a TCP SYN scan in Wireshark. What does an L1 analyst do?
Note the source IP, destination server, targeted ports, packet count and time window, starting from the first SYNs. Check whether any responses came back and any handshakes completed. Correlate with firewall, IDS/IPS and server logs for the same source and time, and look for follow-up: connections to the exposed services, credential attacks, persistence.
Recon, ATT&CK & OT
What is reconnaissance? Passive vs active?
Reconnaissance is the first stage of an attack: gathering information to plan it. Passive: no contact with the target — company website, LinkedIn, public records, social media habits. Very hard to detect. Active scanning: probing the target’s infrastructure — port scans, ping sweeps, vulnerability scans. Visible in logs and Wireshark (bursts of SYNs to many ports).
In MITRE ATT&CK, what is the difference between a tactic and a technique? Give examples.
Tactic = the goal, the "why" (e.g. Persistence). Technique = the method, the "how" (e.g. creating a new account). Each technique has detection and mitigation advice. Examples: Reconnaissance (gather info), Initial Access (phishing), Execution (run malicious code), Persistence (keep access after reboots or password changes), Lateral Movement (move to other systems), Exfiltration (steal data).
Define threat, vulnerability and risk.
Threat: something that can cause harm (an attacker, malware). Vulnerability: a weakness that can be exploited (unpatched software, a weak password). Risk: the chance that a threat exploits a vulnerability, and the impact if it does.
What is OT security, and how is it different from IT security?
OT (operational technology) is the hardware and software that monitors and controls physical processes — sensors, valves, production-line robots, power grids, water plants. IT protects data and systems; its top priority is often confidentiality. OT protects physical processes and people; the top priority is safety and uptime. OT attacks can cause physical damage or loss of life. Patching is hard because machines can’t simply be stopped. Alerts come from physical devices and sensors, and the SOC works with site engineers.
Interviewers will test Module 1 basics quickly. What will you revise the night before?
Port numbers (most common topic), OSI layers, TCP vs UDP and the handshake, MAC vs IP, private IP ranges, IDS vs IPS, DMZ, the six Event IDs, subnetting (2^n − 2), Wireshark filters, and your report template. Practise answers aloud in your own words, with a real example — interviewers judge understanding, not memorised slides. Don’t read; explain as if teaching.
Practise these in Interview IQ
These are the short answers. Enrolled students get all 40 questions in Interview IQ on the student portal, each with a 30-second answer to say aloud, a real-world example, the follow-up questions interviewers ask and the class recording to rewatch.
Prepare in the next SOC Analyst batch
Weekday and weekend batches run for students and working professionals, in the Madhapur classroom and live online.
- SOC Analyst · Nov 26, 2026Classroom & onlineIncludes demo session & webinar
Find Skill IT Education on
Search for Skill IT Education, Madhapur on any of these. A genuine listing always shows our address at LR Towers, 100 Feet Road and the phone number +91 91777 15978.
Preparing for these questions
Which ports should I memorise for a SOC interview?
20/21 FTP, 22 SSH, 23 Telnet, 25 SMTP, 53 DNS, 80 HTTP, 110 POP3, 143 IMAP, 443 HTTPS, 445 SMB and 3389 RDP, and why traffic on each could be suspicious.
Will I be asked about Wireshark?
Often. Expect to explain basic display filters (ip.addr, tcp.port, dns, http) and how you would spot a scan or a suspicious connection in a capture.
How do I explain an investigation in an interview?
Use a simple structure: what fired, the evidence you checked, your analysis, your conclusion and what you escalated or contained.
Get interview-ready for SOC roles
Share your details and our admissions team will call you about the next SOC Analyst batch, mock interviews and placement support.
