Linux Networking Basics

Linux networking is the base layer for everything you operate:

• web services (TCP/HTTP)
• databases (TCP)
• internal APIs and health checks
• Kubernetes ingress/service routing

This tutorial gives you a debugging toolbox and mental model.

Learning outcomes

By the end, you’ll be able to:

• inspect listening ports and active connections
• debug connectivity using DNS + TCP checks
• understand basic routing concepts (gateway/route)

1) Ports and sockets: what to check first

Listeners (server side)

# show listening sockets
sudo ss -lntp

# examples:
# -l : listening
# -n : numeric (no DNS lookup)
# -t : TCP
# -u : UDP

Connections (client side)

# show established connections
sudo ss -ant | findstr ESTAB

# show all connections for a PID
sudo ss -pant | grep 12345

Rule of thumb:

• if a port isn’t listening → service isn’t running or can’t bind
• if it’s listening but clients can’t connect → firewall/network/routing/DNS issue

2) DNS: confirm names resolve correctly

Use dig (preferred) or nslookup.

# basic lookup
dig example.com

# query a specific record type
dig example.com A

dig kubernetes.default.svc.cluster.local

3) Test connectivity with curl

If you expect HTTP:

# get headers only
curl -I http://localhost:8080

# show verbose debug
curl -v http://localhost:8080/health

# follow redirects (if needed)
curl -v -L https://example.com

4) Test TCP reachability (when HTTP might not be enough)

Linux doesn’t always ship nc by default, but if you have it:

# check if TCP port is open
nc -vz 192.168.1.10 5432

5) Routing: how packets leave the machine

Inspect routes:

# routing table
ip route

# route lookup to a specific destination
ip route get 8.8.8.8

6) Trace the path (advanced but useful)

# trace route (ICMP-based depending on network)
traceroute example.com

# alternative with TCP (often blocked, but helpful sometimes)
# requires traceroute with TCP support or use of other tooling

Common troubleshooting flows

Flow A: “Service is down”

1. ss -lntp → is it listening?
2. systemctl status <service>
3. check logs (journalctl)

Flow B: “Service is up but clients can’t reach it”

1. verify DNS (dig)
2. test TCP/HTTP (curl -v)
3. inspect routes (ip route)
4. check firewall rules (local iptables/ufw or cloud security groups)

Next steps

Next tutorials in the DevOps track:

• shell scripting patterns (automation)
• CI/CD pipelines (GitHub Actions)
• IaC (Terraform)
• configuration management (Ansible)