Post

Linux Firewalls: nftables, iptables, firewalld, and ufw

Linux Firewalls: nftables, iptables, firewalld, and ufw

Linux has four firewall front-ends in common use, and they are not interchangeable. iptables is legacy, nftables is its replacement, firewalld is a dynamic manager on top of nftables, and ufw is a simplified front-end for Ubuntu. This post maps them and shows the commands you actually need.

Which One Am I Running?

1
2
3
4
5
6
7
8
9
10
11
12
13
14
# Is firewalld active?
systemctl status firewalld

# Is ufw active?
sudo ufw status

# Is nftables active?
systemctl status nftables

# Are there iptables rules?
sudo iptables -L -n -v

# Are there nftables rules?
sudo nft list ruleset

Tip: On modern systems, iptables commands are usually translated to nftables by iptables-nft. The rules you see with iptables -L may actually live in the nftables ruleset.

nftables: The Modern Default

nftables uses tables, chains, and rules, with a cleaner syntax than iptables.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
# Show the full ruleset
sudo nft list ruleset

# Create a table and chain
sudo nft add table inet filter
sudo nft add chain inet filter input '{ type filter hook input priority 0; policy drop; }'

# Allow established connections
sudo nft add rule inet filter input ct state established,related accept

# Allow SSH
sudo nft add rule inet filter input tcp dport 22 accept

# Allow loopback
sudo nft add rule inet filter input iif lo accept

# Save the ruleset
sudo nft list ruleset > /etc/nftables.conf

Tip: nft rules are applied immediately. If you lock yourself out over SSH, you need console access to recover, so always allow SSH before setting the default policy to drop.

iptables: Legacy but Still Everywhere

1
2
3
4
5
6
7
8
9
10
11
12
13
14
# List rules with counters
sudo iptables -L -n -v

# Allow established connections
sudo iptables -A INPUT -m conntrack --ctstate ESTABLISHED,RELATED -j ACCEPT

# Allow SSH
sudo iptables -A INPUT -p tcp --dport 22 -j ACCEPT

# Drop everything else
sudo iptables -P INPUT DROP

# Save rules (Debian)
sudo iptables-save > /etc/iptables/rules.v4

Tip: iptables -L -n -v shows packet counters. A rule with zero matches is not the rule blocking you, and chasing it wastes time.

firewalld: Dynamic Zones

firewalld organises rules into zones, each with its own trust level. It is the default on RHEL family systems.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
# Show the active zone and its rules
sudo firewall-cmd --list-all

# List all zones
sudo firewall-cmd --get-zones

# Allow a service permanently
sudo firewall-cmd --permanent --add-service=http
sudo firewall-cmd --reload

# Allow a port
sudo firewall-cmd --permanent --add-port=8080/tcp
sudo firewall-cmd --reload

# Allow a source subnet
sudo firewall-cmd --permanent --add-source=10.0.0.0/24 --zone=trusted
sudo firewall-cmd --reload

Tip: --permanent changes are not active until you run --reload. Forgetting the reload is the most common firewalld mistake.

ufw: Simple Front-End

ufw is designed to be easy, and it is the default on Ubuntu.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
# Enable and check status
sudo ufw enable
sudo ufw status verbose

# Allow SSH
sudo ufw allow 22/tcp

# Allow a service by name
sudo ufw allow 'Nginx Full'

# Allow from a specific source
sudo ufw allow from 10.0.0.0/24 to any port 5432

# Delete a rule
sudo ufw delete allow 22/tcp

Tip: ufw allow OpenSSH before ufw enable is the standard way to avoid locking yourself out on a fresh Ubuntu server.

Debugging Firewall Drops

1
2
3
4
5
6
7
8
9
# nftables: log dropped packets
sudo nft add rule inet filter input log prefix "nft-drop: " drop

# iptables: log dropped packets
sudo iptables -A INPUT -j LOG --log-prefix "iptables-drop: "

# Watch the logs
sudo journalctl -k -f | grep -i drop
sudo dmesg -w | grep -i drop

Tip: Logging every drop is noisy. Add a temporary log rule, reproduce the problem, then remove it. Permanent drop logging will flood your disk.

nftables Sets, Maps, and Flowtables

nftables has features that iptables lacks: sets, maps, and flowtables.

1
2
3
4
5
6
7
8
9
10
11
12
# Create a set of allowed IPs
sudo nft add set inet filter allowed_ips '{ type ipv4_addr; }'
sudo nft add element inet filter allowed_ips '{ 10.0.0.1, 10.0.0.2 }'
sudo nft add rule inet filter input ip saddr @allowed_ips accept

# Create a map: port to backend
sudo nft add map inet filter port_map '{ type inet_service : ipv4_addr; }'
sudo nft add element inet filter port_map '{ 80 : 10.0.0.10, 443 : 10.0.0.11 }'

# Create a flowtable for fast-path forwarding
sudo nft add flowtable inet filter f '{ hook ingress priority 0; devices = { eth0, eth1 }; }'
sudo nft add rule inet filter forward ip protocol { tcp, udp } flow add @f

Tip: Flowtables are the single biggest performance win in nftables. They bypass the normal rule evaluation for established flows, which is how nftables handles high-throughput forwarding.

Migrating from iptables to nftables

1
2
3
4
5
6
7
8
# Translate an iptables ruleset to nftables
iptables-save | iptables-restore-translate -f /etc/nftables.conf

# Translate a single rule
iptables-translate -A INPUT -p tcp --dport 22 -j ACCEPT

# Translate an ip6tables rule
ip6tables-translate -A INPUT -p tcp --dport 22 -j ACCEPT

Tip: iptables-translate is the fastest way to convert a single rule. It produces the exact nftables equivalent, which is far less error-prone than translating by hand.

firewalld Rich Rules

Rich rules are firewalld’s way of expressing complex conditions that the simple --add-service and --add-port cannot.

1
2
3
4
5
6
7
8
9
10
11
# Allow a specific source to a specific service
sudo firewall-cmd --permanent --add-rich-rule='rule family="ipv4" source address="10.0.0.0/24" service name="ssh" accept'

# Rate-limit SSH
sudo firewall-cmd --permanent --add-rich-rule='rule service name="ssh" limit value="10/m" accept'

# Log and drop
sudo firewall-cmd --permanent --add-rich-rule='rule family="ipv4" source address="10.0.0.0/24" log prefix="blocked: " level="info" drop'

# Reload
sudo firewall-cmd --reload

Tip: Rich rules are evaluated before the zone’s normal rules, which makes them the right tool for exceptions. They are also the only way to rate-limit in firewalld.

Quick Reference

Tasknftablesiptablesfirewalldufw
Show rulesnft list rulesetiptables -L -n -vfirewall-cmd --list-allufw status verbose
Allow SSHnft add rule ... tcp dport 22 acceptiptables -A INPUT -p tcp --dport 22 -j ACCEPTfirewall-cmd --add-service=sshufw allow 22/tcp
Reloadnft -f /etc/nftables.confiptables-restorefirewall-cmd --reloadufw reload

Tip: Work in order: identify the active firewall, list its rules, add the allow rule, then verify with a connection test. Most “firewall is blocking me” incidents are a rule that was added but never reloaded.

This post is licensed under CC BY 4.0 by the author.