Managing Static Routes in Linux

1. The Linux Routing Table: Your Server's GPS

Every time your Linux system needs to send a network packet, it consults an internal set of rules called the routing table. Think of this table as a GPS for your network traffic. For each packet, the kernel looks at the destination IP address and checks the routing table to answer the question: "What's the best way to get there?"

To understand this better, let's start with a foundational reading.

As you read, focus on these key concepts:

• Route: A rule that maps a destination network to an outgoing network interface and, optionally, a gateway.
• Gateway (or Router): A device on your local network that knows how to forward traffic to other networks.
• Default Gateway: The "catch-all" gateway. If the kernel can't find a specific route for a destination, it sends the packet to the default gateway. This is how your server talks to the internet.
• Routing Table: The collection of all routes that the kernel maintains.

2. Viewing the Current Routing Table

Before you can manage routes, you need to see what's already configured. The modern command for this is part of the iproute2 suite, which you've used before.

To view the main routing table, use this command:

ip route show

You can also shorten this to just ip route.

Let's break down a typical output:

default via 192.168.1.1 dev enp0s3
192.168.1.0/24 dev enp0s3 proto kernel scope link src 192.168.1.100

• default via 192.168.1.1 dev enp0s3: This is the default route. It means "for any traffic that doesn't match a more specific rule, send it to the gateway at 192.168.1.1 using the enp0s3 network interface." This is your server's path to the internet.
• 192.168.1.0/24 dev enp0s3 ...: This is a route for the local network. It means "for any traffic destined for the 192.168.1.0/24 network, send it directly out of the enp0s3 interface." No gateway is needed because the destination is on the same local network.

You might also encounter the older route -n command, especially in legacy scripts. It provides similar information but in a different format. While ip route is the modern standard, it's good to recognize both.

To learn more about viewing the routing table, let's consult our resources.

3. Managing Static Routes

A static route is a route that you, the administrator, add manually. This is necessary when your server needs to communicate with a network that is not on its local subnet and cannot be reached via the default gateway.

A common scenario in a business environment is accessing a separate private network, for instance, a network hosting internal databases or development tools.

Adding and Deleting Routes (Temporarily)

The ip route command is used for both adding and deleting routes. The syntax for adding a route is:

sudo ip route add {NETWORK/CIDR} via {GATEWAY_IP}

Let's imagine your server (192.168.1.100) needs to connect to a database on a private network 10.20.30.0/24. Your local network's router doesn't know about this network, but there is a special gateway at 192.168.1.254 that does.

You would add a static route like this:

sudo ip route add 10.20.30.0/24 via 192.168.1.254

Now, when your server tries to send a packet to, say, 10.20.30.55, the kernel will match this new rule and forward the packet to the 192.168.1.254 gateway.

To remove this route, you use ip route del:

sudo ip route del 10.20.30.0/24

Important: Any routes you add this way are temporary. They will be lost when the system reboots.

The following resources provide detailed syntax and examples for these commands.

4. Making Routes Persistent

In any real-world DevOps role, you can't have your network configuration disappear after a reboot. You need to make your static routes persistent. The method for this depends on the Linux distribution and its network management tools.

On Debian/Ubuntu Systems

For older systems or those configured with /etc/network/interfaces, you can add up or post-up commands to an interface's configuration stanza. This command will be executed after the interface is brought up.

For example, in /etc/network/interfaces:

auto enp0s3
iface enp0s3 inet static
    address 192.168.1.100
    netmask 255.255.255.0
    gateway 192.168.1.1
    # Add a persistent static route
    post-up ip route add 10.20.30.0/24 via 192.168.1.254

On RHEL/CentOS Systems

On Red Hat-based systems, you typically create a dedicated route file for the interface. For an interface named eth0, you would create /etc/sysconfig/network-scripts/route-eth0.

The content of the file would be:
10.20.30.0/24 via 192.168.1.254

The following reading covers these persistence methods clearly. Note that modern systems often use netplan (on Ubuntu) or NetworkManager (nmcli) (on many distributions), which have their own methods. However, understanding these file-based approaches is fundamental.

5. Verifying and Troubleshooting Routes

After adding a route, how do you know if it's working or being used correctly?

The ip route get command is an excellent tool for this. It tells you exactly which route the kernel will use to reach a specific IP address without actually sending any traffic.

To test our earlier example:

ip route get 10.20.30.55

The output will show you the gateway and interface that will be used, confirming your static route is being selected.

10.20.30.55 via 192.168.1.254 dev eth0 src 192.168.1.100 uid 1000
    cache

This is much faster than running a traceroute and confirms the decision part of the routing process.

Conclusion

You've now learned how to inspect and control your Linux server's outbound traffic flow using static routes. This is a fundamental networking skill that moves you beyond managing a single host to configuring how that host interacts within a larger, more complex network topology.

Key Takeaways:

• The routing table is a set of rules used by the Linux kernel to determine where to send network packets.
• The ip route show command is used to view the current routing table.
• A default route handles all traffic not covered by a more specific route.
• Static routes are added manually with ip route add to provide paths to specific networks.
• Routes added via the command line are temporary and are lost on reboot.
• To make routes persistent, you must add them to system-specific network configuration files (e.g., /etc/network/interfaces or /etc/sysconfig/network-scripts/).
• The ip route get {IP} command is a powerful tool for verifying which route will be used for a given destination.

In our next and final lesson for this module, we will put all our recent networking knowledge into practice. You will set up a simple web server (Nginx) and verify network connectivity to it, applying your understanding of IP configuration, firewalls, and routing to expose a service to the network.