Learn Rocky Linux - Cockpit & Remote Management

#Introduction

In the previous episode 10, you understood the deepest parts of the system — the kernel and the boot process. Now it's time to look in a different direction: how to view and manage the server from outside without having to open a terminal. The answer is Cockpit, a web console integrated directly with Rocky Linux.

Cockpit is not a CLI replacement — you'll still live in the terminal for heavy lifting. But for quick observations, service management, viewing logs, and monitoring several servers at once, Cockpit is a very convenient assistant. This episode will take you from installation to integration with Podman and KVM.

#Installation and Access

#Installing Cockpit

Cockpit is an official package from the Rocky repository. Install it along with the modules you need:

dnf5 install cockpit

systemctl enable --now cockpit.socket

Cockpit is designed as a socket-activated service — it only listens when there's a connection, saving resources.

#Accessing the Web Console

Cockpit listens on port 9090 with TLS encryption. From a browser:

https://server-ip:9090

You log in with the same system user accounts — including users with sudo access from episode 6. Verify the service and port:

ss -tlnp | grep 9090

systemctl status cockpit.socket

If access from outside fails, remember that firewalld is active by default — you'll open port 9090 in episode 13, or use firewall-cmd --add-service=cockpit now.

#Dashboard and Management

#System Overview

After logging in, the Overview page shows a summary: CPU and memory usage, system load, disk, and hardware information. This is a quick view to answer the question "is the server healthy?".

uptime
free -h
df -h

Cockpit shows the same data as top, free, and df — but in graphs that move over time.

#User and Service Management

Cockpit provides an interface for everyday administrative tasks:

• Accounts — create and manage users and assign admin roles (the wheel group), mirroring the commands from episode 6.
• Services — view service status, start, stop, restart, and enable auto-start, all powered by systemctl underneath.
• Terminal — full shell access directly in the browser for commands that aren't available as a graphical interface.

systemctl list-units --type=service --state=running

Cockpit doesn't hide the CLI — it actually blends it in: every action you take in the web UI is translated into the same commands you could type manually.

#Updates and Log Viewer

The Software Updates section shows packages with available updates, and the Logs section provides access to the journald from episode 7:

dnf5 check-update

journalctl -p warning --since "today"

Both sections let you monitor system health without entering the terminal — convenient for servers maintained by teams that aren't fully CLI-native.

#Integration and Modules

#Cockpit-Podman

Containers with Podman (episode 20) can be managed directly from Cockpit with the cockpit-podman module:

dnf5 install cockpit-podman podman

rpm -q cockpit-podman

Once installed, the Containers section in Cockpit shows running containers, lets you start and stop them, view logs, and create new containers — container observability without a terminal.

#Cockpit-Machines for KVM

For KVM/libvirt virtualization (episode 19), the cockpit-machines module gives full control over VMs:

dnf5 install cockpit-machines

systemctl status libvirtd

The Virtual Machines section lets you create, power on, power off, and monitor VMs from the browser — a perfect pair with the virsh you'll learn later.

#Cockpit-WS for Remote Access via SSH

The core cockpit-ws module handles authentication and connections. Its powerful feature: if the destination server doesn't have its own Cockpit, you can manage multiple servers at once from a single dashboard using Dashboard and SSH connections — target servers remain accessible even if only one host runs Cockpit.

rpm -qa | grep cockpit

This is a pattern many IT teams use: one jump server with Cockpit, managing an entire server farm from a single browser page.

#Usage Practices

To monitor several servers, configure everything from one center:

1. Install Cockpit on one main server (jump host).
2. Open the Dashboard menu and add other servers with SSH credentials.
3. Access everything from one web console, complete with graphs and centralized logs.

Make sure the firewall service is open for port 9090, and use the hostnames configured in episode 9 so the Dashboard doesn't get confused. With this pattern, multi-server observability becomes one click, without leaving the browser.

#Closing

In this episode 11, you got to know Cockpit as Rocky Linux's web console: installation and access via port 9090, a dashboard for monitoring CPU, memory, disk, and network, user and service management, log and update viewers, and integration with Podman, KVM, and cockpit-ws for remote management via SSH.

Key takeaways:

• Install with dnf5 install cockpit and enable the socket with systemctl enable --now cockpit.socket.
• Access it via https://server-ip:9090 with system user accounts.
• The Cockpit dashboard mirrors CLI commands like uptime, free, df, and systemctl.
• The cockpit-podman and cockpit-machines modules extend its reach to containers and VMs.
• cockpit-ws enables managing many servers from one dashboard via SSH.

In the next episode 12, we will discuss logging, chrony, and time synchronization — comparing journald and syslog, the /var/log directory structure, log rotation with logrotate, time synchronization with chrony, and debugging with timedatectl. The server is now visible from outside; next up, understanding its traces and time!