Installing and Configuring Grafana

The SSH (Secure Shell) protocol is a network protocol for remote command-line management of operating systems, widely considered the standard for remote access to *nix machines. It allows secure login to a server, remote command execution, file management (creating, deleting, copying, etc.), and more. Most cloud and hosting providers require SSH to access their services. In this article, we’ll look at how to copy files over SSH on both Windows and Linux systems. How SSH Works SSH can securely transmit any data (audio, video, application protocol data) through an encrypted communication channel. Unlike outdated and insecure protocols like Telnet and rlogin, SSH ensures data confidentiality and authenticity — essential for internet communications. Here’s how a secure connection between a client and server is established: TCP Connection Setup: By default, the server listens on port 22. Both sides share a list of supported algorithms (compression, encryption, key exchange) and agree on which to use. Authentication: To prevent impersonation, both parties verify each other's identities using asymmetric encryption (public/private key pairs). First, the server is authenticated. On the first connection, the client sees a warning with server details. Trusted server keys are stored in /home/<username>/.ssh/known_hosts. Key Generation: Once the server is verified, both sides generate a symmetric key to encrypt all data exchanged. User Authentication: This is done using either a password or a client-sent public key stored in /home/<username>/.ssh/authorized_keys on the server. The most popular implementation on Linux is OpenSSH, which comes pre-installed on most distributions (Ubuntu, Debian, RHEL-based, etc.). Clients like PuTTY or MobaXterm are used on Windows. Since Windows 10 and Server 2019, OpenSSH tools are also available natively. You can learn more about working with SSH in our tutorial. File Copying via SSH Two main utilities for copying files over SSH in Linux are scp and sftp. Both come with OpenSSH. SSH supports two protocol versions: 1 and 2. OpenSSH supports both, but version 1 is rarely used. Autocompletion Setup To enable Tab-based autocompletion when using scp, set up public key authentication: Generate a key pair: ssh-keygen You’ll see output like: Generating public/private rsa key pair. Enter file in which to save the key (/home/user/.ssh/id_rsa): Enter passphrase (empty for no passphrase): By default, your keys (id_rsa for private and id_rsa.pub for public) are saved to ~/.ssh/. Now copy the public key to the remote machine: ssh-copy-id [username]@[ip-address] After entering the user's password, you’ll see a message confirming the key was added. Secure Copy (SCP) For small data transfers (e.g., service configs), scp is best. Copy from local to remote: scp test.txt user@192.168.1.29:/home/user/ Copy multiple files: scp test1.txt test2.txt user@192.168.1.29:/home/user/ Copy from remote to local: scp user@192.168.1.29:/home/user/test.txt ~/ Copy directories: scp -r testdir user@192.168.1.29:/home/user/ Remote-to-remote copy: scp gendo@192.168.1.25:/home/gendo/test.txt user@192.168.1.29:/home/user/ Secure FTP (SFTP) SFTP is another utility included in OpenSSH. As of OpenSSH 9.0, scp now uses SFTP by default instead of the old SCP/RCP protocol. Unlike classic FTP, sftp transmits encrypted data over a secure tunnel. It does not require a separate FTP server. Example usage: sftp misato@192.168.1.29 sftp> ls sftp> lcd testdir/ sftp> get test.txt sftp> bye Graphical file managers like Midnight Commander and Nautilus use sftp. In Nautilus, the remote server appears like a local folder, e.g., user@ip. Copying Files Over SSH on Windows Use the pscp command-line tool from PuTTY to copy files on Windows. Copy to server: pscp C:\server\test.txt misato@192.168.1.29:/home/misato/ Copy from server: pscp misato@192.168.1.29:/home/misato/test.txt C:\file.txt List files on remote server: pscp -ls user@192.168.1.29:/home/misato Use quotes for paths with spaces: pscp "C:\dir\bad file name.txt" misato@192.168.1.29:/home/misato/ To get help, run: pscp Conclusion We’ve covered how to copy files to and from a server using the secure SSH protocol. If you work with cloud servers, understanding SSH is essential — it’s the standard method for remote access to *nix machines and a vital part of everyday DevOps and system administration.

The term daemon comes from a word in ancient Greek mythology that referred to an immaterial being influencing the human world. In computing, especially in UNIX-like operating systems, a daemon is a background process that runs without direct interaction from the user. It doesn’t depend on a terminal or user interface and typically starts with the system boot or under specific conditions. What is a Daemon The main function of a daemon is to provide specific services to other processes or users. For example, a daemon might listen on network ports waiting for connections, monitor system events and respond when certain conditions are met, manage scheduled jobs (like cron), send emails (sendmail), and more. In Windows, the closest equivalent to a daemon is a service. The difference lies mainly in how they're started, registered, managed, and configured within operating systems. However, their purpose is the same: to ensure continuous background operation of certain functions or services. Key Characteristics of a Daemon Runs in the background: Users typically don’t see the daemon’s interface; it doesn’t write to standard output (or redirect it to logs), nor does it request keyboard input. Autonomous: A daemon starts either at system boot when triggered by an init system (like systemd), or manually by a user (via scripts, cron, etc.). Long-lived: Ideally, a daemon runs indefinitely unless a critical error occurs or it receives an explicit stop signal. Isolated: Usually runs under a separate user/group account to minimize privileges, making services more secure and easier to manage. Logging: Instead of using standard input/output, daemons log information to log files or the system logger (journald, syslog, etc.), which is helpful for debugging and diagnostics. Daemons in Linux Historically, nearly all system background tasks in Linux are implemented as daemons. The OS includes dozens of them, each responsible for a specific function. Here are some examples: sshd (Secure Shell Daemon): Listens on port 22 (by default) and allows remote users to connect via encrypted SSH. Without sshd, remote terminal access is almost impossible. cron: A job scheduler daemon. It checks crontab entries and runs scripts or commands on a schedule, such as log cleanup, sending reports, system checks, etc. syslogd / rsyslog / journald: System logging daemons that collect messages from the kernel, utilities, other daemons, and apps, and save them in log files or the journal. NetworkManager or Wicd: Daemons that manage network settings — automating connections to wired/wireless networks, switching, configuring VPNs, and more. These daemons start at system boot and are registered with the system service manager (e.g., systemd). They run until the system is shut down or rebooted. Users interact with them indirectly — through config files, terminal commands (service, systemctl), or network requests (if the daemon provides HTTP/S, SSH, or another network interface). How to Create and Manage Daemons To implement a daemon, follow these steps: Forking the process: The parent process calls fork() and continues running the daemon code in the child process. Detach from controlling terminal (setsid): To avoid user interference (e.g., closing the terminal), the daemon calls setsid() to start a new session and become its leader. Close standard input/output descriptors: Since the daemon shouldn't write to the screen or wait for input, stdin, stdout, and stderr are closed or redirected to log files. Handle signals and logging: To support graceful shutdown or config reloads, the daemon must handle signals (SIGTERM, SIGHUP, etc.). Logging is usually done via syslog or files. Main loop:  After initialization, the daemon enters its main loop: waiting for events, handling them, and repeating until stopped. Let’s see how to create a daemon on Ubuntu 22.04 using a Hostman cloud server 1. Write the Daemon in C Create a file called mydaemon.c and insert the following code: #include <stdio.h> #include <stdlib.h> #include <unistd.h> #include <syslog.h> int main() { // Open syslog openlog("mydaemon", LOG_PID, LOG_DAEMON); syslog(LOG_NOTICE, "Daemon started"); // Main infinite loop while (1) { // Your background tasks: monitoring, queue handling, etc. syslog(LOG_NOTICE, "Performing task..."); sleep(60); } // If we ever exit the loop syslog(LOG_NOTICE, "Daemon stopped"); closelog(); return 0; } 2. Compile the Program First, update your packages: sudo apt update && sudo apt upgrade Install the GCC compiler if not already installed: sudo apt install gcc Compile the daemon: gcc mydaemon.c -o mydaemon 3. Move the Executable Move the binary to /usr/local/bin/, a standard location for custom utilities: mv mydaemon /usr/local/bin/mydaemon 4. Create a systemd Service Create a unit file called mydaemon.service: sudo nano /etc/systemd/system/mydaemon.service Insert the following content: [Unit] Description=My Daemon After=network.target [Service] Type=simple ExecStart=/usr/local/bin/mydaemon Restart=on-failure [Install] WantedBy=multi-user.target Explanation of the fields: Description: Description shown in systemctl status. After=network.target: Ensures the daemon starts after the network is up. Type=simple: The daemon doesn’t fork, it runs as a single process. ExecStart: Path to the daemon executable. Restart=on-failure: Restart automatically if the daemon crashes. WantedBy=multi-user.target: Makes the service start in the standard multi-user environment. 5. Start and Monitor the Daemon sudo systemctl daemon-reload # Reload systemd configuration sudo systemctl start mydaemon # Start the daemon sudo systemctl status mydaemon # Check its status If everything works, the status will show active. To view logs: journalctl -u mydaemon.service -e Examples of Daemon Usage Web Servers Their job is to listen on a network port (usually 80 or 443), accept HTTP/HTTPS requests, generate a response (an HTML page, JSON data, etc.), and send the result back to the client. In most cases, a web server starts with the system and continues running until the server is shut down or a stop command is issued (e.g., systemctl stop nginx). Database Daemons MySQL/MariaDB, PostgreSQL, MongoDB — all of these are also daemons. They start with the system and continue running in the background, accepting requests from client applications or web services. These daemons typically log activity, support configuration via files, and are managed using special utilities (or systemd). Job Schedulers (cron, atd) The cron daemon checks the schedule table (crontab) and runs programs at the times or intervals specified by the user. This makes it possible to automate backups, system updates, health checks, and many other routine tasks. atd is a similar daemon but executes tasks only once at a specific time (unlike cron, which runs tasks on a regular schedule). Access and Control Services (sshd, xrdp) sshd (Secure Shell Daemon) provides remote access via the SSH protocol. xrdp enables remote desktop connections using the RDP protocol. It acts as a daemon that listens for network connections on a specified port. Init System Daemons (systemd, init, Upstart) In modern systems, the role of the "main daemon" is fulfilled by systemd (replacing the older SysV init system). systemd is the first process to start after the kernel and is responsible for launching and managing all other services and processes. It starts them in parallel and handles their dependencies. Simply put, systemd is itself a daemon that “orchestrates” all others in the system. Advantages and Disadvantages of Daemons Advantages: Automation: Daemons enable system behavior to be automated — from responding to network requests to scheduling tasks — without user intervention. Isolation: Running under separate user/group accounts and detaching from terminals enhances security by limiting potential damage in case of compromise. Continuous Operation: A daemon can keep servicing requests (like a web server) without interruption even if the user logs out or the console is closed. Manageability: Linux provides system tools (e.g., systemd, init scripts) to centrally manage all daemons: starting, stopping, restarting, and logging. Disadvantages: Debugging Complexity: Since daemons run in the background and don’t output to the console, debugging requires thorough logging and more complex setups (debug flags, tracing, etc.). Security Risks: If a daemon runs with elevated privileges (e.g., as root), any vulnerability can potentially compromise the entire system. It's best to run daemons under limited accounts. Dependency Management: Some daemons may fail if, for example, they need network access before the network is up. Modern init systems solve this, but with classic SysV init scripts, this used to be a common issue. Increased Resource Usage: Any constantly running background process consumes system resources (memory, CPU time). If there are too many daemons, this can impact performance, especially on systems with limited resources. Conclusion Daemons are central to Linux operating systems' architecture, offering vast automation and background services capabilities. They allow administrators to flexibly configure network operations, scheduled tasks, logging, security systems, and many other components. Writing your own daemon requires understanding processes, signals, system calls, and careful attention to logging and security. Modern init systems (especially systemd) have simplified daemon management and service logic, making the creation of custom services more structured and flexible. However, it remains a complex field that demands careful design, debugging, and ongoing maintenance.

A personal DNS server can be useful if your provider doesn't offer this service or if existing solutions don't suit your needs. The easiest way to set one up is via a control panel (cPanel, CloudPanel, HestiaCP, etc), but you can also do it manually using the terminal and the Linux DNS Server BIND 9. Preparing the Server Let's say you've rented a Hostman Linux VPS and want to use your own DNS servers. To do that, you need to meet two conditions: Order another public IP address — DNS setup requires at least two IPs. Open DNS port 53, which is necessary for the nameserver to work. Ubuntu/Debian Update the package list: apt update Allow incoming packets on port 53 UDP in the firewall: iptables -I INPUT -p udp --dport 53 -j ACCEPT Save the firewall rules: iptables-save CentOS Install system updates: yum update Install time synchronization utility: yum install chrony Set your timezone, for example: timedatectl set-timezone Europe/Cyprus Enable and start the time synchronization service: systemctl enable chronyd --now Open port 53: firewall-cmd --permanent --add-port=53/udp Apply the updated firewall rules: firewall-cmd --reload Installing the DNS Server This guide uses BIND 9 to create an IP-based DNS server. Ubuntu/Debian Install required packages: apt-get install bind9 dnsutils Enable autostart: systemctl enable bind9 Start the service: systemctl start bind9 Check if it's running: systemctl status bind9 Look for active status in the output. CentOS Install the DNS utility: yum install bind Enable autostart: systemctl enable named Start the service: systemctl start named Check its status: systemctl status named You should see active in the output. Basic DNS Server Configuration The settings are defined in the configuration file. Ubuntu/Debian Open the config file: vi /etc/bind/named.conf.options In the listen-on block, specify the networks, e.g.: listen-on { 10.10.10.0/24; 10.1.0.0/16; }; To allow the DNS server to listen on all interfaces, either omit this line or use any. In the allow-query line, specify who can make queries: allow-query { any; }; Restart the service for changes to take effect: systemctl restart bind9 CentOS Open the config file: vi /etc/named.conf Find these lines: listen-on port 53 { 127.0.0.1; localhost; 192.172.160.14; }; ... allow-query { any; }; In the listen-on line, after localhost, specify the DNS IP address. This is the IP on which the host will accept queries. Use any to listen on all addresses. In the allow-query line, define query permissions. any allows queries from everyone. You can also restrict it to a specific subnet, e.g., 192.172.160.0/24. Apply the config: systemctl restart named Global Options Besides the basics, you can fine-tune the server using other global parameters: Argument What It Configures directory Working directory (default is /var/named if not specified) forwarders IPs to forward unresolved queries to (e.g., Google's DNS) forwarders { 8.8.8.8; 8.8.4.4; }; forward Options: FIRST or ONLY. FIRST tries forwarders first, then internal. ONLY skips internal search. listen-on Interfaces that BIND listens on (usually port 53 UDP) allow-transfer Hosts allowed for zone transfers allow-query Who is allowed to send DNS queries allow-notify Hosts allowed to receive zone change notifications allow-recursion Hosts that can make recursive queries. Default is unrestricted. Testing To check if the DNS server accepts queries from clients, use the nslookup utility. From another computer: nslookup site-example.com 192.172.160.14 This checks the IP address of site-example.com using DNS server 192.172.160.14. Alternatively, use dig: dig @192.172.160.14 site-example.com It works similarly, just a different syntax. BIND Zones Basic DNS server setup is complete. Now, let’s talk about usage. For that, you configure zones: Primary zone – You create and edit domain records directly on this host. Secondary zone – This host pulls data from a primary DNS server. Stub zone – Stores only NS records used for redirection. Caching-only zone – Doesn’t store records; only caches query results for performance. Zone management is handled in the config file and is a larger topic. Creating your own zone lets you assign friendly names to each host, which is helpful when dealing with many nodes instead of using IPs.