Linux Networking and Distributed Services Study Guide

Linux Networking Preliminaries and Virtualization Setup

Networking is a critical capability for system administrators and developers operating within Linux environments. Establishing network functionality allows system setup for internet connectivity, identification of system interface properties and IP configurations, remote file transfer, and remote administrative access. When configuring network connectivity inside a virtual machine environment such as Oracle VirtualBox Manager, internet access issues on a non-responsive VM can be resolved through specific configuration steps. The administrator opens VirtualBox Manager, selects the target virtual machine from the left pane, navigates to the Settings menu located at the top toolbar, selects Network from the left pane of the settings window, and modifies the attached network type by selecting Bridged Adapter from the drop-down menu. Saving these settings by clicking OK and starting the virtual machine bridges the virtual network adapter directly to the physical network interface of the host.

Core networking parameters govern how systems identify themselves and route traffic. An IP address is a unique string of numbers separated by periods that identifies each computer utilizing the Internet Protocol to communicate across a network. A subnet mask is a $32\text{-bit}$ number that masks an IP address, dividing the total address space into a network address component and a host address component. Subnet masks are constructed by setting all network bits to binary $1$ and all host bits to binary $0$. Additional fundamental parameters include the default gateway, static versus dynamic address assignment models, network interface cards (NIC), and hardware Media Access Control (MAC) addresses.

Linux Network Configuration Files and System Commands

System configuration in Linux relies on specific files located within the root file system hierarchy. Resolution of hostnames to IP addresses across various databases is governed by /etc/nsswitch.conf. The system IP address association and system hostname are defined in /etc/hostname. In Red Hat-based distributions, global network parameters such as the overall hostname are specified in /etc/sysconfig/network, while individual interface properties including IP address, subnet mask, and gateway details are configured inside device-specific scripts located in /etc/sysconfig/network-scripts/ifcfg-nic. Primary Domain Name System server addresses are declared inside /etc/resolv.conf.

Network administration and interface manipulation are performed using dedicated terminal commands. Interface detection and configuration are managed via ifconfig. Assigning an IP address and subnet mask to an interface such as loopback (lo) using ifconfig lo 192.168.1.233 netmask 255.255.255.0 requires superuser privileges. Executing this command as a non-root user triggers system permission errors: SIOCSIFADDR: Operation not permitted, SIOCSIFFLAGS: Operation not permitted, and SIOCSIFNETMASK: Operation not permitted. Prepending sudo grants the necessary permissions to apply the configuration. Standard ifconfig output displays active interfaces such as Ethernet (enp1s0), loopback (lo), and wireless (wlp2s0), detailing flags (such as UP, BROADCAST, RUNNING, MULTICAST, LOOPBACK), Maximum Transmission Unit (MTU) values (such as $1500$ or $65536$), MAC addresses (ether), transmit queue lengths (txqueuelen), IPv4 addresses (inet), subnet masks (netmask), broadcast addresses (broadcast), IPv6 addresses (inet6), prefix lengths, scope IDs, packet counters for received (RX) and transmitted (TX) traffic, byte totals, errors, dropped packets, overruns, frame errors, carrier errors, and collisions. Additional essential network commands include ping (used to test ICMP reachability and measure round-trip time statistics including packet loss percentages and minimum, average, maximum, and mean deviation latency values), ifup and ifdown (to bring network interfaces online or offline), netstat (which displays kernel IP routing tables, destination network paths, gateways, and interface mappings), tcpdump (for packet capture and network analysis), and traceroute (which traces packet routing paths to a destination up to a maximum hop count of $30$ hops using $60\text{-byte}$ probe packets).

Hardware Addressing, Local Area Networks, and IPv4 Architecture

A Media Access Control (MAC) address, also referred to as a hardware address, serves as a unique physical identifier assigned to every network interface card (NIC). The MAC address is represented as a $48\text{-bit}$ binary number formatted as six pairs of hexadecimal characters separated by colons, such as 68:05:CA:03:19:9C. A Local Area Network (LAN) comprises a set of interconnected computers linked locally via network hardware such as hubs or switches. Devices within the same LAN reside in a localized physical space and utilize matching network interface types, such as Ethernet or Wi-Fi. Within an individual LAN, data frames are delivered between machines based directly on their physical MAC addresses.

Because enterprise networks do not exist as single isolated LANs, complex network topologies consist of multiple interconnected LANs. While MAC addressing is sufficient for intra-LAN communication, it cannot scale or provide routing capabilities across distinct LAN boundaries. Inter-network communication requires a globally routable hierarchical addressing scheme provided by the Internet Protocol (IP). Under IPv4, addresses consist of $32\text{-bit}$ binary numbers structured into four groups of $8\text{-bit}$ bytes (octets). A $32\text{-bit}$ address space yields a total theoretical capacity of $2^{32} = 4,294,967,296$ unique addresses. IPv4 addresses are expressed in three primary notation formats: binary notation (a $32\text{-bit}$ sequence presented as four $8\text{-bit}$ binary groupings separated by spaces), hexadecimal notation (four $8\text{-bit}$ groupings represented by two hexadecimal characters each, separated by spaces), and dotted-decimal notation (where each $8\text{-bit}$ octet is converted to its decimal equivalent between $0$ and $255$, separated by period delimiters in the form N.N.N.N).

Classful IP addressing categorizes the IPv4 address space into five distinct classes based on leading bit patterns to divide network identification (Net ID) from host identification (Host ID):

• Class A addresses reserve the leading bit as 0. The Net ID occupies $7\text{-bits}$ ($8\text{-bits}$ total length) and the Host ID occupies $24\text{-bits}$. This class supports $2^7 = 128$ distinct networks, each capable of addressing $2^{24} = 16,777,216$ hosts. The address range spans from 0.0.0.0 to 127.255.255.255.

• Class B addresses set the leading bits to 10. The Net ID occupies $14\text{-bits}$ ($16\text{-bits}$ total) and the Host ID occupies $16\text{-bits}$. Class B accommodates $2^{14} = 16,384$ networks with $2^{16} = 65,536$ host addresses per network, spanning from 128.0.0.0 to 191.255.255.255.

• Class C addresses set the leading bits to 110. The Net ID occupies $21\text{-bits}$ ($24\text{-bits}$ total) and the Host ID occupies $8\text{-bits}$. Class C provides $2^{21} = 2,097,152$ networks with $2^8 = 256$ addresses per network, ranging from 192.0.0.0 to 223.255.255.255.

• Class D addresses set the leading bits to 1110 and are dedicated exclusively to multicast addressing. Net ID and Host ID divisions are not defined. The address range spans from 224.0.0.0 to 239.255.255.255.

• Class E addresses set the leading bits to 1111 and are reserved for experimental and future use. Net ID and Host ID divisions are not defined. The address range spans from 240.0.0.0 to 255.255.255.255.

To facilitate efficient network routing, physical networks are logically partitioned into smaller sub-networks termed subnets. Network interfaces operating within the same subnet share identical upper-order network bits in their IP addresses while varying in their lower-order host bits. The exact number of shared bits determines the physical size of the subnet and is defined by the subnet mask. For instance, subnetworks 141.14.2.0, 141.14.7.0, and 141.14.22.0 operating under a subnet mask of 255.255.255.0 share the leading $24\text{-bits}$ as network bits, allocating the remaining $8\text{-bits}$ for local host addressing.

IP Address Allocation and Packet Routing Mechanics

IP addresses can be assigned to network hosts through static or dynamic allocation methods. Static allocation involves manually entering configuration parameters directly into system configuration files or applying them via startup scripts. A statically allocated address remains permanently assigned to the machine regardless of reboots. Dynamic allocation relies on the Dynamic Host Configuration Protocol (DHCP). Upon system startup, the host issues a broadcast request to locate a local DHCP server. The DHCP server selects an unassigned IP address from its designated subnet pool and leases it to the requesting machine. Under dynamic allocation, a host's IP address may change across consecutive boot cycles. Alternatively, DHCP servers can be configured to map specific hardware MAC addresses to fixed IP assignments. Dynamic allocation offers greater administrative flexibility but requires that the DHCP server reside on the same LAN as client hosts or be accessible via a relay agent.

Routing is the process by which network packets are forwarded across distinct subnetworks toward their destination addresses. When a host on subnet 141.14.22.0 (such as IP address 141.14.22.83) transmits a packet to a host on a separate subnet 141.14.2.0 (such as IP address 141.14.2.105), the local host determines that the destination IP does not belong to the local subnet. The packet is directed to the local default gateway router (e.g., 141.14.22.1), which examines its internal IP routing tables and forwards the packet across intermediate router hops until it reaches the destination gateway (e.g., 141.14.2.1) and client. Diagnostic tools such as traceroute trace these intermediate router hops, recording latency measurements for each hop along the path to destination hosts.

File Transfer Protocol (FTP) Architecture and Operations

File Transfer Protocol (FTP) operates at the application layer of the TCP/IP protocol suite to provide a simple, secure mechanism for exchanging files over IP networks. Transferring files from a client machine to a remote server is termed uploading, whereas transferring files from a server to a client is termed downloading. Accessing an FTP server requires an FTP client program, such as FileZilla, Fire FTP, or Dreamweaver. Rather than physically moving files, FTP copies files across network boundaries while preserving source structures and hiding underlying platform-specific file system differences.

FTP functions strictly within a client/server environment. The remote machine operates an FTP daemon task in the background (named ftpd in UNIX environments) that listens continuously for incoming client service requests. The protocol processes single file transfers sequentially and supports administrative actions including creating directories, deleting directories, listing directory contents, deleting files, renaming files, and enforcing file ownership and access permissions.

The FTP protocol separates control signaling from data transfer by utilizing distinct connection channels:

• Control Connection: Established on TCP port $21$. All client authentication commands and server response codes travel across this channel.

• Connection Modes: Modern FTP clients negotiate either an active or passive connection mode to dictate how the data channel is established.

  • Active Connection: The client establishes the control channel on port $21$ and issues a PORT command specifying an unprivileged client port. The FTP server initiates the data connection from its local TCP port $20$ to the client's specified high-range port.

  • Passive Connection: The client transmits a PASV command over the control channel. The server binds an unprivileged port higher than $10,000$ for the session and replies with its IP address and the bound port number. The client then initiates the data connection to the server's newly bound high port. Passive mode is preferred when clients reside behind firewalls or Network Address Translation (NAT) devices. Each individual data transfer request opens a distinct data connection.

FTP interaction is governed by standard protocol commands divided into three categories:

• Access Control Commands: USER (specifies user identification string), PASS (specifies password string; must immediately precede user identification validation steps), ACCT (identifies user account; required if password authentication returns code $332$, unnecessary if response is $230$), CWD (Change Working Directory to specified target path), CDUP (Change to Parent Directory), SMNT (Structure Mount), REIN (Reinitialize connection session), and QUIT (terminates session after completing active transfers).

• Transfer Parameter Commands: PORT (defines client IP and port parameters), PASV (requests server DTP to listen passively on a randomly selected available port), TYPE (specifies data representation format), STRU (specifies file structure: F for File structure, R for Record structure, P for Page structure), and MODE (specifies transfer mode: S for Stream mode, B for Block mode, C for Compressed mode).

• FTP Service Commands: RETR (retrieves a copy of the specified file from server), STOR (stores data received over data channel into target file, overwriting existing files), STOU (stores data into a uniquely named file created by server), APPE (appends transferred data to existing file or creates new file), ALLO (allocates required server storage space), REST (restarts transfer from a specified offset point; must immediately follow a transfer command), RNFR (specifies existing file path to be renamed; must be followed immediately by RNTO), RNTO (specifies new target file name), ABOR (aborts active data transfer while retaining control channel), DELE (deletes specified file; irreversible and validated at client level), RMD (removes directory), MKD (creates directory), PWD (prints current working directory path), LIST (returns detailed directory listing), NLST (returns simple list of file names), SITE (invokes site-specific server extensions), SYST (queries remote operating system type), STAT (returns current server status), HELP (returns server command help information), and NOOP (executes no operation; prevents connection timeout during inactive sessions).

Network File System (NFS) Architecture, Daemons, and Administration

Network File System (NFS) is a widely deployed distributed file system protocol designed by Sun Microsystems (Walsh et al., 1985; Sandberg et al., 1985). Originally evolving from version 2 to version 3 (NFSv3) and version 4 (NFSv4), NFS enables Linux and UNIX systems to share directories and mount remote file systems locally across networks. NFS abstracts interconnected workstations into independent systems with distinct local file systems, permitting explicit file sharing over client-server relationships where a system can simultaneously function as both client and server. NFS relies on stateless protocol semantics, supports standard UNIX file system semantics, operates transport-independently, and utilizes default TCP/UDP port $2049$. Red Hat Enterprise Linux 7 (RHEL7) supports NFSv3, NFSv4.0, and NFSv4.1.

NFS functionality relies on underlying Remote Procedure Call (RPC) daemons and system services:

• rpcbind: Maps RPC program numbers to universal network addresses listening on TCP/UDP port $111$. RPC services register their listening ports and program numbers with rpcbind upon initialization. In Linux, rpcbind (formerly portmap started via /etc/rc.d/init.d/portmap) must be active before RPC services start. The command rpcinfo -p inspects registered RPC services.

• nfs-server: Core service process enabling client access to exported NFS shares.

• nfs-lock / rpc.statd: Implements Network Status Monitor (NSM) protocols to perform file lock recovery when an NFS server reboots following a crash.

• nfs-idmap / rpc.idmapd: Provides NFSv4 name mapping, translating user/group names to local UIDs/GIDs.

• rpc.mountd: NFS mount daemon implementing server-side mount request processing for NFSv3 clients.

• rpc.rquotad: Provides user disk quota information to remote clients.

NFS configuration is managed through three primary files: /etc/exports (the primary server configuration file where exported shared directories and access parameters are declared), /etc/fstab (the client configuration file containing mount entries for mounting NFS shares automatically during system boot), and /etc/sysconfig/nfs or /etc/nfsconfig (defines listening ports and supported NFS versions). Export options declared within /etc/exports include ro (granting read-only client access), rw (granting read-write client access), sync (requiring changes to be committed to non-volatile storage before confirming client requests), no_subtree_check (disables subtree checking to improve transfer reliability when subdirectories of large file systems are exported), and no_root_squash (allows root users on clients to retain root privileges on the exported directory).

Installing and exporting an NFS share on an Ubuntu Linux system requires executing: sudo apt updatesudo apt install nfs-kernel-serversudo mkdir -p /mnt/nfs_sharesudo chown -R nobody:nogroup /mnt/nfs_share/sudo chmod 777 /mnt/nfs_share/ Access configuration is declared in /etc/exports by appending /mnt/nfs_share 192.168.43.0/24(rw,sync,no_subtree_check). Exports are applied by running sudo exportfs -a and restarting the server via sudo systemctl restart nfs-kernel-server. Firewall access is configured using sudo ufw allow from 192.168.43.0/24 to any port nfs, sudo ufw enable, and verified via sudo ufw status.

On RHEL/CentOS systems, package installation is completed using yum install nfs-utils nfs-utils-lib along with yum install portmap (for NFSv2/v3) or yum install rpcbind (for NFSv4). Services are initialized using /etc/init.d/portmap start or /etc/init.d/rpcbind start and /etc/init.d/nfs start, with automatic boot enabling configured via chkconfig --level 35 portmap on (or rpcbind) and chkconfig --level 35 nfs on. Export management uses exportfs utility flags: exportfs -a (exports all shares in /etc/exports), exportfs -v (displays active shared directories and export options), exportfs -u (unexports shares), and exportfs -r (refreshes server share lists after editing /etc/exports).

Domain Name Services (DNS) Architecture and Resolution Mechanisms

Domain Name Services (DNS) provides a globally distributed lookup mechanism that translates human-memorable domain names into numeric IP addresses and resource descriptors. In the 1970s ARPANET era, host resolution relied on a centralized HOST.txt file maintained by SRI-NIC on a single host. As network scale grew, HOST.txt suffered from severe bandwidth load, name collisions, and consistency failures. DNS was created in 1983 by Paul Mockapetris through RFC 1034 and RFC 1035, introducing a distributed architecture. DNS consists of three core components: a hierarchical namespace, servers hosting namespace data, and client resolvers querying servers for records.

DNS architecture exhibits five essential design features:

• Global Distribution: DNS data is maintained locally across thousands of independent zone servers but remains globally retrievable. Remote query responses are cached locally by resolvers to enhance performance.

• Loose Coherency: Internal database consistency is maintained via incremental serial numbers assigned to each zone version. Changes made to master zone files increment the zone serial number, triggering replica synchronization across slave servers according to timing intervals defined by zone administrators. Cached records expire according to Time To Live (TTL) values defined in the zone data.

• Scalability: The database supports unlimited expansion; individual enterprise DNS servers can host over $20,000,000$ names and handle upwards of $24,000\text{ queries per second}$ distributed across master servers, slave servers, and caching resolvers.

• Reliability: Data replication maintains availability. Slave servers maintain authoritative copies received from master servers, allowing clients to query masters, slaves, or local caches. DNS protocol operations utilize UDP port $53$ (with DNS handling packet retransmissions and sequencing) or TCP port 53$.\n- Dynamicity: DNS supports dynamic updates, allowing dynamic record additions, deletions, and modifications. Modifications to master databases trigger slave replication; however, master servers represent single points of failure for dynamic update operations.\n\nDNS domain structures follow a Fully Qualified Domain Name (FQDN) hierarchy, such as `WWW.RIPE.NET.`, where individual labels are separated by dots. FQDNs map to resource records (RRs) within a tree structure. Resource records define name-to-resource mappings, such as `www.ripe.net. A 10.10.10.2` (where `A` designates an IPv4 Address Resource). Domains represent namespaces where all sub-elements fall under parent branches (e.g., everything below `.com` belongs to the `com` domain). Parent administrators can delegate operational authority for subdomains to child zone administrators while retaining structural parent-to-child delegation links. Administrative spaces boundaries are defined as zones.\n\n# Dynamic Host Configuration Protocol (DHCP) Operations and Management\n\nDynamic Host Configuration Protocol (DHCP) was developed by the IETF Dynamic Host Configuration working group as an extension of the BOOTP protocol. Operating over UDP on port 67$(for server listening) and port$68 (for client listening), DHCP automates the supervision, centralized management, and distribution of IP addresses, default gateway router parameters, subnet masks, and network configuration settings to hosts. DHCP provides temporary IP address bindings, making it well-suited for environments where mobile hosts regularly connect and disconnect from networks.\n\nKey characteristics of DHCP include centralized IP administration, backward compatibility with BOOTP (allowing BOOTP client software to request static configurations from DHCP servers), support for multiple redundant servers, and support for both dynamic and static IP address assignments. DHCP operates independently and does not interact with Domain Name Services (DNS).\n\nDHCP architecture relies on three primary software components:\n- Client Software: Runs on client endpoints to generate broadcast discovery requests and process acquired network parameters.\n- Server Software: Operates on centralized servers to manage IP address pools, process client lease requests, and assign configuration data.\n- Relay Agent Software: Because routers block network layer broadcast messages, client DHCP broadcasts cannot cross subnet boundaries. DHCP relay agents operate on local routers, intercepting local DHCP broadcast requests, repackaging them into unicast IP packets, and forwarding them directly to remote DHCP servers. Server replies are received by the relay agent and forwarded back to the requesting client.\n\nDHCP manages address assignments using internal databases and lease mechanisms:\n- Databases: The server maintains a manual database (binding permanent IP addresses to client hardware MAC addresses identically to BOOTP) and a dynamic pool database (containing ranges of unassigned IP addresses allocated on a First-Come, First-Served basis).\n- Lease Types: Addresses are issued via Manual Leases (explicit network manager static binding), Automatic Leases (permanent assignment by server), or Dynamic Leases (temporary assignment for a defined duration).\n- Lease Lifecycle: When a client initializes, it broadcasts a DHCP Discover message. Listening DHCP servers reply with a DHCP Offer. The client accepts an offer and requests the address. The server issues an acknowledgment confirming the lease. When half of the lease duration (\frac{1}{2}\text{ lease time}$) elapses, the client attempts to renew its lease with the server. If a lease expires without renewal, the client must cease using the IP address.</p></li></ul><p>DHCP reduces administrative burdens and eliminates manual IP configuration conflicts. However, DHCP introduces operational dependencies: if the central DHCP server becomes unavailable, non-configured clients cannot access network services. Additionally, client hostnames remain static even when IP addresses change dynamically, and core network infrastructure (such as servers and routers) must retain fixed static IP addresses. Security vulnerabilities exist because DHCP is an unauthenticated protocol. Malicious users with local network access can execute Denial of Service (DoS) attacks by requesting numerous address leases, completely exhausting the available IP address pool.</p><h3 id="afadf2ea-85c3-4429-b17a-0bc68ed12b86" data-toc-id="afadf2ea-85c3-4429-b17a-0bc68ed12b86" collapsed="false" seolevelmigrated="true">Network Information Services (NIS) Centralized Administration</h3><p>Network Information Services (NIS) is an administrative naming service designed to resolve consistency problems in distributed UNIX computing environments. In non-centralized environments, each workstation maintains independent local copies of critical administrative files, such as <code>/etc/passwd</code>, <code>/etc/group</code>, and <code>/etc/hosts</code>. Propagating account updates across every workstation manually is inefficient and prone to synchronization errors. NIS solves this by establishing uniform storage and retrieval of network-wide administrative databases, known as NIS Maps, across network hosts in a transport-protocol and media-independent manner.</p><p>While DNS focuses on resolving machine hostnames to IP addresses to simplify network communication, NIS focuses on network administration by centralizing control over user accounts, security credentials, network groups, and system services. Unlike the hierarchical structure of DNS, the NIS namespace is completely flat.</p><p>NIS operates on a client-server architecture using three primary machine roles:</p><ul><li><p>Master Server: A designated single server per NIS domain that holds the primary source files and master NIS maps. All administrative updates must occur on the master server.</p></li><li><p>Slave Servers: Redundant servers that hold duplicate copies of master NIS maps. Slave servers process client read queries, minimizing master server load and providing high availability.</p></li><li><p>NIS Clients: Host systems that execute NIS client processes to query maps from master or slave servers. Servers must possess local disk storage, whereas clients do not require local disks. The NIS server does not need to reside on the same hardware as the NFS file server.</p></li></ul><p>NIS architecture components are categorized into domains, daemons, utilities, and maps:</p><ul><li><p>NIS Domains: Logical groupings of machines sharing a common set of NIS maps under a unified domain name.</p></li><li><p>NIS Daemons: <code>ypserv</code> (the core server process that responds to client map requests), <code>ypbind</code> (the client binding daemon that locates and connects clients to active <code>ypserv</code> processes), <code>ypxfrd</code> (accelerates high-speed map transfers between servers), <code>rpc.yppasswdd</code> (handles remote NIS password modifications), and <code>rpc.ypupdated</code> (processes modifications to auxiliary maps such as <code>publickey</code>).</p></li><li><p>NIS Utilities: <code>makedbm</code> (converts text source files into <code>ndbm</code> format binary database maps), <code>ypcat</code> (displays all key-value entries stored inside a specified map), <code>ypinit</code> (builds and installs initial NIS map databases and client server lists), <code>ypmatch</code> (searches a map for a specific key entry), <code>yppoll</code> (queries a server for the order/serial number of a map), <code>yppush</code> (propagates updated maps from master to slave servers), <code>ypset</code> (forces <code>ypbind</code> to bind to a specified server), <code>ypwhich</code> (identifies the active NIS server and displays nickname translation tables), and <code>ypxfr</code> (transfers map data directly from master to slave servers).</p></li><li><p>NIS Maps: Two-column key-value tables stored in <code>ndbm</code> format. NIS maps replace standard <code>/etc</code> UNIX configuration files. Information may be duplicated across multiple maps to allow lookups using different keys (e.g., searching users by UID or username).</p></li><li><p>Map Maintenance: The <code>make</code> command updates NIS maps by executing the Makefile located in <code>/var/yp</code>.</p></li></ul><p>NIS client binding occurs through two distinct operational modes:</p><ul><li><p>Server-list Mode: The <code>ypbind</code> process reads the list of authorized NIS servers from <code>/var/yp/binding/domain/ypservers</code> (generated via <code>ypinit -c</code>) and binds exclusively to servers specified in this file.</p></li><li><p>Broadcast Mode: The <code>ypbind</code> process transmits RPC broadcast packets across the local subnet to discover any active <code>ypserv</code> process. In multi-subnet networks, routers are often configured as NIS servers to serve client binding requests across subnet boundaries.</p></li></ul><p></p><ul><li><p>Linux networking is essential for system admins and developers in Linux environments.</p></li><li><p>Establishing network functionality is vital for:</p><ul><li><p>Internet connectivity</p></li><li><p>System interface properties and IP configurations</p></li><li><p>Remote file transfer</p></li><li><p>Remote administrative access</p></li></ul></li><li><p>In Oracle VirtualBox, resolve internet access issues with the following steps:</p><ol><li><p>Open VirtualBox Manager.</p></li><li><p>Select the target virtual machine from the left pane.</p></li><li><p>Navigate to the Settings menu.</p></li><li><p>Select Network from the left pane.</p></li><li><p>Modify the attached network type (select Bridged Adapter).</p></li><li><p>Save settings and start the virtual machine.</p></li></ol></li><li><p>Core networking parameters include:</p><ul><li><p><strong>IP Address</strong>: Unique identifier for computers using Internet Protocol.</p></li><li><p><strong>Subnet Mask</strong>:$32\text{-bit}$number that divides address space into network and host components.</p></li><li><p><strong>Default Gateway</strong>: Network point that acts as an access point to another network.</p></li><li><p><strong>Static vs. Dynamic Address Assignment</strong>: Static is manually set; dynamic is assigned automatically.</p></li><li><p><strong>Network Interface Cards (NIC)</strong>: Hardware that connects computers to a network.</p></li><li><p><strong>MAC Addresses</strong>: Unique identifiers for network interfaces.</p></li></ul></li><li><p>Linux system configuration files include:</p><ul><li><p><strong>/etc/nsswitch.conf</strong>: Hosts resolution mechanism.</p></li><li><p><strong>/etc/hostname</strong>: System IP address and hostname configuration.</p></li><li><p><strong>/etc/sysconfig/network (Red Hat)</strong>: Global network parameters.</p></li><li><p><strong>/etc/sysconfig/network-scripts/ifcfg-nic</strong>: Interface-specific properties.</p></li><li><p><strong>/etc/resolv.conf</strong>: DNS server addresses.</p></li></ul></li><li><p>Key commands for network administration include:</p><ul><li><p><code>ifconfig</code>: Detects and configures interface.</p></li><li><p><code>ping</code>: Tests ICMP connectivity.</p></li><li><p><code>ifup</code> and <code>ifdown</code>: Bring interfaces online or offline.</p></li><li><p><code>netstat</code>: Displays network statistics.</p></li><li><p><code>tcpdump</code>: Captures network packets.</p></li><li><p><code>traceroute</code>: Traces packet paths to a destination.</p></li></ul></li><li><p>MAC address format:$48\text{-bit}$binary displayed as six pairs of hexadecimal characters.</p></li><li><p>Local Area Network (LAN): Interconnected computers using the same network hardware.</p></li><li><p>Subnets: Smaller networks within large networks to improve routing efficacy.</p></li><li><p>IP Addressing:</p><ul><li><p>Static IPs are manually assigned.</p></li><li><p>Dynamic IPs are assigned by the DHCP protocol, leased temporarily.</p></li><li><p>Routing involves forwarding packets through routers to reach destination IPs.</p></li></ul></li><li><p>File Transfer Protocol (FTP) functions at the application layer for file exchanges:</p><ul><li><p>FTP supports two connection modes: active and passive.</p></li><li><p>Utilizes TCP port$21$for control connections and port$20$for data.</p></li></ul></li><li><p>Network File System (NFS) facilitates remote file sharing:</p><ul><li><p>Relies on RPC daemons and services.</p></li><li><p>Core configuration files include <code>/etc/exports</code> and <code>/etc/fstab</code>.</p></li></ul></li><li><p>Domain Name Services (DNS) translates domain names to IP addresses:</p><ul><li><p>Maintains hierarchical data structure.</p></li><li><p>Operates on UDP port$53$.</p></li></ul></li><li><p>Dynamic Host Configuration Protocol (DHCP) automates IP address distribution:</p><ul><li><p>Operates over UDP ports$67$(server) and$68$$ (client).

• Supports both dynamic and static IP allocations.