CompTIA Server+ key terms

A standard vertical measurement equal to 1.75 inches (44.45 mm) used to describe the height of servers and rack-mounted equipment. A 2U server occupies two rack units, which helps when planning cabinet space, power, and airflow.

Two or more power supplies feeding a single server or chassis so the system keeps running if one supply or its power feed fails. Supplies are usually hot-swappable and may connect to separate electrical circuits.

A design feature that lets a component such as a drive, fan, or power supply be removed and replaced while the server stays powered on and serving workloads, avoiding scheduled downtime for routine hardware replacement.

An independent controller on the motherboard that provides remote access to a server's power, console, and sensors even when the operating system is down. It is usually reached over a dedicated network port.

A method of combining multiple physical drives into one logical unit to gain speed, fault tolerance, or both. Common levels include 0, 1, 5, 6, and 10, each with different cost and protection trade-offs.

Memory that detects and corrects single-bit errors and reports larger faults, protecting servers that run continuously. ECC requires support from the motherboard, CPU, and memory modules working together.

A protocol for accessing solid-state storage over a PCIe bus rather than through a legacy disk controller. NVMe drives offer very low latency and deep queues, making them common for databases and virtualization hosts.

A battery-backed device that keeps servers and network gear running during short power outages and smooths voltage sags. Larger units can signal servers to shut down gracefully when battery charge runs low.

A thin modular server that slides into a shared chassis supplying power, cooling, networking, and management. Blades reduce cabling and rack space compared with standalone rack-mounted servers.

Low-level software stored on a device that controls basic hardware behavior, such as BIOS or UEFI, drive controllers, and network adapters. Updates fix bugs and security flaws but must be applied carefully.

Hardware that lets one keyboard, monitor, and mouse control multiple servers. Rack-mounted KVM consoles are often used during maintenance when remote management access is unavailable.

Software that runs virtual machines by sharing a host's CPU, memory, storage, and network among guests. A Type 1 hypervisor runs directly on hardware; Type 2 runs as an application on another operating system.

A Windows server that stores directory data and handles authentication and authorization for a domain. Multiple controllers replicate with each other so logons continue if one server goes offline.

A defined range of IP addresses that a DHCP server can lease to clients, along with options such as default gateway, DNS servers, and lease duration. Separate scopes serve separate subnets.

A portion of the DNS namespace held on a server, containing records that map names to addresses and services. Forward zones resolve names to addresses; reverse zones resolve addresses to names.

A logical network segment created by configuring switches so devices share a broadcast domain regardless of physical location. VLANs separate traffic for security and reduce broadcast congestion on busy networks.

Bonding two or more physical network adapters into one logical link to raise throughput or survive an adapter or cable failure. The attached switch must support the chosen teaming mode.

A collection of configuration settings in Active Directory applied to users and computers in sites, domains, or organizational units. GPOs enforce security, software, and desktop standards across many servers.

A repeatable process for identifying, testing, deploying, and verifying updates to operating systems, applications, and firmware. It reduces exposure to known vulnerabilities while limiting disruption to production services.

A recorded measurement of normal server performance or configuration taken while the system is healthy. Later readings are compared against it to spot drift, bottlenecks, or unauthorized changes.

A point-in-time copy of a volume or virtual machine that can be rolled back quickly after a failed change. Snapshots are not a substitute for real backups kept on separate storage.

A standard protocol and message format for sending event records from servers and network devices to a central collector. Centralized logging helps correlate events and preserve records after a failure.

Granting users and services only the rights required to perform their duties, and no more. It limits the damage that compromised accounts or mistaken administrative changes can cause.

Authentication requiring two or more independent factors, such as a password plus a code from a phone or a hardware token. It reduces the risk that stolen credentials alone lead to a breach.

The maximum acceptable amount of data loss expressed in time before an incident. An RPO of one hour means backups or replication must capture data at least that often.

The maximum acceptable time to restore a service after a disruption. RTO drives the choice of recovery site, spare hardware, and automation in a continuity plan.

A practice of keeping at least three copies of data, on two different media types, with one copy stored off site. It protects against device failure, ransomware, and site-wide disasters.

A documented set of procedures, roles, and resources for restoring IT services after a major outage. It is tested on a schedule and updated as systems and business needs change.

Reducing a server's attack surface by removing unneeded services, closing unused ports, applying secure configuration, restricting accounts, and enabling auditing. The result is validated against a documented standard.

The combination of certificate authorities, policies, and software that issues and validates digital certificates. Servers use it for encrypted web traffic, authentication, and code or configuration signing.

A documented record of who handled evidence, when, and where, from collection through storage and analysis. It keeps forensic findings admissible and defensible during an investigation.

A self-test the server firmware runs at power-on to check the CPU, memory, and core devices before loading the operating system. Failures are reported through beep codes, indicator lights, or log entries.

Indicators on the chassis, motherboard, or components that show power state, drive activity, fan faults, or specific error conditions. They narrow down hardware problems when a server will not boot.

A connector that sends a port's transmit signal back to its own receive pin to verify that the interface and driver work. It helps separate a failed adapter from a cabling or switch problem.

A structured investigation that moves past the immediate symptom to find the underlying reason an incident occurred, then records corrective actions so the same problem does not return.

A controlled process for requesting, reviewing, approving, scheduling, and rolling back changes to production systems. It reduces outages caused by untested or undocumented modifications.

Passing an unresolved issue to a more skilled person, team, or vendor according to defined criteria and time limits. It keeps incidents moving and sets clear expectations for the user.

A fatal error in which the operating system kernel detects an unrecoverable condition and halts the machine. The crash dump and logs help identify faulty drivers, memory, or hardware.