CompTIA Server+ flashcards
CompTIA Server+ · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- RAID 6
- Striping with dual parity that tolerates the simultaneous failure of two drives and requires a minimum of four drives to implement.
- LRDIMM
- Load-reduced DIMM that uses rank multiplication and buffering to support very high memory capacities on servers, at slightly higher latency than RDIMMs.
- Hot-swap backplane
- Drive bay connector plane that lets drives be removed and replaced while the server stays powered on, provided the controller and operating system support it.
- Rack unit height
- One rack unit equals 1.75 inches or 44.45 mm of vertical space; a standard full-height rack is 42U tall.
- Redundant power supply
- Two or more PSUs sharing the server load so that a single PSU or feed failure will not power down the system; size for N+1.
- Baseboard management controller
- Independent chip that provides out-of-band access to power control, remote console, hardware sensors, and event logs even when the operating system is down.
- ECC memory
- Memory that detects and corrects single-bit errors and detects double-bit errors; it requires chipset support and cannot be mixed with non-ECC modules in a server.
- NVMe
- Storage protocol designed for flash over PCIe, providing much lower latency and far higher queue depth than SATA or SAS drives.
- Type 1 hypervisor
- Bare-metal hypervisor that runs directly on the server hardware and hosts guest virtual machines, with examples including ESXi and Hyper-V.
- VM snapshot
- Point-in-time capture of a virtual machine disk delta; convenient for short-term rollback but not a backup because it grows and depends on the base disk.
- DHCP reservation
- Scope entry that always assigns the same IP address to a specific MAC address, combining centralized configuration with a stable, predictable address.
- NTP stratum
- Number indicating how far a time source sits from the reference clock; stratum 1 syncs directly to it and each level adds one.
- SNMP trap
- Unsolicited alert sent from an agent to a management station when a threshold is crossed or an event occurs, rather than waiting for a poll.
- Role-based access control
- Model that assigns permissions to roles instead of individuals, so each user inherits only the access required by their job function.
- Pilot-ring patching
- Deploying updates to a small canary group of servers first, verifying stability, then widening the rollout to the remaining fleet.
- DNS A record
- Maps a hostname to an IPv4 address; AAAA records map to IPv6, while a CNAME aliases one name to another name.
- Three-two-one backup rule
- Keep at least three copies of data, on two different media types, with one copy stored offsite to survive site-level loss.
- Incremental backup
- Copies only data changed since the last backup of any type, so restoring requires the last full plus every incremental in sequence.
- Differential backup
- Copies all data changed since the last full backup, so a restore needs only the last full plus the most recent differential.
- Recovery time objective
- Maximum acceptable downtime before a service must be restored after an outage; it drives the choice of hot, warm, or cold recovery site.
- Recovery point objective
- Maximum acceptable amount of data loss measured in time from the last known good copy, which sets the required backup or replication frequency.
- Trusted Platform Module
- Chip that stores cryptographic keys and boot measurements used for disk encryption and for verifying that the boot chain is unaltered.
- Cold site
- Disaster recovery facility with space and power but no pre-installed equipment or current data, giving the longest recovery time at the lowest cost.
- Chain of custody
- Documented record of who handled forensic evidence, when, and where, required to keep that evidence admissible in an investigation or legal proceeding.
- POST
- Power-on self-test; firmware routine that checks core hardware at boot and reports failures through beep codes, diagnostic LEDs, or error messages.
- SMART
- Self-Monitoring, Analysis and Reporting Technology; a drive feature that reports predictive health attributes such as reallocated sectors and pending failures.
- Kernel panic
- Fatal operating system error in which the kernel halts to prevent data corruption; the cause is traced through crash dumps and system logs.
- Memory ballooning
- Hypervisor technique that reclaims guest RAM by inflating a driver inside the virtual machine, which can force guest swapping and degrade performance.
- CPU ready time
- Percentage of time a virtual machine was ready to run but waiting for physical CPU cycles; high values indicate host CPU oversubscription.
- Traceroute
- Tool that sends packets with progressively increasing TTL values to list each router along the path to a destination, locating where traffic stops.
- Degraded RAID array
- State in which an array still serves data with a failed or missing member, so redundancy is gone and a rebuild is required.
- Stop error
- Bug-check code produced when Windows halts unexpectedly; the code and dump point to the failing driver or hardware condition.