I85PREP
⌘ K
CompTIA Server+ study materials
CompTIA Server+ study resources

CompTIA Server+ practice questions

One question at a time. Build your CompTIA Server+ understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Server hardware installation and management

A server is described in a quote as occupying 2U of rack space. What does the U represent?

  1. A vertical height of 1.75 inches (44.45 mm) Correct answer.
  2. The number of processor sockets on the board
  3. The maximum continuous power draw in watts
  4. A horizontal width of 19 inches (482 mm)

Explanation

Rack unit (U) is a measure of vertical height in a standard 19-inch equipment rack, where 1U equals 1.75 inches (44.45 mm). Width is fixed by the rack standard and has no relationship to the U count.

Question 2Server hardware installation and management

A technician is about to replace memory modules in a running rack environment. Where should the antistatic wrist strap be connected?

  1. Wear it on the ankle instead of the wrist
  2. Use it only when handling hard disk drives
  3. Clip it to unpainted bare metal on the chassis Correct answer.
  4. Attach it to a painted panel on the outside of the rack enclosure

Explanation

A wrist strap works by keeping the technician at the same electrical potential as the equipment, which requires a connection to bare, unpainted metal on the chassis. Paint and coated surfaces act as insulators and defeat the purpose.

Question 3Server hardware installation and management

Which statement accurately describes the power and cooling design of a blade server deployment?

  1. Blades connect directly to the top-of-rack switch without an enclosure
  2. Blades mount in a standard rack without any chassis or enclosure
  3. Each blade must be connected to its own dedicated power supply
  4. Blades share power, cooling, and networking through the enclosure Correct answer.

Explanation

Blade servers are modular compute nodes that slide into a chassis which supplies shared power, shared cooling, and shared network and storage connectivity. This shared infrastructure is what makes blades denser than comparable rack servers.

Question 4Server hardware installation and management

What is the primary function of a KVM switch in a server rack?

  1. Distributing utility power to several servers from one UPS
  2. Aggregating multiple network links into one logical link
  3. Providing shared storage access for a server cluster
  4. Letting one keyboard, monitor, and mouse control several servers Correct answer.

Explanation

A KVM switch multiplexes a single keyboard, video display, and mouse across multiple attached systems, which reduces the peripherals needed in a dense rack. It does not carry power, storage, or network traffic.

Question 5Server hardware installation and management

A drive bay is labeled as hot-swappable. What does this mean for the administrator?

  1. The drive can be installed in any bay regardless of controller support
  2. The drive can be removed only after the operating system shuts down
  3. The drive spins faster than a comparable cold-swap unit
  4. The drive can be replaced while the server stays powered on Correct answer.

Explanation

Hot-swap capability means the drive carrier can be removed and replaced without powering down the system, provided the controller and backplane support it. The remaining condition is that the array or volume is configured to tolerate the removal.

Question 6Server hardware installation and management

A server has two power supplies. Which configuration provides true power redundancy?

  1. One supply stays unplugged as a spare until the other one fails
  2. Each supply is fed from a different power source Correct answer.
  3. Both supplies are plugged into the same PDU so the load is balanced between them
  4. The supplies must be different wattages to provide redundancy

Explanation

Redundancy is only effective when the two supplies draw from independent sources, such as separate circuits, PDUs, or UPSs. If both share one feed, a single breaker trip or PDU failure still takes the server down.

Question 7Server hardware installation and management

After a new server is installed in a rack, what step best supports future maintenance and tracking?

  1. Reimage every other server in the rack to keep firmware versions identical
  2. Log the serial number, location, and configuration in the asset inventory Correct answer.
  3. Write the local administrator password on a label on the front bezel
  4. Disable the management controller to reduce the attack surface

Explanation

Recording serial number, physical location, and hardware configuration creates the record that later supports warranty claims, firmware tracking, and physical locating of the unit. Asset records are the standard mechanism for this.

Question 8Server hardware installation and management

What does UEFI Secure Boot do during the startup process?

  1. Encrypts the boot partition with a key sealed in the TPM
  2. Limits the boot volume size to 2 TB
  3. Verifies the digital signature of boot code before it executes Correct answer.
  4. Forces the server to boot from the network before the local disk

Explanation

Secure Boot validates the signatures of bootloaders and drivers against trusted keys stored in firmware, blocking unsigned or tampered code from running. It does not itself encrypt volumes or set boot order.

Question 9Server hardware installation and management

A RAID 5 array is built from four 2 TB drives. How much usable capacity does the array provide?

  1. 6 TB Correct answer.
  2. 2 TB
  3. 8 TB
  4. 4 TB

Explanation

RAID 5 reserves the equivalent capacity of one drive for distributed parity, so usable space equals the total raw capacity minus one drive. With four 2 TB drives that is 8 TB raw minus 2 TB parity, leaving 6 TB.

Question 10Server hardware installation and management

A four-disk RAID 10 array is created from two mirrored pairs. Which statement describes its fault tolerance?

  1. It cannot tolerate any drive failure until a hot spare is rebuilt
  2. One drive per mirror can fail before the array is lost Correct answer.
  3. It tolerates only single-drive failures because parity spans all four disks
  4. It survives any two simultaneous drive failures without data loss

Explanation

RAID 10 mirrors striped pairs, so each mirror can lose one member and keep running. Losing both drives of the same mirror destroys the array even though a different two-drive combination would have been survivable.

Question 11Server hardware installation and management

Which statement about registered DIMMs is correct?

  1. RDIMMs operate without any support from the integrated memory controller
  2. RDIMMs cannot use error-correcting code
  3. RDIMMs can be mixed with UDIMMs in the same channel
  4. RDIMMs buffer address and command signals to reduce electrical load Correct answer.

Explanation

RDIMMs place a register between the memory controller and the DRAM devices, buffering address and command signals. That reduces electrical loading and allows more modules per channel than unbuffered DIMMs.

Question 12Server hardware installation and management

A server has four DIMM slots arranged as two channels, and only two DIMMs are available. How should they be installed?

  1. Install both DIMMs in channel A to reduce access latency
  2. Install one DIMM per channel to enable interleaving Correct answer.
  3. Install a single DIMM and keep the other as a cold spare
  4. Install both DIMMs in adjacent slots of the same channel

Explanation

Balanced population across channels lets the memory controller interleave accesses, which increases effective bandwidth. Vendor population tables should be followed, and they almost always spread the first pair across channels.

Question 13Server hardware installation and management

A server will not respond on the network and its operating system appears hung. Which component allows an administrator to reach the console and power-cycle the system remotely?

  1. The UPS management card sends a reboot command over the power cable
  2. The BMC provides out-of-band console and power control Correct answer.
  3. The access-layer switch provides a serial console for attached servers
  4. The RAID controller logs the boot sequence for later replay

Explanation

A baseboard management controller operates independently of the operating system and provides remote console redirection plus power control over a dedicated or shared management interface. This is the standard path for recovering an unresponsive host.

Question 14Server hardware installation and management

Which best describes products such as iDRAC and iLO?

  1. They are vendor implementations of out-of-band management controllers Correct answer.
  2. They are firmware update utilities that run inside the operating system
  3. They are PCIe RAID controllers with battery-backed cache
  4. They are hypervisor-based backup agents installed in the guest OS

Explanation

iDRAC and iLO are vendor-specific names for embedded management controllers that provide remote console, power control, and health monitoring independent of the installed OS. They are functionally equivalent to a BMC.

Question 15Server hardware installation and management

A 1U server has no free full-height expansion slots, but a full-height network adapter must be installed. What does a riser card do in this situation?

  1. A riser card supplies redundant power to the motherboard and drives
  2. A riser card reorients PCIe slots to fit a low-profile chassis Correct answer.
  3. A riser card converts PCIe slots into additional memory slots
  4. A riser card increases the number of CPU cores available to the OS

Explanation

A riser card extends or rotates the PCIe slot so an expansion card can sit parallel to the system board, which is how full-height cards fit into low-profile chassis. It also commonly adds slots beyond those on the motherboard.

Question 16Server hardware installation and management

A server needs an NVMe solid-state drive that can be swapped from the front of the chassis without opening the case. Which form factor should be selected?

  1. A U.2 NVMe drive in a front bay Correct answer.
  2. An mSATA module in a mini PCIe slot
  3. An M.2 2280 module mounted on the system board
  4. A 3.5-inch SATA hard disk in a front bay

Explanation

U.2 (also called SFF-8639) carries PCIe/NVMe signaling through a cabled or backplane connection in a 2.5-inch carrier, so the drive can sit in a front-loading hot-swap bay. M.2 and mSATA are internally mounted and not front-accessible.

Question 17Server hardware installation and management

When seating a processor and heatsink, which practice is correct?

  1. Tighten one screw fully before moving to the opposite corner
  2. Apply a thin even layer and tighten the screws diagonally Correct answer.
  3. Coat the socket pins as well as the top of the processor
  4. Apply a thick layer so it fills every gap between CPU and heatsink

Explanation

Thermal compound is needed only to fill microscopic surface imperfections, so a thin, even layer is sufficient. Tightening fasteners in a diagonal sequence applies even pressure and avoids tilting or cracking the processor.

Question 18Server hardware installation and management

How should racks be laid out to support effective front-to-back server airflow?

  1. Blanking panels are removed from empty U space to increase airflow
  2. Perforated floor tiles are placed in the hot aisle to feed cool air
  3. Intakes face a cold aisle and exhausts face a hot aisle Correct answer.
  4. All servers exhaust into the same aisle from which they draw air

Explanation

Server airflow runs front to back, so rows are arranged intake-to-intake and exhaust-to-exhaust to create separate cold and hot aisles. Mixing the two causes recirculation and hot spots.

Question 19Server hardware installation and management

Which capability does SMART provide on server drives?

  1. SMART encrypts data at rest on the drive using a hardware key
  2. SMART copies data to a hot spare before the drive fails
  3. SMART health attributes flag drives that are likely to fail Correct answer.
  4. SMART raises the spindle speed when read errors are detected

Explanation

SMART monitors and reports attributes such as reallocated sectors and pending errors, letting administrators replace a degrading drive before it fails outright. It is a monitoring technology, not a data protection or repair function.

Question 20Server hardware installation and management

A server's measured peak power draw is 600 W. Which supply rating best reflects power-supply sizing best practice?

  1. A 650 W unit
  2. A 600 W unit
  3. A 550 W unit
  4. An 800 W unit Correct answer.

Explanation

A supply should be sized above the measured peak so it does not run at 100 percent of rating, where efficiency and thermal margin are poor. Roughly 20 to 30 percent headroom also leaves room for added components.

Question 21Server hardware installation and management

What role does a SAS expander play in a storage backplane?

  1. It lets one HBA port address many drives Correct answer.
  2. It provides redundant power to the drive bays
  3. It is required to build a RAID 0 array
  4. It converts SAS signaling to SATA signaling

Explanation

A SAS expander fans out a single host port into many downstream connections, which is how a backplane can present dozens of drives to a two-port or four-port HBA. Without it, each drive would need its own controller port.

Question 22Server hardware installation and management

An organization must update BIOS firmware on 200 production servers. Which approach is most appropriate?

  1. Push the update to all servers at once during peak business hours
  2. Test on a pilot group, then roll out in batches Correct answer.
  3. Skip the update because firmware has no effect on stability
  4. Update only the servers that happen to reboot on their own

Explanation

Validating on a small pilot group exposes compatibility problems with limited impact, and phased batches keep the blast radius small while a rollback path stays available. Firmware changes can alter boot behavior, so an untested mass push is high risk.

Question 23Server hardware installation and management

Before populating a 42U rack with dense multi-node servers, which physical constraint must be verified?

  1. Floor load rating and rack weight capacity Correct answer.
  2. Whether every server in the rack has an optical drive
  3. Whether the rack door hinges can be reversed for the room layout
  4. The color coding used for the network patch cables

Explanation

Fully populated racks can weigh well over a ton, so the data center floor and the rack itself must be rated for the total load. Exceeding either rating risks structural damage and is a safety issue.

Question 24Server hardware installation and management

A server must reach 1.5 TB of memory using all 24 available DIMM slots. Which memory type best supports that goal?

  1. Non-ECC unbuffered SODIMMs
  2. Single-rank RDIMMs of the smallest available density
  3. Load-reduced DIMMs (LRDIMMs) Correct answer.
  4. Unbuffered ECC DIMMs (UDIMMs)

Explanation

LRDIMMs buffer the data lines as well as the address and command signals, reducing the electrical load per rank. That allows higher-capacity modules to be populated across every slot, which is the usual route to very large memory configurations.

Question 25Server hardware installation and management

A high-end GPU rated at 300 W is being installed in a server. What must be done to power it correctly?

  1. Connect the auxiliary PCIe power cables and check PSU capacity Correct answer.
  2. Use a SATA-to-PCIe adapter to feed the card from a drive lead
  3. Rely on the PCIe slot, which supplies up to 300 W to the card
  4. Install a second CPU so the card can draw power from its VRM

Explanation

A PCIe x16 slot delivers a maximum of 75 W, so a 300 W accelerator requires the auxiliary 6-pin or 8-pin (or 12VHPWR) feeds from the power supply. The supply rating and number of available PCIe connectors must also be confirmed.

Question 26Server hardware installation and management

In a standard equipment rack, what does one rack unit (1U) represent?

  1. 19 inches (482.6 mm) of horizontal width
  2. One shelf position measuring 3.5 inches (88.9 mm)
  3. A weight allowance of 10 kg per mounting slot
  4. 1.75 inches (44.45 mm) of vertical height Correct answer.

Explanation

A rack unit is a measure of vertical height only, defined as 1.75 inches (44.45 mm). Equipment heights are expressed in multiples of this value, so a 2U server occupies 3.5 inches of vertical space. Rack width is a separate specification, conventionally 19 inches between mounting rails.

Question 27Server hardware installation and management

A technician is about to remove a DIMM from a powered-off server. Which practice best protects the component from electrostatic discharge?

  1. Wiping the DIMM contacts with a dry cloth immediately before seating it
  2. Storing the DIMM on a carpeted surface until it is needed
  3. Handling the DIMM only by its gold contact edges
  4. Wearing a grounded antistatic wrist strap attached to bare metal on the chassis Correct answer.

Explanation

A wrist strap with a current-limiting resistor bleeds static charge from the technician to the chassis ground, keeping both at the same potential. Components should also be handled by their edges rather than their contacts. Carpet and dry cloths generate or retain static charge.

Question 28Server hardware installation and management

Before installing a new RAID controller into an existing server, which action best confirms the component will work in that system?

  1. Verify the server model and controller appear together on the vendor HCL Correct answer.
  2. Check that the controller's firmware is the newest release available
  3. Estimate whether the controller fits within the server's power budget
  4. Compare the controller's price against similar models

Explanation

A hardware compatibility list is the vendor's tested record of which adapters, drives, and processors are supported in a given system. Confirming both the server model and the specific controller on that list avoids unsupported configurations and warranty issues. Firmware currency and power draw matter later but do not establish support.

Question 29Server hardware installation and management

A data center has eight servers in one rack, and administrators need occasional local console access to each. Which device lets a single keyboard, monitor, and mouse set serve all eight servers?

  1. A port aggregating network switch
  2. A KVM switch Correct answer.
  3. A rack-mounted uninterruptible power supply
  4. A switched rack PDU

Explanation

A KVM switch multiplexes one keyboard, video, and mouse console across multiple attached systems, so local access does not require a dedicated console per server. Power distribution and network switching serve entirely different functions.

Question 30Server hardware installation and management

In a rack, what is the primary purpose of installing a cable management arm on a sliding rail kit?

  1. It locks the server into the rack to deter unauthorized removal
  2. It routes and supports cables so they flex safely while the server slides in and out Correct answer.
  3. It delivers redundant power feeds to the server's power supplies
  4. It improves cooling by blocking rear exhaust recirculation

Explanation

A cable management arm carries the power, network, and storage cables behind a sliding server so they are supported and articulate as the chassis moves. Without it, cables can strain connectors, bind, or be pulled loose during service. It has no electrical or security function.

Question 31Server hardware installation and management

A storage array requires every drive to be reachable through two independent paths so that a single cable or controller failure does not cut off access. Which drive characteristic supports this requirement?

  1. A 4K native sector size
  2. A spindle speed of 7200 RPM
  3. A dual-port SAS interface Correct answer.
  4. Support for native command queuing

Explanation

Dual-port SAS drives expose two independent PHYs, allowing each drive to connect to two controllers or expanders for multipath redundancy. If one path fails, the host can continue I/O over the other. Queuing, rotational speed, and sector format affect performance or capacity, not path redundancy.

Question 32Server hardware installation and management

What is the primary function of ECC memory in a server?

  1. It keeps a backup copy of memory contents on the disk controller
  2. It doubles the effective bandwidth of each memory channel
  3. It allows DIMMs of different capacities to be mixed within a channel
  4. It detects and corrects single-bit memory errors Correct answer.

Explanation

ECC modules carry extra check bits that let the memory controller detect and correct single-bit errors, and detect most multi-bit errors. This protects against silent data corruption caused by transient faults. It does not increase bandwidth, enable arbitrary mixing, or write memory to disk.

Question 33Server hardware installation and management

An administrator configures a single global hot spare on an array controller that hosts three RAID 5 arrays. What happens when a drive in one of those arrays fails?

  1. All three arrays rebuild simultaneously onto the spare drive
  2. The controller automatically rebuilds the failed array onto the spare Correct answer.
  3. The spare is assigned only if the failure occurs in the first array
  4. The array remains degraded until an administrator manually installs a replacement

Explanation

A global hot spare is available to any array on the controller and is claimed automatically by the first array that loses a drive. The controller then rebuilds that array onto the spare, restoring redundancy without human intervention. One spare protects one drive failure at a time.

Question 34Server hardware installation and management

A server powers on, the fans spin, but no video appears and the chassis speaker emits a repeating beep pattern. What is the most appropriate first step?

  1. Replace the motherboard immediately
  2. Look up the beep pattern in the server vendor's documentation Correct answer.
  3. Reset the array controller configuration to defaults
  4. Reinstall the operating system from installation media

Explanation

Beep patterns are vendor-specific POST error codes that usually point to a specific subsystem such as memory or video. Decoding the pattern before acting avoids replacing parts at random. Operating system and storage configuration issues cannot occur before POST completes.

Question 35Server hardware installation and management

A PCIe x8 card is installed in a physical x16 slot on a board that supports full x16 links. What link width should the administrator expect?

  1. x16, because the slot forces its full width onto any installed card
  2. x1, because the slot shares lanes with the baseboard management controller
  3. x4, because half of the slot's lanes are reserved for the chipset
  4. x8, limited by the connector and lane count on the card itself Correct answer.

Explanation

PCIe link width is negotiated to the highest width supported by both ends, so an x8 card in an x16 slot trains at x8. The slot's extra lanes remain unused, and no performance penalty beyond the card's own limit applies. Management controllers use sideband interfaces rather than PCIe lanes.

Question 36Server hardware installation and management

A server already holds 2666 MT/s DIMMs and receives additional 3200 MT/s DIMMs in the remaining slots; the platform supports both speeds. What outcome should be expected?

  1. The platform automatically overclocks the slower modules to 3200 MT/s
  2. All memory runs at 2666 MT/s, the speed of the slowest installed modules Correct answer.
  3. The 2666 MT/s DIMMs are disabled so that only the faster modules are used
  4. Each channel runs at the speed of the modules installed in that channel, producing mixed speeds

Explanation

Memory runs at a single clock derived from the slowest populated module so that all channels remain synchronized. The faster DIMMs therefore operate below their rating, and no capacity is lost. Platforms do not selectively disable slower modules or raise their rating.

Question 37Server administration

A server has two drives configured in RAID 1. Which statement describes the fault tolerance this provides?

  1. Write performance is doubled because both drives are used.
  2. Data is striped across both drives with distributed parity.
  3. Both drives must be replaced whenever one drive fails.
  4. Either single drive can fail and the data remains accessible. Correct answer.

Explanation

RAID 1 mirrors identical data onto both drives, so the array continues to serve data if one member fails. The failed drive can be replaced and the mirror rebuilt from the surviving copy. Striping with parity describes RAID 5, and mirroring does not double write throughput.

Question 38Server administration

Which service should be configured on all servers so that log entries from different systems can be correlated accurately?

  1. LDAP
  2. SNMP
  3. DHCP
  4. NTP Correct answer.

Explanation

NTP keeps system clocks synchronized to a common time source. Consistent timestamps are required to place events from multiple servers in the correct order during troubleshooting or forensic review. DHCP, SNMP, and LDAP do not synchronize time.

Question 39Server administration

Which device supplies temporary battery power to a server during a brief utility outage and allows a graceful shutdown?

  1. Line conditioner
  2. Power distribution unit
  3. Uninterruptible power supply Correct answer.
  4. Redundant power supply

Explanation

A UPS contains batteries that carry the load when utility power drops, giving the server time to shut down cleanly or ride out a short outage. A PDU only distributes power, a redundant power supply only provides a second AC path, and a line conditioner only smooths voltage.

Question 40Server administration

A drive in a RAID 5 array fails during production hours. Which server feature allows the failed drive to be replaced without powering the system down?

  1. Redundant power supplies
  2. Memory sparing enabled in the BIOS
  3. Hot-swappable drive bays Correct answer.
  4. Error-correcting memory

Explanation

Hot-swappable bays let a drive be removed and replaced while the system stays powered and running. Combined with RAID redundancy, this restores the array with minimal downtime. Memory features and power redundancy do not affect drive replacement.

Question 41Server administration

Which document records the serial number, warranty expiration, and physical location of each server in a datacenter?

  1. Change log
  2. Asset inventory record Correct answer.
  3. Network topology diagram
  4. Service level agreement

Explanation

An asset inventory tracks identifying and lifecycle details for each device, including serial numbers, warranty dates, and rack location. Change logs record modifications, topology diagrams show connectivity, and SLAs define service commitments.

Question 42Server administration

Which statement best describes a blade server?

  1. A standalone server with its own chassis and power supplies.
  2. A server installed in a two-post rack without redundancy.
  3. A server that runs only virtual machines on shared storage.
  4. A modular server that shares chassis power and cooling. Correct answer.

Explanation

Blade servers are modular units that slide into a shared chassis, which supplies power, cooling, and often networking to all blades. This density reduces cabling and rack space compared with standalone servers.

Question 43Server administration

Which TCP port is used by Remote Desktop Protocol for remote administration of a Windows server?

  1. 22
  2. 3389 Correct answer.
  3. 161
  4. 443

Explanation

RDP listens on TCP 3389 by default, and administrators must permit that port through host and network firewalls. Port 22 is SSH, 443 is HTTPS, and 161 is SNMP.

Question 44Server administration

Which type of memory detects and corrects single-bit errors without interrupting the server?

  1. Registered memory
  2. Error-correcting code memory Correct answer.
  3. Dual-rank memory
  4. Non-parity memory

Explanation

Error-correcting code memory uses extra check bits to detect and correct single-bit errors transparently, which reduces unexpected crashes on servers. Non-parity memory has no error detection, while registered and dual-rank describe module architecture rather than correction capability.

Question 45Server administration

A network printer must always receive the same IP address from DHCP without any configuration being made on the printer itself. Which approach accomplishes this?

  1. Disable the DHCP scope on the printer's local subnet.
  2. Set the lease duration for the scope to infinite.
  3. Assign the address from an excluded address range.
  4. Create a DHCP reservation mapped to the printer's MAC address. Correct answer.

Explanation

A reservation ties a specific address to a client's MAC address, so the same IP is offered every time without touching the device. Exclusions simply prevent the server from handing out addresses, and longer leases do not guarantee the same address.

Question 46Server administration

An administrator needs reverse lookups of server IP addresses to return correct hostnames in log entries. Which DNS record type should be verified?

  1. A
  2. SRV
  3. CNAME
  4. PTR Correct answer.

Explanation

A PTR record lives in the reverse lookup zone and maps an IP address to a hostname, which is what reverse lookups return. A records perform forward mapping, SRV records locate services, and CNAME records create aliases.

Question 47Server administration

Two departments share a single physical switch but must be isolated from each other at layer 2. Which configuration achieves this?

  1. Enable port mirroring on the uplink port.
  2. Configure a separate default gateway on every port.
  3. Assign each department's ports to its own VLAN. Correct answer.
  4. Place both departments in the same IP subnet.

Explanation

VLANs create separate broadcast domains on shared switching hardware, so traffic from one department cannot reach the other at layer 2 without routing. Subnetting alone on one flat segment does not isolate broadcast traffic, and mirroring copies traffic rather than isolating it.

Question 48Server administration

Which SNMP operation does an agent use to send an unsolicited alert to a management station?

  1. Set
  2. Trap Correct answer.
  3. Get
  4. Walk

Explanation

A trap is generated by the agent without a prior request, which is how devices push urgent event notifications to the manager. Get and Walk are manager-initiated read operations, and Set is a manager-initiated write.

Question 49Server administration

A backup schedule runs a full backup on Sunday and differential backups on Monday through Friday. What is required to restore the data as it existed on Thursday?

  1. The Sunday full backup only.
  2. Thursday's differential backup only.
  3. The Sunday full backup plus Thursday's differential. Correct answer.
  4. The Sunday full backup plus every differential through Thursday.

Explanation

A differential backup captures everything changed since the last full backup, so the most recent differential already contains all prior changes. Restoring requires only the full backup and the latest differential. Applying each intermediate differential is the procedure for incremental backups.

Question 50Server administration

Which approach is best practice when deploying critical security patches to a large fleet of production servers?

  1. Pilot the patch on a small group, then deploy in stages. Correct answer.
  2. Defer all patching until the next hardware refresh cycle.
  3. Apply the patches to every server at once during peak hours.
  4. Patch only the servers that are directly internet-facing.

Explanation

Piloting on a representative subset surfaces compatibility problems before wide exposure, and staged rollout limits the blast radius if something goes wrong. Simultaneous deployment risks a fleet-wide outage, and indefinite deferral leaves known vulnerabilities unaddressed.