Server Management: Server Hardware Installation and Management

CompTIA Server+ Certification Preparation (Domain 1.0) Fictional company: Globomantics (veterinary technology company)

Table of Contents

1. Introduction to Server+ Certification
2. Server Concepts
3. Storage
4. Rack Configuration
5. Server Management
6. Advanced Topics
7. Conclusion
8. Design Decisions — Summary
9. Appendix — Glossary

Module 1 — Introduction to Server+ Certification

About the Course

This course is part of the CompTIA Server+ series. It covers domain 1.0: Server Hardware Installation and Management.

Target audience:

• People preparing for the Server+ exam
• Professionals looking to move from technical support to a data center role
• Anyone seeking a better understanding of servers

Recommended prerequisites:

• Basic computer hardware knowledge (equivalent to CompTIA A+)
• 1-2 years of IT industry experience

Domain 1.0 Objectives

The wording “given a scenario” is deliberate: the exam tests your ability to solve real problems, not just memorize definitions.

The Globomantics Thread

timeline
    title Globomantics Evolution
    Start : 5 employees
          : San Jose home office
          : First tower server
    Growth : Expanded team
           : First San Jose office
           : Rack servers
    Regional expansion : ~400 HQ employees
                      : London Tokyo Mumbai offices
                      : ~100 employees per office
    Large enterprise : ~1500 employees
                    : Dedicated data center

Study Tips

• Watch videos in order (the Globomantics story builds progressively)
• Take notes — notes help find important passages later
• Build a lab for hands-on experience — practice makes permanent

Module 2 — Server Concepts

Introduction

Globomantics application software specifications:

Server roles and dominant resources:

Form Factors

flowchart LR
    A[Server need] --> B{Space available?}
    B -->|Large space, low density| C[Tower]
    B -->|Moderate space, medium density| D[Rack Mount]
    B -->|Small space, high density| E[Blade]
    C --> C1["Vertical orientation\nStandalone, expandable\nIdeal for small office\nNo special infrastructure"]
    D --> D1["Horizontal in a rack\nRails + cable arm\nBetter density\nMid-size enterprise"]
    E --> E1["Shared blade chassis\nCPU+RAM on blade\nCentralized power/cooling\nHigh density, high cost"]

Tower

• Standalone vertical server — professional grade components (tested, certified — ≠ desktop)
• Advantages: expandability, easy access, no special infrastructure
• Redundant fans, possibility of redundant power supply
• Globomantics decision: Tower placed in a well-ventilated and locked room

Rack Mount

• Horizontal in a standard 19-inch rack cabinet
• Supplied with telescopic rails (rail kit) and cable management arm (CMA)
• Weight fully supported by the rack (not on lower servers)
• Ball bearing sliders allow pulling the server without fully removing it

Blade

• Rack chassis containing vertically sliding blades
• Each blade: CPU, RAM, sometimes storage
• Chassis provides: power, cooling, network connectivity (backplane/midplane)
• Turbine fans (blowers) provide cooling — very loud
• Limited expandability — high initial cost — may require three-phase power

Comparison: Scale-Up vs Scale-Out

Hardware Compatibility List (HCL)

Created by the software vendor (not hardware).

HCL = Menu. Hardware requirements = shopping list. Buy at the intersection of both.

Benefits:

• Guarantees hardware works with the software
• Facilitates support — vendor can’t blame hardware
• Helps anticipate product lifecycle

CPU — Architecture and Components

CPU Architecture Types

Sockets, Cores and Threads

flowchart TD
    CPU["Physical CPU Processor"]
    CPU --> S0["Socket 0 Physical slot"]
    CPU --> S1["Socket 1 Physical slot"]
    S0 --> C0["Core 0"]
    S0 --> C1["Core 1"]
    S0 --> Cn["Core N"]
    C0 --> T0["Thread 0"]
    C0 --> T1["Thread 1 Hyperthreading"]

• Processor (CPU): the physical chip installed in a socket
• Socket: the physical slot on the motherboard — not backward-compatible between generations
• Core: processor within a processor — enables parallel processing
• Thread: execution unit within a core (hyperthreading = 2 threads per core)

Thermal Dissipation — TDP

• TDP (Thermal Design Power): max power fed to the socket to prevent overheating
• Baffles: channel airflow through components
• Shrouds: plastic guides directing air — never remove! (guaranteed overheating)

CPU Cache

CPU Choice: Cores vs Frequency

RAM — Types and Placement

RAM Types

DDR Evolution

Each DDR generation provides approximately 20-30% energy savings over the previous.

Different keying per generation — physically incompatible. Never force!

Registered vs Unregistered RAM

ECC — Error Correcting Code

• Detects and corrects single-bit errors on the fly
• Detects multi-bit errors
• Indispensable for mission-critical servers

Fault-Tolerant Memory

Motherboard Considerations

GPU — Graphics Processing Unit

PCI Express (PCIe) — Versions and Performance

Backward compatibility: An x1 card works in an x4 slot.

Calculation example: 40 Gbps NIC on PCIe 4.0 → 40/8 = 5 GB/s required → PCIe 4.0 x4 = 8 GB/s → x4 slot sufficient.

Server Maintenance

Hot Swap vs Cold Swap

Check three levels before hot-swap: BIOS/UEFI, OS, and hardware vendor.

BIOS and UEFI

flowchart LR
    subgraph BIOS["Legacy BIOS"]
        B1["Basic Input/Output System\nSince the 80s\nStored in NV-RAM\nLimited to 2TB MBR\n16-bit real mode"]
    end
    subgraph UEFI["Modern UEFI"]
        U1["Unified Extensible Firmware Interface\nBIOS successor\nSupports more than 2TB GPT\nSecure Boot crypto signatures\nTPM integration\n32/64-bit graphical interface"]
    end
    BIOS -->|"BIOS compatibility mode"| UEFI

Module 3 — Storage

Storage Types

flowchart LR
    A[Data access] --> B{Where is the data?}
    B -->|Via internet API/HTTP| C["Object Storage\nImmutable + Atomic\nVersioning possible\nEx: S3 Azure Blob"]
    B -->|On another PC LAN| D["File Storage\nHierarchical folders/files\nSMB/NFS\nEx: NAS file server"]
    B -->|Directly attached| E["Block Storage\nData in blocks\nFormatted by the server\nEx: internal disks SAN"]

Disk Arrays and RAID Levels

RAID Summary Table

flowchart TD
    RAID[RAID Levels]
    RAID --> R0["RAID 0 — Pure Striping\n2+ disks, no redundancy\nMaximum performance\nTotal loss if 1 disk fails"]
    RAID --> R1["RAID 1 — Mirroring\n2 disks, 1 identical copy\nFast read from 2 disks\nLoss if both fail"]
    RAID --> R5["RAID 5 — Striping + XOR Parity\n3+ disks minimum\n1 disk capacity lost\n1 failure tolerance\nRotating parity"]
    RAID --> R6["RAID 6 — Double Parity\n4+ disks\n2 disks capacity lost\n2 simultaneous failure tolerance\nReed-Solomon"]
    RAID --> R10["RAID 10 (1+0) — Recommended\n4+ disks in pairs\nMirror pairs then stripe\nHigh performance + resilience\n50% usable capacity"]

RAID 10 is always preferable to RAID 01: mirroring granularity is better.

Disk Types and Interfaces

HDD

SSD Types

Storage Interfaces

Performance increasing: SATA < SAS < NVMe — and cost increasing the same way.

Clustering and Storage Area Networks (SAN)

Clustering

flowchart TB
    subgraph "Active-Passive Cluster"
        A1["Active Node processes all requests"]
        A2["Passive Node on standby"]
        A1 <-->|"Dedicated network heartbeat"| A2
    end
    SAN["Storage Area Network Shared storage"]
    A1 & A2 --> SAN

Storage Area Network (SAN)

Network dedicated to storage, separate from the normal data network.

Front-end SAN protocols:

Module 4 — Rack Configuration

Server Organization in Racks

Rack Units and Cabinets

• 1 rack unit U = 44.45 mm height
• Standard rack width: 19 inches
• Standard size: 42U or 48U

Safety during installation:

flowchart TD
    A["1. Verify floor load\nMax ~3000 lbs for full rack"] --> B
    B["2. Install stabilizers\nAnti-tipping feet first"] --> C
    C["3. Use rack lift for heavy equipment\nBlade chassis, UPS"] --> D
    D["4. Fill from bottom up\nLow center of gravity"] --> E
    E["5. Never pull out\nMultiple units simultaneously"] --> F
    F["6. Label all cables\nBoth ends"]

Cable management best practices:

Power Supply

Basic Concepts

Single-Phase vs Three-Phase

flowchart LR
    M["Single-Phase\n1 sine wave\nResidential, small offices\n120V or 240V\n2 wires + neutral"]
    T["Three-Phase\n3 waves at 120-degree offset\nData centers, industrial\n208V between phases\n3 phase wires"]
    M --> M1["Tower servers, standard rack servers"]
    T --> T1["Blade chassis\nLarge data centers\nRequires qualified electrician"]

Power Redundancy

flowchart TB
    U1["Utility 1 Provider A"]
    U2["Utility 2 Provider B"]
    G1[Generator 1]
    G2[Generator 2]
    UPS1["UPS 1 Battery"]
    UPS2["UPS 2 Battery"]
    PDU1["PDU 1"]
    PDU2["PDU 2"]
    PSUA["Server PSU A Redundant"]
    PSUB["Server PSU B Redundant"]

    U1 --> G1 --> UPS1 --> PDU1
    U2 --> G2 --> UPS2 --> PDU2
    PDU1 & PDU2 --> PSUA
    PDU1 & PDU2 --> PSUB

• UPS: handles microcuts and switchover time to generator
• Generator: handles prolonged outages (hours, days)

Cooling Methods

Hot/Cold Aisle:

flowchart LR
    subgraph "Cold Aisle"
        AC["Cold air from raised floor or ceiling"]
    end
    subgraph "Server Row"
        S1["Server front - draws cold air"]
        S2["Server"]
    end
    subgraph "Hot Aisle"
        AH["Hot air collected then cooled by chillers"]
    end
    AC --> S1 & S2
    S1 & S2 --> AH
    AH -->|"Return to chillers"| AC

Network Cabling

Copper Twisted Pair Categories

Fiber Optic Cabling

Never replace an aqua cable with an orange one — performance incompatibility.

SFP Transceivers

Module 5 — Server Management

Local Management

Console Connections

KVM Switch

A single monitor/keyboard/mouse for multiple servers.

Crash Cart

Mobile workstation for emergency interventions and initial installation.

Remote Control

KVM over IP = network access to console = logical physical access. Secure access absolutely!

Remote Management IPMI

IPMI — Baseboard Management Controller (BMC)

“Computer within a computer” — independent of the main CPU and OS.

flowchart TB
    subgraph "Physical Server"
        IPMI["IPMI BMC\nAutonomous computer\nOwn CPU memory NIC\nWorks even if OS crashed\nWorks even if server off if powered"]
        CPU_MAIN["Main CPU + OS"]
        PWR["Server power supply"]
    end
    IPMI <-->|"Out-of-band management network"| ADMIN["Remote administrator"]
    PWR --> IPMI
    PWR --> CPU_MAIN

Always change the default password! Dell: root/calvin — publicly known via automatic scans.

Proprietary IPMI Solutions

Full license recommended (iDRAC Enterprise, iLO Advanced) — long-term savings.

Out-of-Band (OOB) Network

flowchart LR
    subgraph "Production network"
        NS[Production switch]
        S1[Server 1]
    end
    subgraph "OOB management network"
        MS["Dedicated isolated management switch"]
        I1["Server 1 IPMI port - separate IP"]
    end
    Admin["Admin via VPN to OOB"] --> MS
    MS --> I1
    S1 --> NS
    I1 -.-> S1

Module 6 — Advanced Topics

POST Process

The POST (Power-On Self-Test) runs on every power-up:

flowchart TD
    PWR["Power on"] --> FIRMWARE["Firmware loading BIOS/UEFI from NV-RAM"]
    FIRMWARE --> CPU_TEST["CPU test - basic registers and instructions"]
    CPU_TEST --> RAM_TEST["RAM test - read/write on all cells"]
    RAM_TEST --> PERIPH["Device inventory - PCIe SATA USB NICs"]
    PERIPH --> VIDEO["Video initialization - display available from here"]
    VIDEO --> IPMI_INIT["IPMI/BMC initialization - before OS, always available"]
    IPMI_INIT --> SPLASH["POST screen / Error codes - LED codes or beep codes"]
    SPLASH --> BOOT["Boot device selection - order defined in BIOS/UEFI"]
    BOOT --> BOOTLOADER["Bootloader loading - GRUB Windows Boot Manager"]
    BOOTLOADER --> OS["OS startup"]

Important POST keys on Dell PowerEdge:

NIC Teaming

Grouping multiple physical network interfaces into a single logical interface.

flowchart TB
    subgraph "Server with NIC Teaming"
        NIC1["NIC 1 Physical"]
        NIC2["NIC 2 Physical"]
        TEAM["Team Interface Logical — 1 IP"]
        NIC1 & NIC2 --> TEAM
    end
    subgraph "Switch with LACP"
        PORT1["Switch Port 1"]
        PORT2["Switch Port 2"]
        LAG["Link Aggregation Group LAG/LACP"]
        PORT1 & PORT2 --> LAG
    end
    NIC1 --> PORT1
    NIC2 --> PORT2

Teaming modes:

Windows Server PowerShell Commands:

# Create an Active-Backup team
New-NetLbfoTeam -Name "ProdTeam" `
    -TeamMembers "NIC1","NIC2" `
    -TeamingMode SwitchIndependent `
    -LoadBalancingAlgorithm HyperVPort

# Create an LACP team
New-NetLbfoTeam -Name "ProdTeam" `
    -TeamMembers "NIC1","NIC2" `
    -TeamingMode LACP `
    -LoadBalancingAlgorithm Dynamic

# Check team status
Get-NetLbfoTeam
Get-NetLbfoTeamMember -Team "ProdTeam"

iDRAC — Configuration and RACADM Commands

iDRAC offers three interfaces:

1. Web interface HTTPS to the iDRAC IP
2. RACADM local or SSH CLI
3. Redfish API standard DMTF REST

Initial iDRAC Configuration

1. Power on -> F2 -> System Setup
2. iDRAC Settings -> Network
3. Enable dedicated NIC port
4. Configure IP DHCP or Static
5. Save and exit
6. Web access: https://IP_iDRAC
   Default login: root / calvin  <- CHANGE IMMEDIATELY

Essential RACADM Commands

# Complete system info
racadm getsysinfo

# iDRAC network configuration
racadm getniccfg
racadm setniccfg -s 192.168.1.100 255.255.255.0 192.168.1.1

# Power control
racadm serveraction graceshutdown
racadm serveraction powercycle
racadm serveraction poweron

# Remote RACADM via network
racadm -r 192.168.1.100 -u root -p MyPassword serveraction poweron

# Sensor and storage monitoring
racadm getsensorinfo
racadm storage get drives
racadm raid get pdisks

# Export/Import complete iDRAC configuration
racadm get -t xml -f idrac_config.xml
racadm set -t xml -f idrac_config.xml

Redfish API — iDRAC Standard DMTF REST

# System info
curl -ks -u root:password \
  https://192.168.1.100/redfish/v1/Systems/System.Embedded.1

# Power on via Redfish
curl -ks -u root:password -X POST \
  https://192.168.1.100/redfish/v1/Systems/System.Embedded.1/Actions/ComputerSystem.Reset \
  -H "Content-Type: application/json" \
  -d '{"ResetType": "On"}'

# View SEL (System Event Log) logs
curl -ks -u root:password \
  https://192.168.1.100/redfish/v1/Managers/iDRAC.Embedded.1/Logs/Sel

iLO — HPE Integrated Lights-Out

iLO 7 Gen 12 introduces Silicon Root of Trust — verification keys embedded in silicon during manufacturing.

DAS vs NAS vs SAN — Complete Comparison

flowchart TB
    subgraph "DAS Direct Attached Storage"
        SERVER_DAS["Server HBA or SATA"]
        STORAGE_DAS["External SAS disk array"]
        SERVER_DAS <-->|"SAS/SATA direct - no network"| STORAGE_DAS
    end
    subgraph "NAS Network Attached Storage"
        NAS_DEV["NAS Device - dedicated file OS"]
        SW_NAS[LAN Switch]
        C1[Client 1]
        C2[Client 2]
        C1 & C2 <-->|"SMB/NFS file protocol"| SW_NAS
        SW_NAS <--> NAS_DEV
    end
    subgraph "SAN Storage Area Network"
        SERVER1["Server 1 HBA/CNA"]
        SERVER2["Server 2 HBA/CNA"]
        SAN_SW["SAN Switch FC/iSCSI"]
        CTRL["Storage controller"]
        DISKS["Disk arrays"]
        SERVER1 & SERVER2 <-->|"iSCSI/FC/FCoE block"| SAN_SW
        SAN_SW <--> CTRL
        CTRL <-->|"SAS/FC"| DISKS
    end

Server Maintenance Checklist

Daily (automated monitoring)

• Check IPMI/iDRAC/iLO alerts (email or monitoring)
• Monitor CPU, RAM, disk temperatures
• Check RAID status — no failed or rebuilding disk
• Disk space monitoring — 80%/90% threshold alerts
• Check system event logs

Weekly

• Review hardware error logs
• Check UPS status — battery level, self-tests
• Check backup status — success/failure
• Check available firmware updates

Monthly

• Firmware updates — BIOS, iDRAC, controllers — after validation
• Test UPS in bypass mode with backup protection in place
• Physical inspection — server LEDs, cables
• Cluster failover test if applicable

Annual

• Complete failover test — simulate server failure
• Complete power cutoff test — UPS+generator together
• Preventive UPS battery replacement if more than 3 years old
• Security audit — IPMI passwords, SSL certificates
• DRP/BCP review

Useful Monitoring Commands

# Linux — CPU Temperature
sensors | grep Core

# Linux — RAID status mdadm
cat /proc/mdstat
mdadm --detail /dev/md0

# Linux — Disk SMART
smartctl -a /dev/sda

# Linux — Disk space
df -h

# Windows PowerShell — Server status
Get-WmiObject Win32_Processor | Select Name, LoadPercentage
Get-Counter '\Memory\Available MBytes'
Get-EventLog -LogName System -Newest 20 -EntryType Error

# iDRAC RACADM — Health report
racadm getsysinfo
racadm storage get drives

# IPMI — Read sensors
ipmitool -H 192.168.1.100 -U root -P password sensor
ipmitool -H 192.168.1.100 -U root -P password sel list

Physical Server Installation Flow

flowchart TD
    A["1. Receive and verify\nCheck against purchase order\nInspect for shipping damage"] --> B
    B["2. HCL verification\nDisk firmware compatible with controller?\nCPU/RAM in OS HCL?"] --> C
    C["3. Room/rack preparation\nVerify floor load\nInstall rails/stabilizers\nPlan U location"] --> D
    D["4. Physical installation\nUse rack lift if heavy\nInstall telescopic rails\nSlide server into rack\nSecure with cage nuts"] --> E
    E["5. Power cabling\nPSU A to PDU A\nPSU B to PDU B\nTwist-lock if applicable"] --> F
    F["6. Network cabling\nNIC 1 to Switch A ToR\nNIC 2 to Switch B ToR\nIPMI port to OOB Switch\nLabel all cables"] --> G
    G["7. IPMI configuration\nF2 to iDRAC Settings\nConfigure static IP\nChange default password\nTest web access"] --> H
    H["8. POST and BIOS\nVerify error-free POST\nConfigure boot order\nEnable UEFI and Secure Boot\nSave BIOS profile"] --> I
    I["9. OS installation\nPXE boot or ISO via iDRAC\nInstall OS from template\nEnable updates"] --> J
    J["10. RAID configuration\nConfigure RAID volumes\nCheck array status\nConfigure hot spare"] --> K
    K["11. Validation and monitoring\nAdd to monitoring system\nConfigure IPMI alerts\nDocument in inventory\nFailover test if cluster"]

Module 7 — Conclusion

Course Summary

This course covered domain 1.0 of the CompTIA Server+ exam:

Server+ Exam Tips

1. Scenario-based questions — not definitions
2. Know RAID levels by heart — tolerance, usable capacity, write penalty
3. Know fiber optic color codes — orange OM1/OM2, aqua OM3/OM4, yellow OS1/OS2
4. Understand differences between iDRAC / iLO / IPMI
5. Master scale-up vs scale-out
6. PCIe calculation: version × lanes × throughput
7. Cable categories and their speeds
8. Hot swap vs cold swap with concrete examples

Design Decisions — Summary

Server Concepts

Storage

Appendix — Glossary

Search Terms

server+ · server · management · hardware · installation · systems · cybersecurity · networking · security · storage · types · cpu · commands · concepts · configuration · idrac · ipmi · network · rack · ram · san · architecture · area · cabling