Tutorial

Novas atualizações e melhorias para a Macfleet.

Aviso importante

Os exemplos de código e scripts fornecidos nestes tutoriais são apenas para fins educacionais. A Macfleet não é responsável por quaisquer problemas, danos ou vulnerabilidades de segurança que possam surgir do uso, modificação ou implementação destes exemplos. Sempre revise e teste o código em um ambiente seguro antes de usá-lo em sistemas de produção.

Managing Spotlight Indexing on macOS Devices

Spotlight is macOS's powerful system-wide search feature that helps users quickly locate files, documents, applications, emails, and more across their system. While Spotlight provides excellent search capabilities, its continuous background indexing can sometimes impact system performance, particularly on older hardware or resource-constrained environments. This comprehensive guide provides methods to manage Spotlight indexing effectively.

Understanding Spotlight and Its Impact

Spotlight uses a sophisticated indexing system that continuously scans and catalogs content on your Mac to provide instant search results. This indexing process involves:

  • File content analysis: Reading and indexing text within documents
  • Metadata extraction: Cataloging file properties, creation dates, and tags
  • Application indexing: Tracking installed applications and their data
  • Email and message indexing: Indexing Mail, Messages, and other communication apps
  • Real-time updates: Monitoring file system changes to keep the index current

Why Manage Spotlight Indexing?

There are several scenarios where managing Spotlight indexing becomes necessary:

  • Performance optimization: Reducing CPU and disk usage on resource-limited systems
  • Privacy concerns: Preventing indexing of sensitive files and directories
  • Storage management: Saving disk space used by index files
  • Troubleshooting: Resolving search issues or corrupted indexes
  • Enterprise deployment: Standardizing search behavior across managed devices
  • Development environments: Preventing indexing of build directories and temporary files

Prerequisites

Before managing Spotlight indexing, ensure you have:

  • Administrative privileges on the Mac
  • Terminal or SSH access
  • Understanding of the impact on search functionality
  • Backup of important data (recommended)

Basic Spotlight Management Commands

Understanding mdutil

The mdutil command is the primary tool for managing Spotlight indexing:

# Check indexing status
mdutil -s /

# Enable indexing for a volume
mdutil -i on /

# Disable indexing for a volume
mdutil -i off /

# Erase and rebuild index
mdutil -E /

# Get help
mdutil -h

Checking Current Indexing Status

Before making changes, check the current indexing status:

#!/bin/bash

# Check Spotlight indexing status for all volumes
echo "Spotlight Indexing Status Report"
echo "==============================="
echo "Date: $(date)"
echo ""

# Get all mounted volumes
volumes=$(df -h | grep "^/dev" | awk '{print $9}')

echo "Volume Indexing Status:"
echo "----------------------"

for volume in $volumes; do
    if [ -d "$volume" ]; then
        status=$(mdutil -s "$volume" 2>/dev/null | grep "Indexing enabled")
        echo "$volume: $status"
    fi
done

echo ""
echo "Active indexing processes:"
ps aux | grep -i mds | grep -v grep

Disabling Spotlight Indexing

Basic Disable Script

Simple script to disable Spotlight indexing on the main volume:

#!/bin/bash

# Disable Spotlight indexing on the root volume
echo "Disabling Spotlight indexing..."

if sudo mdutil -i off /; then
    echo "✓ Spotlight indexing disabled successfully"
    
    # Verify the change
    status=$(mdutil -s / | grep "Indexing enabled")
    echo "Current status: $status"
else
    echo "✗ Failed to disable Spotlight indexing"
    exit 1
fi

echo "Note: Search functionality will be limited until indexing is re-enabled"

Advanced Disable Script with Logging

More comprehensive script with logging and verification:

#!/bin/bash

# Advanced Spotlight disable script with logging
LOG_FILE="/var/log/spotlight_management.log"
TIMESTAMP=$(date '+%Y-%m-%d %H:%M:%S')

# Function to log messages
log_message() {
    echo "[$TIMESTAMP] $1" | tee -a "$LOG_FILE"
}

# Function to disable indexing for a volume
disable_indexing() {
    local volume=$1
    
    log_message "Attempting to disable indexing for volume: $volume"
    
    # Check if volume exists
    if [ ! -d "$volume" ]; then
        log_message "ERROR: Volume $volume does not exist"
        return 1
    fi
    
    # Check current status
    current_status=$(mdutil -s "$volume" 2>/dev/null)
    log_message "Current status for $volume: $current_status"
    
    # Disable indexing
    if sudo mdutil -i off "$volume"; then
        log_message "SUCCESS: Indexing disabled for $volume"
        
        # Verify the change
        new_status=$(mdutil -s "$volume" 2>/dev/null)
        log_message "New status for $volume: $new_status"
        
        return 0
    else
        log_message "ERROR: Failed to disable indexing for $volume"
        return 1
    fi
}

# Main execution
log_message "Starting Spotlight indexing disable process"

# Disable for root volume
if disable_indexing "/"; then
    log_message "Root volume indexing disabled successfully"
else
    log_message "Failed to disable root volume indexing"
    exit 1
fi

# Optional: Disable for other volumes
# Uncomment the following lines to disable for additional volumes
# disable_indexing "/Volumes/ExternalDrive"
# disable_indexing "/Users"

log_message "Spotlight indexing disable process completed"
echo "Process completed. Check log at: $LOG_FILE"

Batch Disable for Multiple Volumes

Script to disable indexing on multiple volumes:

#!/bin/bash

# Batch disable Spotlight indexing for multiple volumes
VOLUMES=(
    "/"
    "/Users"
    "/Applications"
    # Add more volumes as needed
)

echo "Batch Spotlight Indexing Disable"
echo "================================"
echo "Date: $(date)"
echo ""

successful=0
failed=0

for volume in "${VOLUMES[@]}"; do
    echo "Processing volume: $volume"
    
    if [ -d "$volume" ]; then
        if sudo mdutil -i off "$volume" 2>/dev/null; then
            echo "  ✓ Successfully disabled indexing for $volume"
            ((successful++))
        else
            echo "  ✗ Failed to disable indexing for $volume"
            ((failed++))
        fi
    else
        echo "  ⚠ Volume $volume does not exist, skipping"
    fi
done

echo ""
echo "Summary:"
echo "--------"
echo "Successfully disabled: $successful volumes"
echo "Failed: $failed volumes"

if [ $failed -gt 0 ]; then
    echo "Some operations failed. Check individual volume status manually."
    exit 1
fi

Removing Spotlight Index Files

Basic Cleanup Script

After disabling indexing, remove existing index files to free up space:

#!/bin/bash

# Remove Spotlight index files
echo "Removing Spotlight index files..."

# Remove .Spotlight-V100 directories (modern macOS)
if sudo rm -rf /.Spotlight-V100; then
    echo "✓ Removed .Spotlight-V100 directory"
else
    echo "⚠ No .Spotlight-V100 directory found or removal failed"
fi

# Remove legacy .Spotlight directories
if sudo rm -rf /.Spotlight*; then
    echo "✓ Removed legacy Spotlight directories"
else
    echo "⚠ No legacy Spotlight directories found"
fi

# Calculate space freed
echo "Spotlight index cleanup completed"

Advanced Cleanup with Size Calculation

More detailed cleanup script that shows space savings:

#!/bin/bash

# Advanced Spotlight index cleanup with size reporting
echo "Spotlight Index Cleanup Utility"
echo "==============================="
echo "Date: $(date)"
echo ""

# Function to get directory size
get_size() {
    local dir=$1
    if [ -d "$dir" ]; then
        du -sh "$dir" 2>/dev/null | cut -f1
    else
        echo "0B"
    fi
}

# Function to remove spotlight directories
cleanup_spotlight_dirs() {
    local base_path=$1
    local total_size=0
    
    echo "Cleaning up Spotlight directories in: $base_path"
    
    # Find all Spotlight directories
    spotlight_dirs=$(find "$base_path" -name ".Spotlight*" -type d 2>/dev/null)
    
    if [ -z "$spotlight_dirs" ]; then
        echo "  No Spotlight directories found"
        return 0
    fi
    
    for dir in $spotlight_dirs; do
        size=$(get_size "$dir")
        echo "  Found: $dir ($size)"
        
        if sudo rm -rf "$dir"; then
            echo "    ✓ Removed successfully"
        else
            echo "    ✗ Failed to remove"
        fi
    done
}

# Clean up common locations
cleanup_spotlight_dirs "/"
cleanup_spotlight_dirs "/Users"

# Clean up on external volumes if any
for volume in /Volumes/*; do
    if [ -d "$volume" ]; then
        echo ""
        cleanup_spotlight_dirs "$volume"
    fi
done

echo ""
echo "Spotlight index cleanup completed"
echo "Run 'df -h' to see updated disk usage"

Enabling Spotlight Indexing

Basic Enable Script

Simple script to re-enable Spotlight indexing:

#!/bin/bash

# Enable Spotlight indexing on the root volume
echo "Enabling Spotlight indexing..."

if sudo mdutil -i on /; then
    echo "✓ Spotlight indexing enabled successfully"
    
    # Verify the change
    status=$(mdutil -s / | grep "Indexing enabled")
    echo "Current status: $status"
    
    echo ""
    echo "Note: Indexing will begin automatically and may take some time to complete"
    echo "You can monitor progress using Spotlight search"
else
    echo "✗ Failed to enable Spotlight indexing"
    exit 1
fi

Advanced Enable with Progress Monitoring

Script that enables indexing and provides progress monitoring:

#!/bin/bash

# Advanced Spotlight enable script with progress monitoring
LOG_FILE="/var/log/spotlight_management.log"
TIMESTAMP=$(date '+%Y-%m-%d %H:%M:%S')

# Function to log messages
log_message() {
    echo "[$TIMESTAMP] $1" | tee -a "$LOG_FILE"
}

# Function to monitor indexing progress
monitor_indexing() {
    local volume=$1
    local timeout=60  # Monitor for 60 seconds
    local elapsed=0
    
    echo "Monitoring indexing progress for $volume..."
    
    while [ $elapsed -lt $timeout ]; do
        # Check if indexing is in progress
        if ps aux | grep -v grep | grep -q "mds_stores\|mdworker"; then
            echo "  Indexing in progress... (${elapsed}s elapsed)"
        else
            echo "  Indexing may have completed or paused"
            break
        fi
        
        sleep 5
        elapsed=$((elapsed + 5))
    done
    
    if [ $elapsed -ge $timeout ]; then
        echo "  Monitoring timeout reached. Indexing may still be in progress."
    fi
}

# Function to enable indexing for a volume
enable_indexing() {
    local volume=$1
    
    log_message "Attempting to enable indexing for volume: $volume"
    
    # Check if volume exists
    if [ ! -d "$volume" ]; then
        log_message "ERROR: Volume $volume does not exist"
        return 1
    fi
    
    # Check current status
    current_status=$(mdutil -s "$volume" 2>/dev/null)
    log_message "Current status for $volume: $current_status"
    
    # Enable indexing
    if sudo mdutil -i on "$volume"; then
        log_message "SUCCESS: Indexing enabled for $volume"
        
        # Verify the change
        new_status=$(mdutil -s "$volume" 2>/dev/null)
        log_message "New status for $volume: $new_status"
        
        # Monitor initial progress
        monitor_indexing "$volume"
        
        return 0
    else
        log_message "ERROR: Failed to enable indexing for $volume"
        return 1
    fi
}

# Main execution
log_message "Starting Spotlight indexing enable process"

# Enable for root volume
if enable_indexing "/"; then
    log_message "Root volume indexing enabled successfully"
else
    log_message "Failed to enable root volume indexing"
    exit 1
fi

log_message "Spotlight indexing enable process completed"
echo "Process completed. Check log at: $LOG_FILE"

Rebuilding Spotlight Index

Basic Rebuild Script

Script to completely rebuild the Spotlight index:

#!/bin/bash

# Rebuild Spotlight index
echo "Rebuilding Spotlight index..."
echo "Warning: This will erase the current index and rebuild from scratch"
echo ""

read -p "Continue? (y/N): " confirm
if [[ ! $confirm =~ ^[Yy]$ ]]; then
    echo "Operation cancelled"
    exit 0
fi

echo "Starting index rebuild..."

if sudo mdutil -E /; then
    echo "✓ Index rebuild initiated successfully"
    
    # Show status
    status=$(mdutil -s /)
    echo "Current status: $status"
    
    echo ""
    echo "The index rebuild process has started and will continue in the background"
    echo "This may take several hours depending on the amount of data"
    echo "You can monitor progress using Activity Monitor (look for mds_stores and mdworker processes)"
else
    echo "✗ Failed to initiate index rebuild"
    exit 1
fi

Advanced Rebuild with Selective Volumes

Script to rebuild indexes for specific volumes:

#!/bin/bash

# Advanced Spotlight index rebuild script
VOLUMES=(
    "/"
    "/Users"
    # Add specific volumes or use discovery
)

AUTO_DISCOVER=${1:-false}

echo "Spotlight Index Rebuild Utility"
echo "==============================="
echo "Date: $(date)"
echo ""

# Function to discover mounted volumes
discover_volumes() {
    echo "Discovering mounted volumes..."
    
    # Get all mounted HFS+/APFS volumes
    local discovered_volumes=()
    while IFS= read -r line; do
        volume=$(echo "$line" | awk '{print $9}')
        if [ -d "$volume" ] && [ "$volume" != "/dev" ]; then
            discovered_volumes+=("$volume")
        fi
    done < <(df -h | grep "^/dev")
    
    echo "Discovered volumes:"
    for vol in "${discovered_volumes[@]}"; do
        echo "  - $vol"
    done
    
    VOLUMES=("${discovered_volumes[@]}")
}

# Function to rebuild index for a volume
rebuild_index() {
    local volume=$1
    
    echo "Rebuilding index for: $volume"
    
    # Check if volume exists and is mounted
    if [ ! -d "$volume" ]; then
        echo "  ✗ Volume $volume does not exist or is not mounted"
        return 1
    fi
    
    # Show current status
    current_status=$(mdutil -s "$volume" 2>/dev/null)
    echo "  Current status: $current_status"
    
    # Rebuild index
    if sudo mdutil -E "$volume"; then
        echo "  ✓ Index rebuild initiated for $volume"
        
        # Show new status
        new_status=$(mdutil -s "$volume" 2>/dev/null)
        echo "  New status: $new_status"
        
        return 0
    else
        echo "  ✗ Failed to rebuild index for $volume"
        return 1
    fi
}

# Auto-discover volumes if requested
if [ "$AUTO_DISCOVER" = "true" ]; then
    discover_volumes
fi

echo "Volumes to rebuild:"
for volume in "${VOLUMES[@]}"; do
    echo "  - $volume"
done

echo ""
read -p "Proceed with rebuilding indexes? (y/N): " confirm
if [[ ! $confirm =~ ^[Yy]$ ]]; then
    echo "Operation cancelled"
    exit 0
fi

echo ""
echo "Starting index rebuild process..."
echo ""

successful=0
failed=0

for volume in "${VOLUMES[@]}"; do
    if rebuild_index "$volume"; then
        ((successful++))
    else
        ((failed++))
    fi
    echo ""
done

echo "Rebuild Summary:"
echo "---------------"
echo "Successfully initiated: $successful volumes"
echo "Failed: $failed volumes"
echo ""
echo "Note: Index rebuilding will continue in the background"
echo "Monitor progress with: ps aux | grep mds"

Advanced Spotlight Management

Selective Directory Exclusion

Script to exclude specific directories from indexing:

#!/bin/bash

# Exclude specific directories from Spotlight indexing
EXCLUDE_DIRS=(
    "/tmp"
    "/var/tmp"
    "/private/tmp"
    "/Users/Shared/Build"
    # Add more directories as needed
)

echo "Spotlight Directory Exclusion Manager"
echo "===================================="
echo ""

# Function to add directory to exclusion list
exclude_directory() {
    local dir_path=$1
    
    echo "Excluding directory: $dir_path"
    
    # Check if directory exists
    if [ ! -d "$dir_path" ]; then
        echo "  ⚠ Directory does not exist: $dir_path"
        return 1
    fi
    
    # Add to Spotlight privacy list using defaults
    if sudo defaults write /System/Library/CoreServices/Search.bundle/Contents/Resources/SpotlightExclusions.plist Exclusions -array-add "$dir_path"; then
        echo "  ✓ Added to exclusion list"
        return 0
    else
        echo "  ✗ Failed to add to exclusion list"
        return 1
    fi
}

# Function to show current exclusions
show_exclusions() {
    echo "Current Spotlight exclusions:"
    echo "----------------------------"
    
    # Read from system preferences
    if defaults read /System/Library/CoreServices/Search.bundle/Contents/Resources/SpotlightExclusions.plist Exclusions 2>/dev/null; then
        echo ""
    else
        echo "No exclusions found or unable to read exclusions list"
    fi
}

# Show current exclusions
show_exclusions

echo ""
echo "Adding new exclusions..."

for dir in "${EXCLUDE_DIRS[@]}"; do
    exclude_directory "$dir"
done

echo ""
echo "Exclusion process completed"
echo "Restart Spotlight for changes to take effect:"
echo "sudo killall mds && sudo mdutil -E /"

Performance Monitoring Script

Script to monitor Spotlight's impact on system performance:

#!/bin/bash

# Spotlight Performance Monitor
DURATION=${1:-60}  # Monitor for 60 seconds by default
INTERVAL=5

echo "Spotlight Performance Monitor"
echo "============================"
echo "Monitoring duration: ${DURATION} seconds"
echo "Sample interval: ${INTERVAL} seconds"
echo ""

# Function to get Spotlight process stats
get_spotlight_stats() {
    local timestamp=$(date '+%H:%M:%S')
    
    # Get mds processes
    mds_stats=$(ps aux | grep -E "mds|mdworker" | grep -v grep | awk '{
        cpu += $3; 
        mem += $4; 
        count++
    } END {
        printf "Processes: %d, CPU: %.1f%%, Memory: %.1f%%", count, cpu, mem
    }')
    
    # Get disk activity (if available)
    disk_activity=$(iostat -c 1 2>/dev/null | tail -1 | awk '{print "Disk: " $4 "% utilization"}' || echo "Disk: N/A")
    
    echo "[$timestamp] $mds_stats, $disk_activity"
}

# Header
echo "Time     | Spotlight Stats"
echo "---------|------------------------------------------------"

# Monitor loop
elapsed=0
while [ $elapsed -lt $DURATION ]; do
    get_spotlight_stats
    sleep $INTERVAL
    elapsed=$((elapsed + INTERVAL))
done

echo ""
echo "Monitoring completed"

# Show summary
echo ""
echo "Final Process Summary:"
echo "---------------------"
ps aux | grep -E "mds|mdworker" | grep -v grep | head -10

Automation and Scheduling

Automated Maintenance Script

Script for scheduled Spotlight maintenance:

#!/bin/bash

# Automated Spotlight maintenance script
# Can be run via cron or launchd

CONFIG_FILE="/etc/spotlight_maintenance.conf"
LOG_FILE="/var/log/spotlight_maintenance.log"

# Default configuration
REBUILD_INTERVAL=30  # days
CLEANUP_TEMP=true
MONITOR_PERFORMANCE=true

# Function to log with timestamp
log_message() {
    echo "$(date '+%Y-%m-%d %H:%M:%S') - $1" | tee -a "$LOG_FILE"
}

# Load configuration if exists
if [ -f "$CONFIG_FILE" ]; then
    source "$CONFIG_FILE"
    log_message "Configuration loaded from $CONFIG_FILE"
fi

# Function to check if rebuild is needed
needs_rebuild() {
    local index_age_file="/var/db/.spotlight_last_rebuild"
    
    if [ ! -f "$index_age_file" ]; then
        return 0  # No record, assume rebuild needed
    fi
    
    local last_rebuild=$(cat "$index_age_file")
    local current_time=$(date +%s)
    local age_days=$(( (current_time - last_rebuild) / 86400 ))
    
    if [ $age_days -gt $REBUILD_INTERVAL ]; then
        log_message "Index age: $age_days days, rebuild needed"
        return 0
    else
        log_message "Index age: $age_days days, rebuild not needed"
        return 1
    fi
}

# Function to perform rebuild
perform_rebuild() {
    log_message "Starting scheduled index rebuild"
    
    if sudo mdutil -E /; then
        log_message "Index rebuild initiated successfully"
        echo "$(date +%s)" | sudo tee /var/db/.spotlight_last_rebuild > /dev/null
        return 0
    else
        log_message "Index rebuild failed"
        return 1
    fi
}

# Function to cleanup temporary files
cleanup_temp_files() {
    if [ "$CLEANUP_TEMP" = true ]; then
        log_message "Cleaning up temporary Spotlight files"
        
        # Remove temporary index files
        sudo find /tmp -name ".Spotlight*" -exec rm -rf {} \; 2>/dev/null
        sudo find /var/tmp -name ".Spotlight*" -exec rm -rf {} \; 2>/dev/null
        
        log_message "Temporary file cleanup completed"
    fi
}

# Function to check performance
check_performance() {
    if [ "$MONITOR_PERFORMANCE" = true ]; then
        local mds_cpu=$(ps aux | grep mds | grep -v grep | awk '{sum += $3} END {print sum}')
        
        if [ -n "$mds_cpu" ] && [ "$mds_cpu" -gt 50 ]; then
            log_message "WARNING: High Spotlight CPU usage detected: ${mds_cpu}%"
        else
            log_message "Spotlight performance normal: ${mds_cpu:-0}% CPU"
        fi
    fi
}

# Main maintenance routine
log_message "Starting automated Spotlight maintenance"

check_performance
cleanup_temp_files

if needs_rebuild; then
    perform_rebuild
fi

log_message "Automated Spotlight maintenance completed"

Best Practices and Recommendations

1. Performance Considerations

  • Monitor system resources before disabling indexing
  • Consider partial exclusions instead of complete disable
  • Schedule maintenance during off-hours
  • Test changes on non-production systems first

2. Security and Privacy

  • Exclude sensitive directories from indexing
  • Regularly review and update exclusion lists
  • Consider the security implications of search functionality
  • Document all changes for compliance purposes

3. Enterprise Management

  • Standardize Spotlight policies across devices
  • Use configuration management tools for deployment
  • Implement monitoring and alerting for index issues
  • Maintain documentation of all customizations

4. Troubleshooting Guidelines

  • Always check system logs when issues occur
  • Test individual commands before running complex scripts
  • Keep backups of important data before major changes
  • Document the steps taken for future reference

Troubleshooting Common Issues

Index Corruption

#!/bin/bash

# Fix corrupted Spotlight index
echo "Fixing corrupted Spotlight index..."

# Stop indexing
sudo mdutil -i off /

# Remove corrupted index
sudo rm -rf /.Spotlight-V100

# Re-enable and rebuild
sudo mdutil -i on /
sudo mdutil -E /

echo "Index repair initiated"

High CPU Usage

#!/bin/bash

# Address high Spotlight CPU usage
echo "Addressing high Spotlight CPU usage..."

# Check current processes
echo "Current Spotlight processes:"
ps aux | grep -E "mds|mdworker" | grep -v grep

# Temporarily throttle indexing
sudo launchctl unload -w /System/Library/LaunchDaemons/com.apple.metadata.mds.plist
sleep 5
sudo launchctl load -w /System/Library/LaunchDaemons/com.apple.metadata.mds.plist

echo "Spotlight services restarted"

Conclusion

Effective Spotlight management is crucial for maintaining optimal macOS performance while preserving search functionality where needed. The scripts and techniques provided in this guide offer comprehensive solutions for various Spotlight management scenarios.

Key takeaways:

  • Understand the impact of indexing on system performance
  • Use appropriate scripts for your specific use case
  • Implement proper monitoring and maintenance procedures
  • Always test changes in a controlled environment
  • Document all modifications for future reference

Remember that disabling Spotlight indexing will significantly impact search functionality, so carefully consider the trade-offs before implementing these changes across your Mac fleet.

Tutorial

Novas atualizações e melhorias para a Macfleet.

Configurando um Runner do GitHub Actions em um Mac Mini (Apple Silicon)

Runner do GitHub Actions

GitHub Actions é uma plataforma poderosa de CI/CD que permite automatizar seus fluxos de trabalho de desenvolvimento de software. Embora o GitHub ofereça runners hospedados, runners auto-hospedados fornecem maior controle e personalização para sua configuração de CI/CD. Este tutorial o guia através da configuração e conexão de um runner auto-hospedado em um Mac mini para executar pipelines do macOS.

Pré-requisitos

Antes de começar, certifique-se de ter:

  • Um Mac mini (registre-se no Macfleet)
  • Um repositório GitHub com direitos de administrador
  • Um gerenciador de pacotes instalado (preferencialmente Homebrew)
  • Git instalado em seu sistema

Passo 1: Criar uma Conta de Usuário Dedicada

Primeiro, crie uma conta de usuário dedicada para o runner do GitHub Actions:

# Criar a conta de usuário 'gh-runner'
sudo dscl . -create /Users/gh-runner
sudo dscl . -create /Users/gh-runner UserShell /bin/bash
sudo dscl . -create /Users/gh-runner RealName "GitHub runner"
sudo dscl . -create /Users/gh-runner UniqueID "1001"
sudo dscl . -create /Users/gh-runner PrimaryGroupID 20
sudo dscl . -create /Users/gh-runner NFSHomeDirectory /Users/gh-runner

# Definir a senha para o usuário
sudo dscl . -passwd /Users/gh-runner sua_senha

# Adicionar 'gh-runner' ao grupo 'admin'
sudo dscl . -append /Groups/admin GroupMembership gh-runner

Mude para a nova conta de usuário:

su gh-runner

Passo 2: Instalar Software Necessário

Instale Git e Rosetta 2 (se estiver usando Apple Silicon):

# Instalar Git se ainda não estiver instalado
brew install git

# Instalar Rosetta 2 para Macs Apple Silicon
softwareupdate --install-rosetta

Passo 3: Configurar o Runner do GitHub Actions

  1. Vá para seu repositório GitHub
  2. Navegue para Configurações > Actions > Runners

Runner do GitHub Actions

  1. Clique em "New self-hosted runner" (https://github.com/<username>/<repository>/settings/actions/runners/new)
  2. Selecione macOS como imagem do runner e ARM64 como arquitetura
  3. Siga os comandos fornecidos para baixar e configurar o runner

Runner do GitHub Actions

Crie um arquivo .env no diretório _work do runner:

# arquivo _work/.env
ImageOS=macos15
XCODE_15_DEVELOPER_DIR=/Applications/Xcode.app/Contents/Developer
  1. Execute o script run.sh em seu diretório do runner para completar a configuração.
  2. Verifique se o runner está ativo e ouvindo por trabalhos no terminal e verifique as configurações do repositório GitHub para a associação do runner e status Idle.

Runner do GitHub Actions

Passo 4: Configurar Sudoers (Opcional)

Se suas ações requerem privilégios de root, configure o arquivo sudoers:

sudo visudo

Adicione a seguinte linha:

gh-runner ALL=(ALL) NOPASSWD: ALL

Passo 5: Usar o Runner em Fluxos de Trabalho

Configure seu fluxo de trabalho do GitHub Actions para usar o runner auto-hospedado:

name: Fluxo de trabalho de exemplo

on:
  workflow_dispatch:

jobs:
  build:
    runs-on: [self-hosted, macOS, ARM64]
    steps:
      - name: Instalar NodeJS
        run: brew install node

O runner está autenticado em seu repositório e rotulado com self-hosted, macOS, e ARM64. Use-o em seus fluxos de trabalho especificando estes rótulos no campo runs-on:

runs-on: [self-hosted, macOS, ARM64]

Melhores Práticas

  • Mantenha seu software do runner atualizado
  • Monitore regularmente os logs do runner para problemas
  • Use rótulos específicos para diferentes tipos de runners
  • Implemente medidas de segurança adequadas
  • Considere usar múltiplos runners para balanceamento de carga

Solução de Problemas

Problemas comuns e soluções:

  1. Runner não conectando:

    • Verifique conectividade de rede
    • Verifique validade do token GitHub
    • Certifique-se de permissões adequadas
  2. Falhas de build:

    • Verifique instalação do Xcode
    • Verifique dependências necessárias
    • Revise logs do fluxo de trabalho
  3. Problemas de permissão:

    • Verifique permissões do usuário
    • Verifique configuração sudoers
    • Revise permissões do sistema de arquivos

Conclusão

Agora você tem um runner auto-hospedado do GitHub Actions configurado em seu Mac mini. Esta configuração fornece mais controle sobre seu ambiente CI/CD e permite executar fluxos de trabalho específicos do macOS de forma eficiente.

Lembre-se de manter regularmente seu runner e mantê-lo atualizado com os patches de segurança e versões de software mais recentes.

Aplicativo Nativo

Aplicativo nativo do Macfleet

Guia de Instalação do Macfleet

Macfleet é uma solução poderosa de gerenciamento de frota projetada especificamente para ambientes Mac Mini hospedados na nuvem. Como provedor de hospedagem na nuvem Mac Mini, você pode usar o Macfleet para monitorar, gerenciar e otimizar toda sua frota de instâncias Mac virtualizadas.

Este guia de instalação o conduzirá através da configuração do monitoramento do Macfleet em sistemas macOS, Windows e Linux para garantir supervisão abrangente de sua infraestrutura na nuvem.

🍎 macOS

  • Baixe o arquivo .dmg para Mac aqui
  • Clique duas vezes no arquivo .dmg baixado
  • Arraste o aplicativo Macfleet para a pasta Aplicativos
  • Ejete o arquivo .dmg
  • Abra Preferências do Sistema > Segurança e Privacidade
    • Aba Privacidade > Acessibilidade
    • Marque Macfleet para permitir monitoramento
  • Inicie o Macfleet a partir de Aplicativos
  • O rastreamento inicia automaticamente

🪟 Windows

  • Baixe o arquivo .exe para Windows aqui
  • Clique com o botão direito no arquivo .exe > "Executar como administrador"
  • Siga o assistente de instalação
  • Aceite os termos e condições
  • Permita no Windows Defender se solicitado
  • Conceda permissões de monitoramento de aplicativo
  • Inicie o Macfleet a partir do Menu Iniciar
  • O aplicativo começa o rastreamento automaticamente

🐧 Linux

  • Baixe o pacote .deb (Ubuntu/Debian) ou .rpm (CentOS/RHEL) aqui
  • Instale usando seu gerenciador de pacotes
    • Ubuntu/Debian: sudo dpkg -i Macfleet-linux.deb
    • CentOS/RHEL: sudo rpm -ivh Macfleet-linux.rpm
  • Permita permissões de acesso X11 se solicitado
  • Adicione o usuário aos grupos apropriados se necessário
  • Inicie o Macfleet a partir do menu Aplicativos
  • O aplicativo começa o rastreamento automaticamente

Nota: Após a instalação em todos os sistemas, faça login com suas credenciais do Macfleet para sincronizar dados com seu painel de controle.