Linux: アーカイブと圧縮 — tar / gzip / zipとデータバックアップ

tar -czf project.tar.gz \
    --exclude=node_modules \
    --exclude=.git \
    --exclude=__pycache__ \
    project/

# Create tar archive
tar -cf archive.tar file1 file2 dir/    # Archive
tar -czf archive.tar.gz dir/            # Archive + gzip compression

> 💡 ヒント：`tar -czf`オプションの覚え方 — `c`（作成）、`z`（gzip圧縮）、`f`（ファイル名指定）。これが最も一般的な組み合わせで、結果の`.tar.gz`ファイルはLinux世界の標準配布フォーマットです。展開する時は`c`を`x`（展開）に替えるだけ：`tar -xzf`。
tar -cjf archive.tar.bz2 dir/           # Archive + bzip2 compression
tar -cJf archive.tar.xz dir/            # Archive + xz compression

# Extract
tar -xf archive.tar                     # Extract (auto-detects format)
tar -xzf archive.tar.gz                 # Extract .tar.gz
tar -xf archive.tar -C /target/dir/     # Extract to specified directory

# View contents
tar -tf archive.tar.gz                  # Preview file list inside archive

> ⚠️ 警告：tarアーカイブを展開する前に、`tar -tf`でファイルリストをプレビューしてください — 「爆弾ファイル」（例：大量の小さなファイルに展開されるもの、または`/`で始まる絶対パス）がないことを確認し、システムファイルの上書きやディスク容量の枯渇を防ぎます。

# Exclude
tar -czf backup.tar.gz --exclude="*.log" --exclude=node_modules dir/

# Compress
zip -r project.zip project/             # Recursively compress directory
zip -r project.zip project/ -x "*.git*" # Exclude .git
zip -r project.zip project/ -P password # Encrypt

# Extract
unzip project.zip                       # Extract to current directory
unzip project.zip -d /target/dir/       # Extract to specified directory
unzip -l project.zip                    # View contents without extracting

# zip advantage: natively supported on all operating systems

# Basic usage
rsync -av source/ destination/          # Local sync (recursive + preserve attributes)
rsync -av --delete source/ dest/        # Sync and delete files in dest that aren't in source

> ℹ️ 注意：`rsync --delete`は送信先を送信元と完全に一致させます — 送信先にあって送信元にないファイルは削除されます。間違えた場合（例：送信元パスの間違い）、送信先の大量データが削除される可能性があります。実行前に必ず`--dry-run`でプレビューしてください。
rsync -avz source/ user@host:/dest/     # Remote sync (compressed transfer)
rsync -avz -e "ssh -p 2222" source/ host:/dest/  # Specify SSH port

# Exclude
rsync -av --exclude="node_modules" --exclude=".git" src/ dst/

# Incremental backup (hard links)
rsync -av --link-dest=../backup-yesterday/ src/ backup-today/

# Archive the projects directory under home
tar -czf projects-backup.tar.gz ~/projects/

# View archive contents
tar -tf projects-backup.tar.gz | head -10

# Extract to /tmp
tar -xzf projects-backup.tar.gz -C /tmp/

# With verbose output
tar -czvf my-backup.tar.gz --exclude="*.log" /var/log/

# Prepare test data
mkdir -p /tmp/compress-test
cp -r /usr/share/doc/* /tmp/compress-test/ 2>/dev/null

# gzip
time tar -czf test.tar.gz -C /tmp compress-test
ls -lh test.tar.gz

# bzip2
time tar -cjf test.tar.bz2 -C /tmp compress-test
ls -lh test.tar.bz2

# xz
time tar -cJf test.tar.xz -C /tmp compress-test
ls -lh test.tar.xz

# Comparison results
# Typically: test.tar.gz is largest but fastest, test.tar.xz is smallest but slowest

# Archive a project (excluding unneeded directories)
cd ~/projects
zip -r myapp.zip myapp/ -x "myapp/node_modules/*" "myapp/.git/*" "myapp/__pycache__/*"

# View contents
unzip -l myapp.zip | head -20

# Encrypt archive
zip -r -e secret.zip confidential/    # Will prompt for password

# Local backup
rsync -av --delete ~/projects/ /backup/projects/

# Remote backup to server
rsync -avz -e "ssh -i ~/.ssh/mykey" \
    --exclude="node_modules" \
    --progress \
    ~/projects/ user@backup-server:/backups/

# Incremental backup (using hard links)
# Day 1
rsync -av /data/ /backup/2026-07-07/

# Day 2 (only stores changes)
rsync -av --link-dest=/backup/2026-07-07/ /data/ /backup/2026-07-08/

#!/bin/bash
# auto-backup.sh - Automated backup of critical directories

BACKUP_ROOT="/backup"
DATE=$(date +%Y%m%d)
RETENTION_DAYS=30

# Directories to back up
declare -A DIRS
DIRS["projects"]="$HOME/projects"
DIRS["configs"]="/etc"
DIRS["notes"]="$HOME/notes"

echo "=== Starting backup: $DATE ==="

# Create backup directory
mkdir -p "$BACKUP_ROOT/$DATE"

for name in "${!DIRS[@]}"; do
    src="${DIRS[$name]}"
    dest="$BACKUP_ROOT/$DATE/$name.tar.gz"
    
    if [ -d "$src" ]; then
        echo "Backing up $src → $dest"
        tar -czf "$dest" \
            --exclude="node_modules" \
            --exclude=".git" \
            --exclude="*.log" \
            "$src" 2>/dev/null
        echo "  ✅ $(du -h "$dest" | cut -f1)"
    else
        echo "  ⚠️ Directory does not exist: $src"
    fi
done

# Delete expired backups
echo ""
echo "Cleaning backups older than $RETENTION_DAYS days..."
find "$BACKUP_ROOT" -maxdepth 1 -type d -mtime +$RETENTION_DAYS -exec rm -rf {} \;

echo "=== Backup complete ==="

#!/bin/bash
# rsync-backup.sh - Incremental backup using rsync

SOURCE="$HOME/projects/webapp"
DEST_USER="backup"
DEST_HOST="backup.example.com"
DEST_PATH="/backups/webapp"
TIMESTAMP=$(date +%Y%m%d_%H%M%S)
LOGFILE="/var/log/backup.log"

log() {
    echo "[$(date '+%Y-%m-%d %H:%M:%S')] $1" | tee -a "$LOGFILE"
}

log "Starting backup of $SOURCE..."

# Perform remote incremental backup using rsync
rsync -avz \
    --delete \
    --partial \
    --progress \
    --exclude="node_modules" \
    --exclude=".git" \
    --exclude=".env" \
    --exclude="tmp/" \
    -e "ssh -i ~/.ssh/backup_key -p 22" \
    "$SOURCE/" \
    "${DEST_USER}@${DEST_HOST}:${DEST_PATH}/current/" >> "$LOGFILE" 2>&1

if [ $? -eq 0 ]; then
    log "✅ Backup succeeded"
    # Create timestamped snapshot
    ssh -i ~/.ssh/backup_key "${DEST_USER}@${DEST_HOST}" \
        "cp -al ${DEST_PATH}/current ${DEST_PATH}/${TIMESTAMP}"
    log "✅ Snapshot created: $TIMESTAMP"
else
    log "❌ Backup failed (exit code: $?)"
    exit 1
fi