Script Bash de sauvegarde de sites web

July 7, 2025 · 6 min read

Fullstack Developer

Dans la lignée de mon dernier post à propos de sauvegardes de volumes et de base de données Docker, je publie le script que j'utilise pour sauvegarder des sites web sur un serveur. J'utilise ce script depuis... 15 ans ? Je viens de le mettre à jour en y ajoutant les informations de statuts de sauvegarde.

Fichier des sites

Ce script va lire un simple fichier texte qui contient 1 ligne par site. Une ligne de ce fichier possède 3 valeurs séparées par un espace, de la forme :

nom base_de_donnees chemin/vers/racine/du/site

Si un site n'a pas de base de données, on indique nodb. Voici un exemple de fichier, appelé backup_websites_websites_vars.txt dans le script ci-dessous :

monsite.net monsite_db var/www/monsite.net/www/
unautresite.org db_autre var/www/unautresite.org/
static.net nodb var/www/static.net/

Pour info, j'ai un fichier de 25 lignes, je sauvegarde donc 25 sites sur ce serveur.

Le script

Ci-dessous le script de sauvegarde que j'utilise avec un cron qui l'exécute toutes les nuits. Le script :

Fait le dump MySQL (+ compression)
Sauvegarde les fichiers (+ compression)
Fait le ménage
Envoi un rapport détaillé par email

#!/bin/bash

#======================================================================
# backup_websites.sh - Websites Backup script
#======================================================================
# by fabien a. [f@feub.net]
# https://fabienamann.dev/
#======================================================================
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.&nbsp; See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not,
#  - write to the Free Software
#               Foundation, Inc.,
#               51 Franklin Street,
#               Fifth Floor,
#               Boston, MA  02110-1301
#               USA.
#               - See http://www.gnu.org/licenses/gpl.html
#======================================================================
# VARIABLES
#======================================================================

BKP_NAME=backup_websites
CIBLE_TAR=home/utilisateur/backup/${BKP_NAME}
CIBLE=/${CIBLE_TAR}

REF_DATE=$(date +"%Y%m%d-%H%M%S")

# DB info
MYSQL_USER=utilisateur_db
MYSQL_PW=password_db

# IMPORTANT:
# This file defines what has to be backed up
# Each line has 3 parameters separated by a space and represents a website:
# name database_name path/to/website/root/to/backup
# For example:
# monsite.net monsite_db /var/www/monsite.net/html/
# Be sure not to have any empty lines (even the last one)
SITES_FILE=/home/utilisateur/bin/backup_websites_websites_vars.txt

# Files and/or directories to exclude
# Ex. EXCLUDE_LIST="--exclude=var/www/storage/cache/ --exclude=var/www/tmp/ --exclude=var/www/badfile.tgz"
EXCLUDE_LIST=""

MAIL_DEST="utilisateur@monsite.net"

FINALUSER=utilisateur

# Status tracking
OVERALL_BACKUP_STATUS="SUCCESS"
BACKUP_RESULTS=()
TOTAL_WEBSITES=0
SUCCESSFUL_WEBSITES=0
FAILED_WEBSITES=0

#======================================================================
# EXECUTION
#======================================================================

# Make sure only root can run our script
if [[ $EUID -ne 0 ]]; then
    echo "This script must be run as root"
    exit 1
fi

logger Start ${BKP_NAME}.sh at $(date +"%D - %T")

# The script deletes all previous backups before starting a new one.
# Be sure to get them locally or somewhere else before
# or comment this line out to keep them all if you've the space
rm -f ${CIBLE}/*

echo "BACKUP STARTING - $(date +%d-%m-%Y\ %H:%M:%S)"

if [ ! -d ${CIBLE} ]; then
    echo "\"${CIBLE}\" directory doesn't exist, creating it now."
    mkdir -p ${CIBLE}/log || exit 1
fi

while read LINE; do
    NAME=$(echo $LINE | awk '{print $1}')
    DB_NAME=$(echo $LINE | awk '{print $2}')
    SITE_ROOT_TAR=$(echo $LINE | awk '{print $3}')
    SITE_ROOT=/${SITE_ROOT_TAR}

    TOTAL_WEBSITES=$((TOTAL_WEBSITES + 1))

    # Initialize status for this website
    WEBSITE_STATUS="SUCCESS"
    DB_STATUS="N/A"
    COMPRESSION_STATUS="FAILED"
    DB_SIZE=""
    ARCHIVE_SIZE=""

    echo "Backup $NAME"
    echo "-----------------------------------------------------------"

    # If there is a database to backup for that site
    if [ $DB_NAME != "nodb" ]; then
        # Clean old dump
        rm -f ${SITE_ROOT}/dump-${DB_NAME}-*

        FILEOUT=dump-${DB_NAME}-${REF_DATE}.sql
        echo " - Dump SQL file $FILEOUT :"

        # Let's dump
        if mysqldump -u $MYSQL_USER -p$MYSQL_PW $DB_NAME >$FILEOUT 2>/dev/null; then
            echo " - Compressing file"
            GZ=$FILEOUT.tar.gz

            if tar -zvcf $GZ $FILEOUT >/dev/null 2>&1; then
                echo " - Remove uncompressed file"
                rm -f $FILEOUT
                FILEOUT=$GZ
                DB_SIZE=$(du -sh ${FILEOUT} | awk '{print $1}')
                echo " - Size : ${DB_SIZE}"
                DB_STATUS="SUCCESS"
            else
                echo " /!\ tar failed. Will NOT remove uncompressed SQL file"
                DB_STATUS="FAILED"
                WEBSITE_STATUS="FAILED"
            fi
        else
            echo " /!\ MySQL dump failed. Check that server is up and check the username + password"
            DB_STATUS="FAILED"
            WEBSITE_STATUS="FAILED"
        fi

        if [ "$DB_STATUS" == "SUCCESS" ]; then
            echo " - Move the file to ${SITE_ROOT}"
            mv $FILEOUT ${SITE_ROOT}
        fi

        echo "--%<------------------------------------------------------------------"
    fi

    #======================================================================
    # COMPRESSION
    #======================================================================
    BASENAME_STAMP=${REF_DATE}-${NAME}

    echo " - Compression tgz"

    if tar -czvf ${CIBLE}/${BASENAME_STAMP}.tgz -C / ${SITE_ROOT_TAR} ${EXCLUDE_LIST} >/dev/null 2>&1; then
        ARCHIVE_SIZE=$(du -sh $CIBLE/${BASENAME_STAMP}.tgz | awk '{print $1}')
        echo " - Archive created successfully: ${ARCHIVE_SIZE}"
        COMPRESSION_STATUS="SUCCESS"
    else
        echo " /!\ Archive creation failed"
        COMPRESSION_STATUS="FAILED"
        WEBSITE_STATUS="FAILED"
    fi

    echo "-----------------------------------------------------------"
    echo " - Total size : ${ARCHIVE_SIZE}"
    echo "-----------------------------------------------------------"

    # Store results for this website
    BACKUP_RESULTS+=("$NAME|$DB_NAME|$DB_STATUS|$COMPRESSION_STATUS|$DB_SIZE|$ARCHIVE_SIZE|$WEBSITE_STATUS")

    if [ "$WEBSITE_STATUS" == "SUCCESS" ]; then
        SUCCESSFUL_WEBSITES=$((SUCCESSFUL_WEBSITES + 1))
    else
        FAILED_WEBSITES=$((FAILED_WEBSITES + 1))
        OVERALL_BACKUP_STATUS="FAILED"
    fi

done <$SITES_FILE

echo "End of the backup : $(date +%D), $(date +%T) "
echo "======================================================================"

# Change permissions
/bin/chmod -R +r ${CIBLE}
/bin/chown -R ${FINALUSER} ${CIBLE}

#======================================================================
# EMAIL CONSTRUCTION AND SENDING
#======================================================================

EMAIL_SUBJECT="[${OVERALL_BACKUP_STATUS}] Backup Websites $(hostname) - $(date +%d-%m-%Y\ %H:%M:%S) (${SUCCESSFUL_WEBSITES}/${TOTAL_WEBSITES} successful)"

EMAIL_BODY=$(
    cat <<EOF
Website Backup Report for $(hostname)
========================================

Backup Date: $(date +%d-%m-%Y\ %H:%M:%S)
Backup Directory: ${CIBLE}

Summary:
- Total Websites: ${TOTAL_WEBSITES}
- Successful: ${SUCCESSFUL_WEBSITES}
- Failed: ${FAILED_WEBSITES}
- Overall Status: ${OVERALL_BACKUP_STATUS}

Detailed Results:
========================================
EOF
)

# Adding detailed results for each website
for result in "${BACKUP_RESULTS[@]}"; do
    IFS='|' read -r name db_name db_status compression_status db_size archive_size website_status <<<"$result"

    EMAIL_BODY+="\n\nWebsite: $name"
    EMAIL_BODY+="\n- Status: $website_status"
    EMAIL_BODY+="\n- Database: $db_name"
    EMAIL_BODY+="\n- DB Backup: $db_status"
    if [ "$db_size" != "" ]; then
        EMAIL_BODY+="\n- DB Size: $db_size"
    fi
    EMAIL_BODY+="\n- Archive: $compression_status"
    if [ "$archive_size" != "" ]; then
        EMAIL_BODY+="\n- Archive Size: $archive_size"
    fi
    EMAIL_BODY+="\n$([[ "$website_status" == "SUCCESS" ]] && echo "✓" || echo "✗") $name backup completed"
done

EMAIL_BODY+="\n\nDisk Usage:"
EMAIL_BODY+="\n$(df -h ${CIBLE})"

echo -e "${EMAIL_BODY}" | /bin/mail -s "${EMAIL_SUBJECT}" ${MAIL_DEST}

if [ $? -eq 0 ]; then
    echo "- Email notification sent successfully"
else
    echo "- WARNING: Failed to send email notification"
fi

logger End ${BKP_NAME}.sh at $(date +"%D - %T")

# Exit with appropriate code
if [[ "$OVERALL_BACKUP_STATUS" == "SUCCESS" ]]; then
    exit 0
else
    exit 1
fi

Ne pas oublier de donner les droits d'exécution au fichier : chmod +x backup_websites_.sh.

Cron

Pour automatiser les sauvegardes, il suffit d'éditer le crontab de l'utilisateur : crontab -e :

# Tous les jours à 2h00
0 2 * * * /home/utilisateur/bin/backup_websites.sh

Fichier des sites​

Le script​

Cron​

Fichier des sites

Le script

Cron