备份与恢复 #

备份原因：
├── 硬件故障
├── 软件错误
├── 人为误操作
├── 数据损坏
├── 安全攻击
└── 合规要求

位置：
├── Linux: $NEO4J_HOME/bin/neo4j-admin
├── Windows: %NEO4J_HOME%\bin\neo4j-admin.bat
└── Docker: docker exec <container> neo4j-admin

neo4j-admin backup --help
neo4j-admin restore --help
neo4j-admin database --help

neo4j stop

cp -r $NEO4J_HOME/data /backup/neo4j_data_$(date +%Y%m%d)

neo4j-admin database backup neo4j --to-path=/backup

neo4j start

server.backup.enabled=true
server.backup.listen_address=0.0.0.0:6362

neo4j-admin database backup neo4j --from=localhost:6362 --to-path=/backup

neo4j-admin database backup neo4j,system --from=localhost:6362 --to-path=/backup

neo4j-admin database backup neo4j --from=localhost:6362 --to-path=/backup --fallback-to-full=false

neo4j stop

neo4j-admin database restore neo4j --from-path=/backup/neo4j-2024-01-01

neo4j-admin database restore neo4j --from-path=/backup/neo4j-2024-01-01 --database=neo4j_restored

neo4j start

MATCH (n)
RETURN count(n) AS node_count

docker exec neo4j neo4j-admin database backup neo4j --to-path=/backups
docker cp neo4j:/backups ./local_backup

docker cp ./local_backup neo4j:/backups
docker exec neo4j neo4j-admin database restore neo4j --from-path=/backups/neo4j-2024-01-01
docker restart neo4j

neo4j-admin database dump neo4j --to-path=/backup

neo4j-admin database load neo4j --from-path=/backup/neo4j.dump

neo4j-admin database dump mydb --to-path=/backup/mydb.dump

neo4j-admin database load mydb_restored --from-path=/backup/mydb.dump

CALL apoc.export.cypher.all('/backup/nodes.cypher', {format: 'cypher-shell'})

MATCH (p:Person)
WITH collect(p) AS persons
CALL apoc.export.cypher.data(persons, [], '/backup/persons.cypher', {})
YIELD file, nodes, relationships
RETURN file, nodes, relationships

CALL apoc.export.json.all('/backup/data.json', {})

MATCH (p:Person)
RETURN p.name, p.age

建议：
├── 每日增量备份
├── 每周全量备份
├── 每月归档备份
├── 异地备份存储
└── 定期测试恢复

#!/bin/bash

BACKUP_DIR="/backup"
DATE=$(date +%Y%m%d_%H%M%S)
NEO4J_HOME="/opt/neo4j"

$NEO4J_HOME/bin/neo4j-admin database backup neo4j --from=localhost:6362 --to-path=$BACKUP_DIR/$DATE

find $BACKUP_DIR -type d -mtime +30 -exec rm -rf {} \;

echo "Backup completed: $BACKUP_DIR/$DATE"

0 2 * * * /opt/scripts/neo4j_backup.sh

流程：
├── 评估数据丢失范围
├── 选择合适的备份点
├── 停止Neo4j服务
├── 执行恢复操作
├── 验证数据完整性
├── 重启服务
└── 监控运行状态

LOAD CSV WITH HEADERS FROM 'file:///recovery_data.csv' AS row
MERGE (p:Person {id: row.id})
SET p += {name: row.name, age: toInteger(row.age)}

neo4j-admin database restore neo4j --from-path=/backup/neo4j --transaction-log-recovery

建议：
├── 使用专用备份存储
├── 异地备份存储
├── 加密敏感数据
├── 保留多个备份版本
└── 定期清理过期备份

neo4j-admin database restore neo4j_test --from-path=/backup/neo4j-2024-01-01 --database=neo4j_test

监控项：
├── 备份是否成功
├── 备份文件大小
├── 备份时间
├── 存储空间使用
└── 恢复测试结果

#!/bin/bash

NEO4J_HOME="/opt/neo4j"
BACKUP_DIR="/backup/neo4j"
DATE=$(date +%Y%m%d_%H%M%S)
LOG_FILE="/var/log/neo4j_backup.log"

log() {
    echo "$(date '+%Y-%m-%d %H:%M:%S') - $1" >> $LOG_FILE
}

log "Starting backup..."

mkdir -p $BACKUP_DIR/$DATE

$NEO4J_HOME/bin/neo4j-admin database backup neo4j --from=localhost:6362 --to-path=$BACKUP_DIR/$DATE

if [ $? -eq 0 ]; then
    log "Backup completed successfully"
else
    log "Backup failed"
    exit 1
fi

find $BACKUP_DIR -type d -mtime +7 -exec rm -rf {} \;

log "Old backups cleaned"

步骤：
1. 评估灾难情况
2. 准备恢复环境
3. 从异地获取备份
4. 执行恢复操作
5. 验证数据完整性
6. 切换服务
7. 监控运行状态