# 创建快照

# 1. 创建全量快照
# nodetool snapshot

Requested creating snapshot(s) for: all keyspaces
Snapshot directory: 1705312800000
Snapshot name: 1705312800000

# 输出说明
– 快照目录: 时间戳命名
– 快照名称: 时间戳
– 范围: 所有Keyspace

# 2. 创建指定Keyspace快照
# nodetool snapshot -kt fgedudb

Requested creating snapshot(s) for: [fgedudb]
Snapshot directory: 1705312800000
Snapshot name: 1705312800000

# 3. 创建指定表快照
# nodetool snapshot -kt fgedudb.fgedu_users

Requested creating snapshot(s) for: [fgedudb.fgedu_users]
Snapshot directory: 1705312800000
Snapshot name: 1705312800000

# 4. 创建命名快照
# nodetool snapshot -t backup_20240115

Requested creating snapshot(s) for: all keyspaces
Snapshot directory: backup_20240115
Snapshot name: backup_20240115

# 5. 查看快照列表
# nodetool listsnapshots

Snapshot name Keyspace Table Size Created on
1705312800000 fgedudb fgedu_users 100 GB 2024-01-15T10:00:00Z
backup_20240115 fgedudb fgedu_orders 50 GB 2024-01-15T10:30:00Z

Total snapshots: 2

# 6. 查看快照详细信息
# ls -la /cassandra/fgdata/fgedudb/fgedu_users/snapshots/

total 0
drwxr-xr-x 2 cassandra cassandra 100 Jan 15 10:00 1705312800000
drwxr-xr-x 2 cassandra cassandra 100 Jan 15 10:30 backup_20240115

# 7. 查看快照文件
# ls -la /cassandra/fgdata/fgedudb/fgedu_users/snapshots/backup_20240115/

total 100000
-rw-r–r– 2 cassandra cassandra 100000 Jan 15 10:00 na-1-big-Data.db
-rw-r–r– 2 cassandra cassandra 1000 Jan 15 10:00 na-1-big-Index.db
-rw-r–r– 2 cassandra cassandra 100 Jan 15 10:00 na-1-big-Filter.db
-rw-r–r– 2 cassandra cassandra 100 Jan 15 10:00 na-1-big-Summary.db
-rw-r–r– 2 cassandra cassandra 100 Jan 15 10:00 na-1-big-Statistics.db

# 快照管理

# 1. 查看快照信息
# nodetool listsnapshots

Snapshot name Keyspace Table Size Created on
1705312800000 fgedudb fgedu_users 100 GB 2024-01-15T10:00:00Z
backup_20240115 fgedudb fgedu_orders 50 GB 2024-01-15T10:30:00Z

# 2. 删除指定快照
# nodetool clearsnapshot -t backup_20240115

Snapshot backup_20240115 cleared

# 3. 删除所有快照
# nodetool clearsnapshot

All snapshots cleared

# 4. 快照归档
# 复制快照到备份目录
# cp -r /cassandra/fgdata/fgedudb/fgedu_users/snapshots/backup_20240115 /backup/local/fgedudb_fgedu_users_20240115/

# 5. 快照压缩
# tar -czf /backup/local/fgedudb_fgedu_users_20240115.tar.gz -C /cassandra/fgdata/fgedudb/fgedu_users/snapshots backup_20240115

# 6. 快照传输
# scp /backup/local/fgedudb_fgedu_users_20240115.tar.gz backup-server:/backup/remote/

# 7. 快照管理脚本
#!/bin/bash
# snapshot_management.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

BACKUP_DIR=”/backup/local”
SNAPSHOT_NAME=”backup_$(date +%Y%m%d)”
LOG_FILE=”/cassandra/logs/backup/snapshot_$(date +%Y%m%d).log”

echo “=== 快照管理开始 ===” | tee -a ${LOG_FILE}
echo “时间: $(date)” | tee -a ${LOG_FILE}

# 创建快照
echo “创建快照: ${SNAPSHOT_NAME}” | tee -a ${LOG_FILE}
nodetool snapshot -t ${SNAPSHOT_NAME} 2>&1 | tee -a ${LOG_FILE}

# 等待快照完成
sleep 10

# 归档快照
echo “归档快照…” | tee -a ${LOG_FILE}
for keyspace in $(nodetool listsnapshots | grep ${SNAPSHOT_NAME} | awk ‘{print $2}’ | sort -u); do
for table in $(nodetool listsnapshots | grep ${SNAPSHOT_NAME} | grep ${keyspace} | awk ‘{print $3}’); do
SRC_DIR=”/cassandra/fgdata/${keyspace}/${table}/snapshots/${SNAPSHOT_NAME}”
DST_DIR=”${BACKUP_DIR}/${keyspace}_${table}_${SNAPSHOT_NAME}”

if [ -d “${SRC_DIR}” ]; then
echo “归档: ${keyspace}.${table}” | tee -a ${LOG_FILE}
mkdir -p ${DST_DIR}
cp -r ${SRC_DIR}/* ${DST_DIR}/
fi
done
done

# 压缩快照
echo “压缩快照…” | tee -a ${LOG_FILE}
cd ${BACKUP_DIR}
tar -czf ${SNAPSHOT_NAME}.tar.gz *_${SNAPSHOT_NAME} 2>&1 | tee -a ${LOG_FILE}

# 清理临时文件
rm -rf *_${SNAPSHOT_NAME}

# 清理旧快照
echo “清理旧快照…” | tee -a ${LOG_FILE}
nodetool clearsnapshot -t ${SNAPSHOT_NAME}

# 清理旧备份
find ${BACKUP_DIR} -name “*.tar.gz” -mtime +30 -delete

echo “=== 快照管理完成 ===” | tee -a ${LOG_FILE}

# 全量恢复

# 1. 停止Cassandra服务
# nodetool drain
# systemctl stop cassandra

# 输出
Draining nodes…
Node is drained
Stopping cassandra service: OK

# 2. 清空数据目录
# rm -rf /cassandra/fgdata/fgedudb/fgedu_users/*

# 3. 解压备份文件
# tar -xzf /backup/local/backup_20240115.tar.gz -C /tmp/

# 4. 复制快照文件
# cp -r /tmp/backup_20240115/* /cassandra/fgdata/fgedudb/fgedu_users/

# 5. 修改文件权限
# chown -R cassandra:cassandra /cassandra/fgdata/fgedudb/fgedu_users/

# 6. 启动Cassandra服务
# systemctl start cassandra

# 输出
Starting cassandra service: OK

# 7. 验证数据
# cqlsh -e “SELECT COUNT(*) FROM fgedudb.fgedu_users;”

count
——-
100000

(1 rows)

# 8. 运行repair
# nodetool repair -pr fgedudb

[2024-01-15 11:00:00,000] Starting repair…
[2024-01-15 11:30:00,000] Repair completed successfully

# 9. 验证集群状态
# nodetool status

Datacenter: dc1
===============
Status=Up/Down
|/ State=Normal/Leaving/Joining/Moving
— Address Load Tokens Owns (effective) Host ID Rack
UN 192.168.1.101 100 GB 256 100.0% a1b2c3d4… RAC1
UN 192.168.1.102 100 GB 256 100.0% e5f6g7h8… RAC1
UN 192.168.1.103 100 GB 256 100.0% i9j0k1l2… RAC2

# 表级恢复

# 1. 停止Cassandra服务（可选）
# 如果只恢复单个表，可以不停止服务
# 但建议停止服务以保证数据一致性

# 2. 清空表数据目录
# rm -rf /cassandra/fgdata/fgedudb/fgedu_users/*

# 3. 解压备份文件
# tar -xzf /backup/local/fgedudb_fgedu_users_20240115.tar.gz -C /tmp/

# 4. 复制快照文件
# cp -r /tmp/backup_20240115/* /cassandra/fgdata/fgedudb/fgedu_users/

# 5. 修改文件权限
# chown -R cassandra:cassandra /cassandra/fgdata/fgedudb/fgedu_users/

# 6. 刷新表
# nodetool refresh fgedudb fgedu_users

# 输出
Refreshing table fgedudb.fgedu_users

# 7. 验证数据
# cqlsh -e “SELECT COUNT(*) FROM fgedudb.fgedu_users;”

count
——-
100000

(1 rows)

# 8. 运行repair
# nodetool repair -pr fgedudb fgedu_users

# 9. 恢复脚本
#!/bin/bash
# table_restore.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

KEYSPACE=”fgedudb”
TABLE=”fgedu_users”
BACKUP_FILE=”/backup/local/fgedudb_fgedu_users_20240115.tar.gz”
LOG_FILE=”/cassandra/logs/backup/restore_$(date +%Y%m%d).log”

echo “=== 表级恢复开始 ===” | tee -a ${LOG_FILE}
echo “时间: $(date)” | tee -a ${LOG_FILE}
echo “Keyspace: ${KEYSPACE}” | tee -a ${LOG_FILE}
echo “Table: ${TABLE}” | tee -a ${LOG_FILE}

# 检查备份文件
if [ ! -f “${BACKUP_FILE}” ]; then
echo “错误: 备份文件不存在” | tee -a ${LOG_FILE}
exit 1
fi

# 清空表数据
echo “清空表数据…” | tee -a ${LOG_FILE}
rm -rf /cassandra/fgdata/${KEYSPACE}/${TABLE}/*

# 解压备份
echo “解压备份…” | tee -a ${LOG_FILE}
tar -xzf ${BACKUP_FILE} -C /tmp/ 2>&1 | tee -a ${LOG_FILE}

# 复制快照文件
echo “复制快照文件…” | tee -a ${LOG_FILE}
cp -r /tmp/backup_20240115/* /cassandra/fgdata/${KEYSPACE}/${TABLE}/ 2>&1 | tee -a ${LOG_FILE}

# 修改权限
echo “修改权限…” | tee -a ${LOG_FILE}
chown -R cassandra:cassandra /cassandra/fgdata/${KEYSPACE}/${TABLE}/

# 刷新表
echo “刷新表…” | tee -a ${LOG_FILE}
nodetool refresh ${KEYSPACE} ${TABLE} 2>&1 | tee -a ${LOG_FILE}

# 验证数据
echo “验证数据…” | tee -a ${LOG_FILE}
COUNT=$(cqlsh -e “SELECT COUNT(*) FROM ${KEYSPACE}.${TABLE};” | head -4 | tail -1 | tr -d ‘ ‘)
echo “数据量: ${COUNT}” | tee -a ${LOG_FILE}

# 运行repair
echo “运行repair…” | tee -a ${LOG_FILE}
nodetool repair -pr ${KEYSPACE} ${TABLE} 2>&1 | tee -a ${LOG_FILE}

echo “=== 表级恢复完成 ===” | tee -a ${LOG_FILE}

#!/bin/bash
# full_backup.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

# Cassandra全量备份脚本

# 配置变量
CASSANDRA_HOME=”/cassandra/app/current”
DATA_DIR=”/cassandra/fgdata”
BACKUP_DIR=”/backup/local”
REMOTE_BACKUP_DIR=”/backup/remote”
SNAPSHOT_NAME=”full_backup_$(date +%Y%m%d)”
LOG_FILE=”/cassandra/logs/backup/full_backup_$(date +%Y%m%d).log”
RETENTION_DAYS=30

# 创建日志目录
mkdir -p $(dirname ${LOG_FILE})
mkdir -p ${BACKUP_DIR}

# 日志函数
log() {
echo “[$(date ‘+%Y-%m-%d %H:%M:%S’)] $1″ | tee -a ${LOG_FILE}
}

# 发送告警
send_alert() {
local subject=”Cassandra备份告警: $1″
local message=”$2”
echo “${message}” | mail -s “${subject}” admin@fgedu.net.cn
}

# 主函数
main() {
log “=== Cassandra全量备份开始 ===”

# 1. 检查磁盘空间
log “检查磁盘空间…”
AVAILABLE=$(df -BG ${BACKUP_DIR} | tail -1 | awk ‘{print $4}’ | sed ‘s/G//’)
if [ ${AVAILABLE} -lt 100 ]; then
log “警告: 磁盘空间不足 (${AVAILABLE}GB)”
send_alert “磁盘空间不足” “备份目录剩余空间: ${AVAILABLE}GB”
fi

# 2. 创建快照
log “创建快照: ${SNAPSHOT_NAME}”
nodetool snapshot -t ${SNAPSHOT_NAME} 2>&1 | tee -a ${LOG_FILE}
if [ $? -ne 0 ]; then
log “错误: 快照创建失败”
send_alert “快照创建失败” “快照名称: ${SNAPSHOT_NAME}”
exit 1
fi

# 3. 等待快照完成
sleep 10

# 4. 归档快照
log “归档快照…”
ARCHIVE_DIR=”${BACKUP_DIR}/${SNAPSHOT_NAME}”
mkdir -p ${ARCHIVE_DIR}

# 遍历所有Keyspace和Table
for keyspace_dir in ${DATA_DIR}/*; do
if [ -d “${keyspace_dir}” ]; then
keyspace=$(basename ${keyspace_dir})
log “处理Keyspace: ${keyspace}”

for table_dir in ${keyspace_dir}/*; do
if [ -d “${table_dir}” ]; then
table=$(basename ${table_dir})
snapshot_dir=”${table_dir}/snapshots/${SNAPSHOT_NAME}”

if [ -d “${snapshot_dir}” ]; then
log “归档表: ${keyspace}.${table}”
mkdir -p ${ARCHIVE_DIR}/${keyspace}/${table}
cp -r ${snapshot_dir}/* ${ARCHIVE_DIR}/${keyspace}/${table}/
fi
fi
done
fi
done

# 5. 压缩备份
log “压缩备份…”
cd ${BACKUP_DIR}
tar -czf ${SNAPSHOT_NAME}.tar.gz ${SNAPSHOT_NAME} 2>&1 | tee -a ${LOG_FILE}
if [ $? -ne 0 ]; then
log “错误: 压缩失败”
send_alert “压缩失败” “备份名称: ${SNAPSHOT_NAME}”
exit 1
fi

# 6. 计算MD5
log “计算MD5…”
md5sum ${SNAPSHOT_NAME}.tar.gz > ${SNAPSHOT_NAME}.tar.gz.md5

# 7. 清理临时文件
log “清理临时文件…”
rm -rf ${ARCHIVE_DIR}

# 8. 清理本地快照
log “清理本地快照…”
nodetool clearsnapshot -t ${SNAPSHOT_NAME} 2>&1 | tee -a ${LOG_FILE}

# 9. 传输到远程
log “传输到远程服务器…”
scp ${SNAPSHOT_NAME}.tar.gz ${SNAPSHOT_NAME}.tar.gz.md5 backup-server:${REMOTE_BACKUP_DIR}/ 2>&1 | tee -a ${LOG_FILE}
if [ $? -ne 0 ]; then
log “警告: 远程传输失败”
send_alert “远程传输失败” “备份名称: ${SNAPSHOT_NAME}”
fi

# 10. 清理旧备份
log “清理旧备份…”
find ${BACKUP_DIR} -name “full_backup_*.tar.gz” -mtime +${RETENTION_DAYS} -delete
find ${BACKUP_DIR} -name “full_backup_*.tar.gz.md5” -mtime +${RETENTION_DAYS} -delete

# 11. 验证备份
log “验证备份…”
BACKUP_SIZE=$(du -sh ${BACKUP_DIR}/${SNAPSHOT_NAME}.tar.gz | awk ‘{print $1}’)
log “备份大小: ${BACKUP_SIZE}”

# 12. 记录备份信息
log “记录备份信息…”
echo “${SNAPSHOT_NAME},${BACKUP_SIZE},$(date ‘+%Y-%m-%d %H:%M:%S’)” >> ${BACKUP_DIR}/backup_history.csv

log “=== Cassandra全量备份完成 ===”
}

# 执行主函数
main

#!/bin/bash
# incremental_backup.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

# Cassandra增量备份脚本

# 配置变量
DATA_DIR=”/cassandra/fgdata”
BACKUP_DIR=”/backup/incremental”
LOG_FILE=”/cassandra/logs/backup/incremental_$(date +%Y%m%d%H%M).log”
RETENTION_DAYS=7

# 创建目录
mkdir -p ${BACKUP_DIR}
mkdir -p $(dirname ${LOG_FILE})

# 日志函数
log() {
echo “[$(date ‘+%Y-%m-%d %H:%M:%S’)] $1” | tee -a ${LOG_FILE}
}

# 主函数
main() {
log “=== Cassandra增量备份开始 ===”

# 备份目录
BACKUP_DATE=$(date +%Y%m%d)
BACKUP_TIME=$(date +%H%M)
BACKUP_PATH=”${BACKUP_DIR}/${BACKUP_DATE}/${BACKUP_TIME}”
mkdir -p ${BACKUP_PATH}

# 遍历所有Keyspace和Table
for keyspace_dir in ${DATA_DIR}/*; do
if [ -d “${keyspace_dir}” ]; then
keyspace=$(basename ${keyspace_dir})

for table_dir in ${keyspace_dir}/*; do
if [ -d “${table_dir}” ]; then
table=$(basename ${table_dir})
backups_dir=”${table_dir}/backups”

if [ -d “${backups_dir}” ] && [ “$(ls -A ${backups_dir})” ]; then
log “备份增量: ${keyspace}.${table}”
mkdir -p ${BACKUP_PATH}/${keyspace}/${table}
cp -r ${backups_dir}/* ${BACKUP_PATH}/${keyspace}/${table}/

# 清空增量备份目录
rm -rf ${backups_dir}/*
fi
fi
done
fi
done

# 压缩增量备份
if [ “$(ls -A ${BACKUP_PATH})” ]; then
log “压缩增量备份…”
cd ${BACKUP_DIR}/${BACKUP_DATE}
tar -czf ${BACKUP_TIME}.tar.gz ${BACKUP_TIME} 2>&1 | tee -a ${LOG_FILE}
rm -rf ${BACKUP_TIME}

# 计算MD5
md5sum ${BACKUP_TIME}.tar.gz > ${BACKUP_TIME}.tar.gz.md5
else
log “无增量数据”
rm -rf ${BACKUP_PATH}
fi

# 清理旧备份
log “清理旧备份…”
find ${BACKUP_DIR} -type d -mtime +${RETENTION_DAYS} -exec rm -rf {} \; 2>/dev/null

log “=== Cassandra增量备份完成 ===”
}

# 执行主函数
main