搜索结果: "backup"

tags: [database, mysql, backup, troubleshooting, storage]

| XtraBackup redo 追赶 | 拷贝数据文件 + 持续拷贝 redo，追到一致点 | 物理热备 |

| `LOCK INSTANCE FOR BACKUP` | MySQL 8.0，备份期间允许 DML、阻塞 DDL | 需 `BACKUP_ADMIN` 权限 |

- **LSN**: XtraBackup 的 `xtrabackup_checkpoints` 中的 `from_lsn / to_lsn`

| **XtraBackup** | 物理 | 极短 | 多线程 | ✅ 是 | qpress/zstd | 100 GB ~ 10 TB |

tags: [redis, database, backup, data-loss, production]

sources: [raw/articles/redis-rdb-aof-backup-strategy.md]

# redis_rdb_backup.sh

BACKUP_DIR="/data/redis_backup"

BACKUP_FILE="${BACKUP_DIR}/dump_${DATE}.rdb"

tags: [tools, linux, automation, monitoring, backup, mysql, debugging]

BACKUP_DIR="/data/backup/mysql"

DB_USER="backup_user"

mkdir -p "$BACKUP_DIR"

mysqldump -u"$DB_USER" -p"$DB_PASS" "$db" | gzip > "$BACKUP_DIR/${db}_${DATE}.sql.gz"

tags: [linux, command, networking, performance, backup, compression]

[backup]

path = /data/backup/

auth users = backupuser

rsync -avz rsync://backupuser@/backup/ /local/backup/

tags: [linux, troubleshooting, backup, security, disk, filesystem, command, case-study, data-loss]

find -L /data/backup -type d -mtime +30 -exec rm -rf {} \;

新任运维小 A 接手后做了一次目录重构，把 `/data/backup` 拆成了 `mysql/`、`app/`、`logs/`、`tmp/` 四个子目录。他改了 `cleanup.sh` 里的路径，把 `/data/backup` 改成了 `/data/backup/tmp`——因为他觉得临时目录总归要清。但这一步没有经过 Code Review，也没有在预发环境演练过。

**致命陷阱：** `/data/backup/tmp` 是一个指向父目录 `/data/backup` 的软链——上一任运维为了偷懒创建的。`find -L` 主动跟随软链，结果把整个 `/data/backup` 当成"临时目录"展开去清理了。`/data/backup/mysql`（mysqldump 文件）和 `/data/backup/app`（应用归档包）全部被删，`/data/backup/logs` 因为被后台进程打开了几个文件（rm 删掉了硬链接但 inode 还在进程手上），看起来"还在"实际也是假象。

df /data/backup

tags: [linux, command, backup, networking, troubleshooting]

Host prod-backup

User backup_user

scp prod-server:/remote/file.txt prod-backup:/backup/

rsync -av /source/dir/ /backup/dir/

两台机器都设置为 `BACKUP` + `nopreempt`（禁用抢占模式），由 priority 决定谁为主：

state BACKUP # 两台都设为 BACKUP

BACKUP_DIR="/etc/nginx/backups/$(date +%Y%m%d_%H%M%S)"

mkdir -p $BACKUP_DIR

cp -r $NGINX_CONF_DIR/* $BACKUP_DIR/

- 备节点（Backup）监听广播，主节点失联（3 次未收到广播）时自动接管 VIP

| state | MASTER | BACKUP | 初始状态，最终由 priority 决定 |

state BACKUP

Backup: 无 VIP + Nginx 待命

Master: check_nginx 脚本返回非 0 → priority -= weight → priority < Backup

- 新增标签：redis, database, persistence, backup, data-loss

- Created raw: raw/articles/redis-rdb-aof-backup-strategy.md

- Created concepts: redis-persistence-strategy, redis-backup-recovery

- Updated: redis-backup-recovery, redis-connection-management

- redis-backup-recovery: +2（→ redis-persistence-strategy, redis-memory-optimization）

nice -n 10 python3 backup.py # 启动时设优先级（-20~19，-20 最高）

cat > /etc/systemd/system/backup.timer << 'EOF'

Description=Daily Backup

systemctl enable --now backup.timer

mysqldump --single-transaction --source-data=2 --all-databases > full_backup.sql

mysql -u root -p < full_backup.sql

| [[mysql-backup-selection-guide]] | MySQL 备份方案完整工程手册 — 17 种工具对比/mysqldump/XtraBackup/m |

cp /var/log/messages /tmp/messages_backup

cp /var/log/secure /tmp/secure_backup

cp /var/log/cron /tmp/cron_backup

| **1236** binlog position error | binlog 被清理或损坏 | 重建从库（xtrabackup/mysqldump） |

主从复制修复回滚——最终方案是重建从库（xtrabackup 全量备份 → 从库恢复 → 重新配置复制）。

sources: [raw/articles/redis-rdb-aof-backup-strategy.md]

- [[redis-backup-recovery]] — Redis 备份恢复与容灾

name: jenkins-backup-20260430

ssh backup-server "cat > /backups/jenkins/jenkins-team-a-$(date +%Y%m%d).tar.gz"

tar -czvf backup.tar.gz --exclude='*.log' /home/

tar -czvf - /home/ | ssh remote "cd /backup && tar -xzvf -"

- [[mysql-backup-selection-guide]] — MySQL 备份方案完整工程手册 — 17 种工具对比/mysqldump/XtraBackup/mydumper 实战/PITR 时间点恢复/LVM 快照/云厂商快照/按数据量级选型/翻车案例

- [[redis-backup-recovery]] — Redis 备份恢复实战：备份脚本 / 恢复流程 / 容灾方案

- backup: 备份恢复

| [[mysql-backup-selection-guide]] | MySQL 备份方案完整工程手册 — 17 种工具对比/mysqldump/XtraBackup/m |

| [[mysql-backup-selection-guide]] | MySQL 备份方案完整工程手册 — 17 种工具对比/mysqldump/XtraBackup/m |

| [[redis-backup-recovery]] | Redis 备份恢复与容灾 |

| [[redis-backup-recovery]] | 备份应在从库执行，主从不是备份的替代 |

| [[redis-backup-recovery]] | 备份恢复（内存与持久化配合） |

server 10.0.0.3:8080 backup;

kubectl get deployment -n -o yaml > backup_.yaml

2. **配置备份** — `kubectl get -n -o yaml > backup.yaml`

gzip -c /var/log/nginx/access.log > /backup/nginx/access.log.gz && rm access.log

| `rsync -avz /backup/ user@remote:/backup/` | 远程同步文件 |

sudo cp /var/lib/aide/aide.db /mnt/readonly-backup/aide.db.$(date +%Y%m%d)

server 10.0.1.12:8080 backup;

*.tfstate.backup