灾备演练脚本示例

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

# 定义变量
PRIMARY_HOST="192.168.1.10"
STANDBY_HOST="192.168.2.10"
DB_USER="fgedu"
DB_PASS="Fgedu@123"
DATA_DIR="/opengauss/fgdata"

# 检查主库状态
check_primary() {
    echo "检查主库状态..."
    ssh $DB_USER@$PRIMARY_HOST "gs_ctl status -D $DATA_DIR"
    if [ $? -ne 0 ]; then
        echo "主库状态异常"
        return 1
    fi
    return 0
}

# 检查备库状态
check_standby() {
    echo "检查备库状态..."
    ssh $DB_USER@$STANDBY_HOST "gs_ctl status -D $DATA_DIR"
    if [ $? -ne 0 ]; then
        echo "备库状态异常"
        return 1
    fi
    return 0
}

# 检查复制状态
check_replication() {
    echo "检查复制状态..."
    ssh $DB_USER@$PRIMARY_HOST "gsql -U $DB_USER -d postgres -c 'SELECT slot_name, active, pg_wal_lsn_diff(pg_current_wal_lsn(), restart_lsn) as lag_bytes FROM pg_replication_slots; 
'"
}

# 执行故障切换
do_failover() {
    echo "开始执行故障切换演练..."
    ssh $DB_USER@$STANDBY_HOST "gs_ctl promote -D $DATA_DIR"
    if [ $? -eq 0 ]; then
        echo "故障切换成功，备库已提升为主库"
        # 验证新主库状态
        ssh $DB_USER@$STANDBY_HOST "gs_ctl status -D $DATA_DIR"
    else
        echo "故障切换失败"
    fi
}

# 恢复主备关系
restore_replication() {
    echo "恢复主备关系..."
    # 在原主库上执行基础备份
    ssh $DB_USER@$STANDBY_HOST "pg_basebackup -h $STANDBY_HOST -U $DB_USER -D $DATA_DIR -F p -X stream -P"
    # 创建recovery.conf
    ssh $DB_USER@$PRIMARY_HOST "cat > $DATA_DIR/recovery.conf << EOF
standby_mode = 'on'
primary_conninfo = 'host=$STANDBY_HOST port=5432 user=$DB_USER password=$DB_PASS'
recovery_target_timeline = 'latest'
EOF"
    # 启动原主库作为备库
    ssh $DB_USER@$PRIMARY_HOST "gs_ctl start -D $DATA_DIR"
}

# 主流程
echo "=== 灾备演练开始 ==="

# 检查初始状态
check_primary
check_standby
check_replication

# 执行故障切换
do_failover

# 验证切换结果
check_standby

# 恢复主备关系
restore_replication

# 检查最终状态
check_primary
check_standby
check_replication

echo "=== 灾备演练完成 ==="