# 服务器环境准备
$ hostname
fgedu.net.cn

$ nproc
32

$ free -h
total used free shared buff/cache available
Mem: 62G 2.1G 58G 9.8M 1.8G 59G

$ df -h /yashandb
Filesystem Size Used Avail Use% Mounted on
/dev/sda2 500G 20G 450G 5% /yashandb

# 网络配置
$ cat /etc/hosts
192.168.1.100 fgedu-primary
192.168.1.101 fgedu-standby

# 防火墙配置
$ firewall-cmd –add-port=5432/tcp –permanent
$ firewall-cmd –reload

# 共享存储配置
$ ls -l /dev/mapper/
crw——-. 1 root root 10, 236 Apr 11 10:00 control
brw-rw—-. 1 root disk 253, 0 Apr 11 10:00 vg_shared-lv_data

# 集群配置
$ cat /yashandb/app/yasdb/etc/yasdb.conf
cluster_name = “fgedu-cluster”
node_list = “node1,node2”
shared_storage = “/dev/mapper/vg_shared-lv_data”

# 复制参数
wal_level = ‘hot_standby’
max_wal_senders = 10
wal_keep_segments = 100
hot_standby = on

# 备库配置
$ cat /yashandb/app/yasdb/etc/yasdb.conf
# 备库参数
listen_addresses = ‘0.0.0.0’
port = 5432
max_connections = 1000
shared_buffers = ’16GB’

# 复制参数
wal_level = ‘hot_standby’
max_wal_senders = 10
wal_keep_segments = 100
hot_standby = on
primary_conninfo = ‘host=fgedu-primary port=5432 user=fgedu password=xxxx’
recovery_target_timeline = ‘latest’

# 心跳检测参数
heartbeat_interval = 10
heartbeat_timeout = 60

# 复制模式
# 同步复制
replication_mode = ‘synchronous’

# 异步复制
# replication_mode = ‘asynchronous’

# 半同步复制
# replication_mode = ‘semi-synchronous’

fgedudb=# select * from pg_stat_replication;
pid | usesysid | usename | application_name | client_addr | client_hostname | client_port | backend_start | backend_xmin | state | sent_lsn | write_lsn | flush_lsn | replay_lsn | write_lag | flush_lag | replay_lag | sync_priority | sync_state
——-+———-+——–+——————+—————+—————–+————-+——————————-+————–+———–+———–+———–+———–+————+———–+———–+————+—————+————
12345 | 10001 | fgedu | walreceiver | 192.168.1.101 | fgedu-standby | 54321 | 2026-04-11 10:00:00.000000+08 | | streaming | 0/1234567 | 0/1234567 | 0/1234567 | 0/1234567 | | | | 1 | sync
(1 row)

# 查看数据库状态
fgedudb=# select pg_is_in_recovery();
pg_is_in_recovery
——————-
f
(1 row)

# 查看复制延迟
fgedudb=# select now() – pg_last_xact_replay_timestamp() as replication_lag;
replication_lag
—————–
00:00:00.000000
(1 row)

# 监控主备状态
function check_replication() {
echo “=== 检查主备复制状态 ===”
yassql -U fgedu -d fgedudb -c “select * from pg_stat_replication;”
echo “”
}

# 监控数据库状态
function check_db_status() {
echo “=== 检查数据库状态 ===”
yassql -U fgedu -d fgedudb -c “select pg_is_in_recovery();”
echo “”
}

# 监控复制延迟
function check_replication_lag() {
echo “=== 检查复制延迟 ===”
yassql -U fgedu -d fgedudb -c “select now() – pg_last_xact_replay_timestamp() as replication_lag;”
echo “”
}

# 执行监控
check_replication
check_db_status
check_replication_lag

# 保存监控日志
echo “$(date ‘+%Y-%m-%d %H:%M:%S’) – 高可用监控完成” >> /yashandb/log/ha_monitor.log