正常运行状态

  在主节点或从节点运行命令  ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf service status 展示的信息中没有 WARNING； Status和 Upstream字段没有出现 ?和 !符号

$ ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf service status
 ID | Name                          | Role    | Status    | Upstream                      | ltclusterd | PID   | Paused? | Upstream last seen
----+-------------------------------+---------+-----------+-------------------------------+------------+-------+---------+--------------------
 1  | lightdbCluster102014812210001 | primary | * running |                               | running    | 13257 | no      | n/a                
 2  | lightdbCluster102014812610001 | standby |   running | lightdbCluster102014812210001 | running    | 29747 | no      | 0 second(s) ago

在各个节点运行命令 ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf node check展示的各个检查项的均为 OK。

示例结果：

$ ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf node check
Node "lightdbCluster102014812210001":
	Server role: OK (node is primary)
	Replication lag: OK (N/A - node is primary)
	WAL archiving: OK (0 pending archive ready files)
	Upstream connection: OK (N/A - node is primary)
	Downstream servers: OK (1 of 1 downstream nodes attached)
	Replication slots: OK (1 of 1 physical replication slots are active)
	Missing physical replication slots: OK (node has no missing physical replication slots)
	Configured data directory: OK (configured "data_directory" is "/data2/lightdb/lightdb-test/data")

异常状态一：主备不同步

在主节点或从节点运行命令  ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf service status  展示的信息有 WARNING: node "lightdbCluster102014812610001" (ID: 2) is running but the ltcluster node record is inactive ；  standby 的running出现 ! 符号。表示，主备状态不同步 ：

$ ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf service status

 ID | Name                          | Role    | Status    | Upstream                      | ltclusterd | PID   | Paused? | Upstream last seen
----+-------------------------------+---------+-----------+-------------------------------+------------+-------+---------+--------------------
 1  | lightdbCluster102014812210001 | primary | * running |                               | running    | 13257 | no      | n/a                
 2  | lightdbCluster102014812610001 | standby | ! running | lightdbCluster102014812210001 | running    | 29747 | no      | n/a                
WARNING: following issues were detected
  - node "lightdbCluster102014812610001" (ID: 2) is running but the ltcluster node record is inactive

原因分析：网络延迟过高，同步起来比较慢

解决方案：1. 建议部署时检查网络延迟；2. 备节点可能重新一段时间的失联，如果网络没有问题，可等待主节点同步完。

异常状态二：复制槽丢失

在主节点或从节点运行命令  ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf service status  展示的信息有 WARNING:

node "lightdbCluster102014812610001" (ID: 2) is not attached to its upstream node "lightdbCluster102014812210001" (ID: 1) ；  standbyd的上游节点找不到 lightdbCluster102014812210001 出现 ! 符号。表示主节点无法同步到备节点 ，链接不到复制槽：

$ ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf service status
WARNING: node "lightdbCluster102014812610001" not found in "pg_stat_replication"
 ID | Name                          | Role    | Status    | Upstream                        | ltclusterd | PID   | Paused? | Upstream last seen
----+-------------------------------+---------+-----------+---------------------------------+------------+-------+---------+--------------------
 1  | lightdbCluster102014812210001 | primary | * running |                                 | running    | 13257 | no      | n/a                
 2  | lightdbCluster102014812610001 | standby |   running | ! lightdbCluster102014812210001 | running    | 8988  | no      | 0 second(s) ago    
WARNING: following issues were detected
  - node "lightdbCluster102014812610001" (ID: 2) is not attached to its upstream node "lightdbCluster102014812210001" (ID: 1)

在各个节点运行命令 ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf node check 展示的各个检查项，提示找不到逻辑复制槽

$ ltcluster -f $LTHOME/etc/ltcluster/ltcluster.conf node check
WARNING: node "lightdbCluster102014812610001" not found in "pg_stat_replication"
Node "lightdbCluster102014812210001":
        Server role: OK (node is primary)
        Replication lag: OK (N/A - node is primary)
        WAL archiving: OK (0 pending archive ready files)
        Upstream connection: OK (N/A - node is primary)
        Downstream servers: CRITICAL (1 of 1 downstream nodes not attached; missing: lightdbCluster102014812610001 (ID: 2))
        Replication slots: OK (node has no physical replication slots)
        Missing physical replication slots: CRITICAL (1 physical replication slots are missing: ltcluster_slot_2)
        Configured data directory: OK (configured "data_directory" is "/data2/lightdb/lightdb-test/data")

查看同步表  select * from pg_stat_replication; 是否有同步记录，如果没有则不能同步