Repmgr未能在claster中注册待机，我错过了什么？

Question

repmgr的问题：failed to register in claster

请告诉我我错过了什么？

主配置：

failover=automatic
follow_command='repmgr standby follow -f /etc/repmgr.conf --log-to-file'
cluster=cluster1
node=1
node_name=node100
conninfo='host=192.168.22.200 port=6432 user=repmgr dbname=repmgr'
master_response_timeout=30
pg_bindir=/usr/local/bin/
service_start_command = systemctl start postgresql-9.6
service_stop_command = systemctl stop postgresql-9.6
service_restart_command = systemctl restart postgresql-9.6

待机配置：

failover=automatic
follow_command='repmgr standby follow -f /etc/repmgr.conf --log-to-file'
cluster=cluster1
node=3
node_name=replica
conninfo='host=slav port=6432 user=repmgr dbname=repmgr'
master_response_timeout=30
pg_bindir=/usr/local/bin/
service_start_command = systemctl start postgresql-9.6
service_stop_command = systemctl stop postgresql-9.6
service_restart_command = systemctl restart postgresql-9.6

REPMGR日志：

[2017-08-09 10:36:37] [WARNING] No record found for node 3 
[2017-08-09 10:36:37] [ERROR] No metadata record found for this node - terminating 
[2017-08-09 10:36:37] [HINT] Check that 'repmgr (master|standby) register' was executed for this node

但当尝试将节点注册为备用节点时：

-bash-4.2$ repmgr -f /etc/repmgr.conf standby register
ERROR: Node 3 exists already with node_name "replica"

Answer 1

nub回答，也许可以帮助1.取消注册待机2.修改用户repmgr将search_path设置为repmgr_test，"$user"，public；3.尝试注册

conninfo='host=slav端口....也使用ip