EnterpriseDB / repmgr

A lightweight replication manager for PostgreSQL (Postgres)

Home Page:https://repmgr.org/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

crosscheck / matrix issue with non default pg port

seadba opened this issue · comments

commented

i have a v11 cluster on a non-default port 5443 - i have passwordless authentication set up and repgmr config works - nodes are registered and standby cloned and recovering - however if i run crosscheck or matrix i get the following errors
"could not connect to server: Connection refused
Is the server running on host "10.99.100.9" and accepting
TCP/IP connections on port 5432?"

  • node "10.99.100.9" (ID: 2) is registered as an active standby but is unreachable

conninfo on both servers

conninfo='host=10.99.100.9 port=5443 user=repmgr dbname=repmgr connect_timeout=2'

i cant figure out where the default 5432 is derived from - cluster show works fine from either node - however the port is not in repmgr.nodes

conninfo | host=10.99.100.10 user=repmgr dbname=repmgr connect_timeout=
2

any idea? thanks

commented

other repmgr commands recognize the port

$ repmgr -f /etc/repmgr/11/repmgr.conf standby unregister
INFO: connecting to local standby
DEBUG: connecting to: "user=repmgr connect_timeout=2 dbname=repmgr host=10.99.100.9 port=5443 fallback_application_name=repmgr options=-csearch_path="
INFO: connecting to primary database
DEBUG: connecting to: "user=repmgr connect_timeout=2 dbname=repmgr host=10.99.100.10 fallback_application_name=repmgr options=-csearch_path="
NOTICE: unregistering node 2
INFO: standby unregistration complete

commented

i switched to default 5432 and that seems to work - will test again with non-default port