[pgpool-general: 6088] pgpool3.7.2 not detecting master node?

[Yong Zhang]

Hi, all

I have pgpool2 3.7.2 with 2 postgresql-10 nodes (version 10.4) in the backend, pgpool starts with below logs:

May 16 16:17:49 pgpool[4496]: 2018-05-16 16:17:49: pid 4496: LOG:  Backend status file /var/log/postgresql/pgpool_status discarded
May 16 16:17:49 pgpool[4496]: 2018-05-16 16:17:49: pid 4496: LOG:  Setting up socket for 0.0.0.0:9999
May 16 16:17:49 pgpool[4496]: 2018-05-16 16:17:49: pid 4496: LOG:  Setting up socket for :::9999
May 16 16:17:49 pgpool[4496]: 2018-05-16 16:17:49: pid 4496: LOG:  find_primary_node_repeatedly: waiting for finding a primary node
May 16 16:17:49 pgpool[4496]: 2018-05-16 16:17:49: pid 4496: LOG:  find_primary_node: checking backend no 0
May 16 16:17:49 pgpool[4496]: 2018-05-16 16:17:49: pid 4496: LOG:  find_primary_node: checking backend no 1
May 16 16:17:49 pgpool[4496]: 2018-05-16 16:17:49: pid 4496: LOG:  find_primary_node: primary node id is 1

Backend0 is my slave node and backend1 is my master node, master and slave are in stream replication mode.

It looks fine from logs, but when clients connected to pgpool, found requests were routed to backend0 not backend1, why???

If I use PostgreSQL-9.6, everything works fine. Please help, thanks.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20180516/0945f997/attachment.html>