[pgpool-general: 5200] Pgpool 3.4.8 is not stopping with wathdog. Problem with socket

[Lucas Luengas]

Hello.

I notice some problems with pgpool 3.4.8 with watchdog on Centos 6.7. Only
with watchdog enabled.

I have no problems with pgpool 3.4.6.

Using command "service pgpool restart" does not stop pgpool service, and
pgpool listen sockets are not closed.

With detailed log, I get these lines en /var/log/pgpool.log

2016-12-24 16:06:33: pid 9821: LOG:  pgpool-II successfully started.
version 3.4.8 (tataraboshi)
2016-12-24 16:06:33: pid 9821: LOCATION:  pgpool_main.c:332
2016-12-24 16:06:33: pid 9821: LOG:  find_primary_node: checking backend no
0

2016-12-24 16:06:33: pid 9821: LOCATION:  pgpool_main.c:2604
2016-12-24 16:06:33: pid 9821: LOG:  find_primary_node: primary node id is 0
2016-12-24 16:06:33: pid 9821: LOCATION:  pgpool_main.c:2634
.2016-12-24 16:06:34: pid 9906: LOG:  set SO_REUSEPORT option to the socket
2016-12-24 16:06:34: pid 9906: LOCATION:  wd_heartbeat.c:674
2016-12-24 16:06:34: pid 9906: LOG:  creating watchdog heartbeat receive
socket.
2016-12-24 16:06:34: pid 9906: DETAIL:  set SO_REUSEPORT
2016-12-24 16:06:34: pid 9906: LOCATION:  wd_heartbeat.c:208
2016-12-24 16:06:34: pid 9907: LOG:  set SO_REUSEPORT option to the socket
2016-12-24 16:06:34: pid 9907: LOCATION:  wd_heartbeat.c:674
2016-12-24 16:06:34: pid 9907: LOG:  creating socket for sending heartbeat
2016-12-24 16:06:34: pid 9907: DETAIL:  set SO_REUSEPORT
2016-12-24 16:06:34: pid 9907: LOCATION:  wd_heartbeat.c:126
..................


With listen_addresses = '0.0.0.0' I get same issue.

This issue is similar to http://www.pgpool.net/mantisbt/view.php?id=202

Kind regards.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pgpool.net/pipermail/pgpool-general/attachments/20161224/8060dd84/attachment.htm>