[pgpool-general: 1293] pgpool stopped accepting client connections after 1 node hung

[Lonni J Friedman]

I'm running pgpool-3.2.1 on a Linux-x86_64 server.  Late last night,
one of my 3 postgresql-9.2.2 servers controlled by pgpool hung (not
the master), and pgpool stopped accepting client connections for no
apparent reason.  In the pgpool log at the time of the hang, I see the
following errors generated thousands of times repeatedly:

2013-01-05 20:08:47 ERROR: pid 31413: connect_inet_domain_socket:
connect() failed: Connection timed out
2013-01-05 20:08:47 ERROR: pid 31413: connection to cuda-db5(5432) failed
2013-01-05 20:08:47 ERROR: pid 31413: new_connection: create_cp() failed
2013-01-05 20:08:47 ERROR: pid 9476: connect_inet_domain_socket:
connect() failed: Connection timed out
2013-01-05 20:08:47 ERROR: pid 9476: connection to cuda-db5(5432) failed
2013-01-05 20:08:47 ERROR: pid 9476: new_connection: create_cp() failed
2013-01-05 20:08:47 ERROR: pid 7850: connect_inet_domain_socket:
connect() failed: Connection timed out
2013-01-05 20:08:47 ERROR: pid 7850: connection to cuda-db5(5432) failed
2013-01-05 20:08:47 ERROR: pid 7850: new_connection: create_cp() failed

I don't understand why pgpool stopped accepting client connections.
I'd expect that if any single node goes down, pgpool should continue
to work and accept connections, and simply mark the unresponsive node
as unavailable.

I've attached my pgpool.conf, in case this is some sort of
configuration error on my part.

thanks
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pgpool.conf
Type: application/octet-stream
Size: 24916 bytes
Desc: not available
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20130106/e3ba26dc/attachment-0001.obj>