[pgpool-hackers: 4451] Re: Pgpool PCP get disabled. How it is possible?

[Tatsuo Ishii]

Hi,

> "0.0.0.0")}, 16) = 0
> [pid 20484] bind(6, {sa_family=AF_INET6, sin6_port=htons(9999),
> inet_pton(AF_INET6, "::", &sin6_addr), sin6_flowinfo=htonl(0),
> sin6_scope_id=0}, 28) = 0
> [pid 20508] bind(4, {sa_family=AF_INET, sin_port=htons(9694),
> sin_addr=inet_addr("0.0.0.0")}, 16) = 0
> [root at pg-mgrdb2 ~]#
> 
> I hope it gives any hint of what is going on. I'll try to check the source
> code a bit later.

I think the pgpool main process had been trying to find the primary
node. grep find_primary_node from the log shows:

[pid 20657] fcntl(5, F_GETFL2024-05-01 18:02:33: pid 20484: LOG:  find_primary_node_repeatedly: waiting for finding a primary node
[pid 20484] write(2, "2024-05-01 18:02:33: pid 20484: "..., 752024-05-01 18:02:33: pid 20484: LOG:  find_primary_node: standby node is 1
[pid 20484] write(2, "2024-05-01 18:02:34: pid 20484: "..., 752024-05-01 18:02:34: pid 20484: LOG:  find_primary_node: standby node is 1
:
:
[pid 20484] write(2, "2024-05-01 18:02:51: pid 20484: "..., 752024-05-01 18:02:51: pid 20484: LOG:  find_primary_node: standby node is 1
[pid 20484] write(2, "2024-05-01 18:02:52: pid 20484: "..., 752024-05-01 18:02:52: pid 20484: LOG:  find_primary_node: standby node is 1

By default pgpool tries to find the primary node up to
search_primary_node_timeout, default to 300 seconds. Since The search
started at: 18:02:33, and the search still continues. After it succeeds
to find the primary node or it's timed out, pcp sockets will be
created. I don't know why pgpool was failing to find the primary node
though.

Best reagards,
--
Tatsuo Ishii
SRA OSS LLC
English: http://www.sraoss.co.jp/index_en/
Japanese:http://www.sraoss.co.jp