[pgpool-general: 5530] Re: Problem with PCP socket directory setting
m.usama at gmail.com
Wed May 24 05:17:26 JST 2017
This is expected behaviour, All PCP client side tools are independent of
Pgpool-II server side settings and have no visibility and access to the
pgpool.conf file, Since the utility can be used to connect to any local or
remote Pgpool, So PCP utilities always try to connect on the default socket
directory and default port unless explicitly specified in the command's
If you are using a different pcp_socket_dir and/or different pcp_port then
you would be required to specify the socket file location and port in the
pcp command argument,
For example to make PCP utilities use /var/run/postgresql as socket
directory you should specify the directory in the host argument.
# with pcp_port = 9898 and pcp_socket_dir = '/var/run/postgresql'
pcp_pool_status -h /var/run/postgresql -p 9898 -U <user>
On Tue, May 23, 2017 at 8:34 PM, Jeremiah Penery <jpenery at redskytech.com>
> I'm using the latest pgpool RPMs: pgpool-II-pg96-3.6.4-1pgdg.rhel6.x86_64
> There seems to be some problem with the PCP socket directory setting. I
> have it set as such:
> pcp_socket_dir = '/var/run/postgresql'
> The file is indeed created in that directory:
> [root at jp1 redsky]# locate 9898
> However, if I try to run some pcp command (e.g. pcp_node_info) without
> giving a host parameter, it seems to always want to use /tmp for the socket
> [root at jp1]# pcp_node_info -U admin -n 0
> ERROR: connection to socket "/tmp/.s.PGSQL.9898" failed with error "No
> such file or directory"
> If I provide the host parameter, it works fine, because I guess it's going
> over the network instead of using the socket file.
> [root at jp1 ]# pcp_node_info -U admin -h localhost -n 0
> 192.168.20.175 5433 2 1.000000 up
> pgpool-general mailing list
> pgpool-general at pgpool.net
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the pgpool-general