MIME-Version: 1.0
From: Pablo Calientos <mputzu@gmail.com>
Date: Tue, 24 Jun 2025 10:44:42 +0200
Message-ID: <CA+BBeFCAHRj3PTTi3shWdpLyMK7KPP+eAs63TYgmfLK_6Backg@mail.gmail.com>
Subject: PGPool II does not recognize backend_server0
To: pgsql-general@postgresql.org
Content-Type: multipart/alternative; boundary="00000000000092dee806384d5795"
Archived-At: <https://www.postgresql.org/message-id/CA%2BBBeFCAHRj3PTTi3shWdpLyMK7KPP%2BeAs63TYgmfLK_6Backg%40mail.gmail.com>
Precedence: bulk

--00000000000092dee806384d5795
Content-Type: text/plain; charset="UTF-8"

Hi all,
I'm new to pgpool, I'm testing from scratch with version
[postgres@pgpool-01 ~]$ pgpool --version
pgpool-II version 4.6.2 (chirikoboshi)

I'm following the ufficial guide at
https://www.pgpool.net/docs/latest/en/html/example-basic.html.
I started testing native replication mode, I use this configuration in
virtualbox VMs:

server pgppol 192.168.1.118  OEL 9.6
server post01 192.168.1.116  OEL 9.6
server post02 192.168.1.117  OEL 9.6

All VMs can ping/telnet between them, from server pgpool I can connect
to others postgres server via psql:

[postgres@pgpool-01 ~]$ psql -h 192.168.1.116
psql (13.20)
Type "help" for help.

postgres=# \q
[postgres@pgpool-01 ~]$ psql -h 192.168.1.117
psql (13.20)
Type "help" for help.

postgres=#

The relevant parameters in pgpool.conf are the following:

backend_clustering_mode = 'native_replication'

backend_hostname0 = '192.168.1.116'
backend_port0 = 5432
backend_weight0 = 1
backend_data_directory0 = '/PGDATA'
backend_application_name0 = 'server0'

backend_hostname1 = '192.168.1.117'
backend_port1 = 5432
backend_weight1 = 1
backend_data_directory1 = '/PGDATA'
backend_application_name1 = 'server1'

following the guide I sent creation for a database "test" from pgpool:

[postgres@pgpool-01 ~]$ createdb -p 9999 test

but the db has been created only in backend_hostname1.
IN fact pgpool report as down backend_hostname0:


[postgres@pgpool-01 ~]$ psql -h localhost -p 9999
psql (13.20)
Type "help" for help.

postgres=# show pool_nodes;
 node_id |   hostname    | port | status | pg_status | lb_weight |
role   | pg_role | select_cnt | load_balance_no
de | replication_delay | replication_state | replication_sync_state |
last_status_change
---------+---------------+------+--------+-----------+-----------+---------+---------+------------+----------------
---+-------------------+-------------------+------------------------+---------------------
 0       | 192.168.1.116 | 5432 | down   | unknown   | 0.500000  |
replica | replica | 0          | false
   | 0                 |                   |                        |
2025-06-19 16:28:12
 1       | 192.168.1.117 | 5432 | up     | unknown   | 0.500000  |
main    | main    | 0          | true
   | 0                 |                   |                        |
2025-06-19 16:28:12
(2 rows)


But I dont see errors in configuration. If I change the backend
server, switching the IP addresses (to be sure there are not
network or access misconfiguration) the server down results always the
same backend_hostname0, I mean the problem is not in
the postgres server (changing IP does not change the status of show pool_nodes):

new configurazion:

backend_hostname0 = '192.168.1.117'		<<< switched IP
backend_port0 = 5432
backend_weight0 = 1
backend_data_directory0 = '/PGDATA'
backend_application_name0 = 'server0'

backend_hostname1 = '192.168.1.116'		<<< switched IP
backend_port1 = 5432
backend_weight1 = 1
backend_data_directory1 = '/PGDATA'
backend_application_name1 = 'server1'

reboot pgpopol and resent show pool_nodes:

[postgres@pgpool-01 ~]$ pgpool stop
2025-06-19 16:44:23.112: main pid 6247: LOG:  stop request sent to
pgpool (pid: 6124). waiting for termination...
.done.
[1]+  Done                    pgpool -n

[postgres@pgpool-01 ~]$ pgpool -n &
[1] 6248
[postgres@pgpool-01 ~]$ 2025-06-19 16:44:31.934: main pid 6248: LOG:
reading status file: 0 th backend is set to down status
2025-06-19 16:44:31.934: main pid 6248: LOG:  reading status file: 1
th backend is set to up status
2025-06-19 16:44:31.935: main pid 6248: LOG:  redirecting log output
to logging collector process
2025-06-19 16:44:31.935: main pid 6248: HINT:  Future log output will
appear in directory "/var/log/pgpool_log".

[postgres@pgpool-01 ~]$ psql -h localhost -p 9999
psql (13.20)
Type "help" for help.

postgres=# show pool_nodes;
 node_id |   hostname    | port | status | pg_status | lb_weight |
role   | pg_role | select_cnt | load_balance_no
de | replication_delay | replication_state | replication_sync_state |
last_status_change
---------+---------------+------+--------+-----------+-----------+---------+---------+------------+----------------
---+-------------------+-------------------+------------------------+---------------------
 0       | 192.168.1.117 | 5432 | down   | unknown   | 0.500000  |
replica | replica | 0          | false
   | 0                 |                   |                        |
2025-06-19 16:44:31
 1       | 192.168.1.116 | 5432 | up     | unknown   | 0.500000  |
main    | main    | 0          | true
   | 0                 |                   |                        |
2025-06-19 16:44:31
(2 rows)

postgres=#

I cannot understand what I'm missing.
Any advice is welcome.

thanks in advance.

MP

--00000000000092dee806384d5795
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><pre style=3D"margin:0px;padding:0px;color:rgb(0,0,0);font=
-size:12px">Hi all,<br>I&#39;m new to pgpool, I&#39;m testing from scratch =
with version <br>[postgres@pgpool-01 ~]$ pgpool --version<br>pgpool-II vers=
ion 4.6.2 (chirikoboshi)<br><br>I&#39;m following the ufficial guide at <a =
href=3D"https://www.pgpool.net/docs/latest/en/html/example-basic.html">http=
s://www.pgpool.net/docs/latest/en/html/example-basic.html</a>.<br>I started=
 testing native replication mode, I use this configuration in virtualbox VM=
s:<br><br>server pgppol 192.168.1.118  OEL 9.6<br>server post01 192.168.1.1=
16  OEL 9.6<br>server post02 192.168.1.117  OEL 9.6<br><br>All VMs can ping=
/telnet between them, from server pgpool I can connect to others postgres s=
erver via psql:<br><br>[postgres@pgpool-01 ~]$ psql -h 192.168.1.116<br>psq=
l (13.20)<br>Type &quot;help&quot; for help.<br><br>postgres=3D# \q<br>[pos=
tgres@pgpool-01 ~]$ psql -h 192.168.1.117<br>psql (13.20)<br>Type &quot;hel=
p&quot; for help.<br><br>postgres=3D#<br><br>The relevant parameters in pgp=
ool.conf are the following:<br><br>backend_clustering_mode =3D &#39;native_=
replication&#39;<br><br>backend_hostname0 =3D &#39;192.168.1.116&#39;<br>ba=
ckend_port0 =3D 5432<br>backend_weight0 =3D 1<br>backend_data_directory0 =
=3D &#39;/PGDATA&#39;<br>backend_application_name0 =3D &#39;server0&#39;<br=
><br>backend_hostname1 =3D &#39;192.168.1.117&#39;<br>backend_port1 =3D 543=
2<br>backend_weight1 =3D 1<br>backend_data_directory1 =3D &#39;/PGDATA&#39;=
<br>backend_application_name1 =3D &#39;server1&#39;<br><br>following the gu=
ide I sent creation for a database &quot;test&quot; from pgpool:<br><br>[po=
stgres@pgpool-01 ~]$ createdb -p 9999 test<br><br>but the db has been creat=
ed only in backend_hostname1.<br>IN fact pgpool report as down backend_host=
name0:<br><br><br>[postgres@pgpool-01 ~]$ psql -h localhost -p 9999<br>psql=
 (13.20)<br>Type &quot;help&quot; for help.<br><br>postgres=3D# show pool_n=
odes;<br> node_id |   hostname    | port | status | pg_status | lb_weight |=
  role   | pg_role | select_cnt | load_balance_no<br>de | replication_delay=
 | replication_state | replication_sync_state | last_status_change<br>-----=
----+---------------+------+--------+-----------+-----------+---------+----=
-----+------------+----------------<br>---+-------------------+------------=
-------+------------------------+---------------------<br> 0       | 192.16=
8.1.116 | 5432 | down   | unknown   | 0.500000  | replica | replica | 0    =
      | false<br>   | 0                 |                   |              =
          | 2025-06-19 16:28:12<br> 1       | 192.168.1.117 | 5432 | up    =
 | unknown   | 0.500000  | main    | main    | 0          | true<br>   | 0 =
                |                   |                        | 2025-06-19 1=
6:28:12<br>(2 rows)<br><br><br>But I dont see errors in configuration. If I=
 change the backend server, switching the IP addresses (to be sure there ar=
e not<br>network or access misconfiguration) the server down results always=
 the same backend_hostname0, I mean the problem is not in <br>the postgres =
server (changing IP does not change the status of show pool_nodes):<br><br>=
new configurazion:<br><br>backend_hostname0 =3D &#39;192.168.1.117&#39;		&l=
t;&lt;&lt; switched IP<br>backend_port0 =3D 5432<br>backend_weight0 =3D 1<b=
r>backend_data_directory0 =3D &#39;/PGDATA&#39;<br>backend_application_name=
0 =3D &#39;server0&#39;<br><br>backend_hostname1 =3D &#39;192.168.1.116&#39=
;		&lt;&lt;&lt; switched IP<br>backend_port1 =3D 5432<br>backend_weight1 =
=3D 1<br>backend_data_directory1 =3D &#39;/PGDATA&#39;<br>backend_applicati=
on_name1 =3D &#39;server1&#39;<br><br>reboot pgpopol and resent show pool_n=
odes:<br><br>[postgres@pgpool-01 ~]$ pgpool stop<br>2025-06-19 16:44:23.112=
: main pid 6247: LOG:  stop request sent to pgpool (pid: 6124). waiting for=
 termination...<br>.done.<br>[1]+  Done                    pgpool -n<br><br=
>[postgres@pgpool-01 ~]$ pgpool -n &amp;<br>[1] 6248<br>[postgres@pgpool-01=
 ~]$ 2025-06-19 16:44:31.934: main pid 6248: LOG:  reading status file: 0 t=
h backend is set to down status<br>2025-06-19 16:44:31.934: main pid 6248: =
LOG:  reading status file: 1 th backend is set to up status<br>2025-06-19 1=
6:44:31.935: main pid 6248: LOG:  redirecting log output to logging collect=
or process<br>2025-06-19 16:44:31.935: main pid 6248: HINT:  Future log out=
put will appear in directory &quot;/var/log/pgpool_log&quot;.<br><br>[postg=
res@pgpool-01 ~]$ psql -h localhost -p 9999<br>psql (13.20)<br>Type &quot;h=
elp&quot; for help.<br><br>postgres=3D# show pool_nodes;<br> node_id |   ho=
stname    | port | status | pg_status | lb_weight |  role   | pg_role | sel=
ect_cnt | load_balance_no<br>de | replication_delay | replication_state | r=
eplication_sync_state | last_status_change<br>---------+---------------+---=
---+--------+-----------+-----------+---------+---------+------------+-----=
-----------<br>---+-------------------+-------------------+----------------=
--------+---------------------<br> 0       | 192.168.1.117 | 5432 | down   =
| unknown   | 0.500000  | replica | replica | 0          | false<br>   | 0 =
                |                   |                        | 2025-06-19 1=
6:44:31<br> 1       | 192.168.1.116 | 5432 | up     | unknown   | 0.500000 =
 | main    | main    | 0          | true<br>   | 0                 |       =
            |                        | 2025-06-19 16:44:31<br>(2 rows)<br><=
br>postgres=3D#<br><br>I cannot understand what I&#39;m missing.<br>Any adv=
ice is welcome.<br><br>thanks in advance.<br><br>MP</pre><br class=3D"gmail=
-Apple-interchange-newline"></div>

--00000000000092dee806384d5795--