Hello,
I have a problem with pgpool package that I installed on a nixos 23.11.
context : 3 VMs , a postgresql and pgpool installed on each of theam, they can all be reachable on the port 5432
nc -v clustering-0 5432
succeeded
nc -v clustering-1 5432
succeeded
nc -v clustering-2 5432
succeeded
On each, I installed pgpool package
environment.systemPackages = [
pkgs.pgpool
];
I’ve configured it :
wd_lifecheck_method = 'heartbeat'
wd_interval = 10
wd_heartbeat_keepalive = 2
wd_heartbeat_deadtime = 30
heartbeat_hostname0 = 'clustering-0'
heartbeat_port0 = 9694
heartbeat_device0 = ''
heartbeat_hostname1 = 'clustering-1'
heartbeat_port1 = 9694
heartbeat_device1 = ''
heartbeat_hostname2 = 'clustering-2'
heartbeat_port2 = 9694
heartbeat_device2 = ''
However when I turn on gppool, I have this result in my logs on the VM clustering-0 :
2024-05-19 19:55:26.209: heart_beat_sender pid 1596: ERROR: failed to send watchdog heartbeat, gethostbyname() failed
2024-05-19 19:55:26.209: heart_beat_sender pid 1596: DETAIL: gethostbyname on "clustering-1" failed with reason: "No address associated with name"
2024-05-19 19:55:26.284: health_check2 pid 1828: LOG: health check retrying on DB node: 2 (round:2)
2024-05-19 19:55:26.518: heart_beat_sender pid 1600: ERROR: failed to send watchdog heartbeat, gethostbyname() failed
2024-05-19 19:55:26.518: heart_beat_sender pid 1600: DETAIL: gethostbyname on "clustering-2" failed with reason: "No address associated with name"
2024-05-19 19:55:27.287: health_check2 pid 1828: LOG: health check retrying on DB node: 2 (round:3)
2024-05-19 19:55:27.694: health_check1 pid 1827: LOG: health check retrying on DB node: 1 (round:1)
2024-05-19 19:55:28.212: heart_beat_sender pid 1596: ERROR: failed to send watchdog heartbeat, gethostbyname() failed
2024-05-19 19:55:28.212: heart_beat_sender pid 1596: DETAIL: gethostbyname on "clustering-1" failed with reason: "No address associated with name"
2024-05-19 19:55:28.291: health_check2 pid 1828: LOG: health check failed on node 2 (timeout:0)
2024-05-19 19:55:28.291: health_check2 pid 1828: LOG: health check failed on quarantine node 2 (timeout:0)
2024-05-19 19:55:28.291: health_check2 pid 1828: DETAIL: ignoring..
Is there a DNS configuration, or a package dependency involved by the usage of pgpool ?