Re: Pgpool with high availability

Hi Adrian,

After modifying the pgpool.conf configuration to correct a subnet mistake,
Previously, the wrong subnet was specified as /26, and it has now been
corrected to /24. The configuration changes were as follows:

Previous Configuration:

delegate_ip = '10.127.1.18'

if_up_cmd = '/sbin/ip addr add $_IP_$/26 dev eth0 label eth0:1'

if_down_cmd = '/sbin/ip addr del $_IP_$/26 dev eth0'

arping_cmd = '/usr/sbin/arping -U $_IP_$ -w 1 -I eth0'

Updated Configuration:

delegate_ip = '10.127.1.18'

if_up_cmd = '/sbin/ip addr add $_IP_$/24 dev eth0 label eth0:1'

if_down_cmd = '/sbin/ip addr del $_IP_$/24 dev eth0'

arping_cmd = '/usr/sbin/arping -U $_IP_$ -w 1 -I eth0'

Current Issue:-

Following the subnet correction, the Virtual IP (VIP) 10.127.1.18 is now
reachable only from the leader node (ha0002), while it remains unreachable
from the standby nodes (ha0001 and ha0003). Below are the details of the
connectivity status and the commands executed:

*Leader Node (**ha0002**)*

[root(at)staging-ha0002 PG_LOGS]# ping 10.127.1.18

PING 10.127.1.18 (10.127.1.18) 56(84) bytes of data.

64 bytes from 10.127.1.18: icmp_seq=1 ttl=64 time=0.041 ms

64 bytes from 10.127.1.18: icmp_seq=2 ttl=64 time=0.058 ms

64 bytes from 10.127.1.18: icmp_seq=3 ttl=64 time=0.060 ms

--- 10.127.1.18 ping statistics ---

3 packets transmitted, 3 received, 0% packet loss, time 2080ms

rtt min/avg/max/mdev = 0.041/0.053/0.060/0.008 ms

[root(at)staging-ha0002 PG_LOGS]# ping 10.127.1.18

PING 10.127.1.18 (10.127.1.18) 56(84) bytes of data.

64 bytes from 10.127.1.18: icmp_seq=1 ttl=64 time=0.041 ms

64 bytes from 10.127.1.18: icmp_seq=2 ttl=64 time=0.058 ms

64 bytes from 10.127.1.18: icmp_seq=3 ttl=64 time=0.060 ms

--- 10.127.1.18 ping statistics ---

3 packets transmitted, 3 received, 0% packet loss, time 2080ms

rtt min/avg/max/mdev = 0.041/0.053/0.060/0.008 ms

[pgbigboss(at)staging-ha0002 ~]$ pcp_watchdog_info -h 10.127.1.18 -p 9898 -U
pgbigboss -W

Password:

3 3 YES ha0002:9999 Linux staging-ha0002 ha0002

ha0002:9999 Linux staging-ha0002 ha0002 9999 9000 4 LEADER 0 MEMBER

ha0001:9999 Linux staging-ha0001 ha0001 9999 9000 7 STANDBY 0 MEMBER

ha0003:9999 Linux staging-ha0003 ha0003 9999 9000 7 STANDBY 0 MEMBER

*Standby Node (**ha0001**)*

*[root(at)staging-ha0001 ~]# ping 10.127.1.18*

*PING 10.127.1.18 (10.127.1.18) 56(84) bytes of data.*

*From 10.127.1.10 icmp_seq=1 Destination Host Unreachable*

*From 10.127.1.10 icmp_seq=2 Destination Host Unreachable*

*From 10.127.1.10 icmp_seq=3 Destination Host Unreachable*

*--- 10.127.1.18 ping statistics ---*

*5 packets transmitted, 0 received, +3 errors, 100% packet loss, time
4126ms*

*pipe 3*

*[pgbigboss(at)staging-ha0001 ~]$ pcp_watchdog_info -h 10.127.1.18 -p 9898 -U
pgbigboss -W*

*Password:*

*ERROR: connection to host "10.127.1.18" failed with error "No route to
host"*

The VIP 10.127.1.18 is accessible from the leader node (ha0002) but not
from the standby nodes (ha0001 and ha0003).

Thanks

Vijay

On Tue, May 28, 2024 at 11:31 PM Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>
wrote:

>
>
> On 5/28/24 1:31 AM, vijay patil wrote:
> >
> > HI Team,
> >
> > "I'm encountering challenges while configuring Pgpool with high
> > availability. The initial setup is completed, and Pgpool is operational
> > on a single node, functioning without issues. However, upon attempting
> > to start Pgpool on any additional nodes, particularly node 2, it becomes
> > immediately unreachable.
>
> And how we are supposed to arrive at an answer with essentially no
> information provided?
>
> Need:
>
> 1) Configuration for initial setup.
>
> 2) A more detailed explanation of what "... upon attempting
> to start Pgpool on any additional nodes" means? Include configuration
> changes.
>
> 3) The error messages.
>
> 4) Where the nodes are located?
>
>
> >
> > I'm seeking assistance to address this issue. My setup consists of three
> > nodes, each hosting both PostgreSQL and Pgpool services."
> >
> >
> > Thanks
> >
> > Vijay
> >
>
> --
> Adrian Klaver
> adrian(dot)klaver(at)aklaver(dot)com
>