[pgpool-general: 9328] Re: pgpool - FATAL: unable to read data from DB node 0

[Tatsuo Ishii]

>> 2025-01-29 06:52:46.705: sr_check_worker pid 404220: FATAL:  unable to
>> read data from DB node 0
Pgpool-II's streaming replication process failed to connect to PostgreSQL.

The reason is in the PostgreSQL log.
>> 2025-01-29 06:52:43.987 UTC [404104] FATAL:  remaining connection slots
>> are reserved for non-replication superuser connections

It seems there's no usable PostgreSQL's connection slot rmaining. You
need to either increase max_connections or reduce num_init_children
(or both).

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

> Can someone help to understand this case
> 
> thank you.
> 
> On Wed, Jan 29, 2025 at 4:22 PM Mukesh Tanuku <mukesh.postgres at gmail.com>
> wrote:
> 
>> Hello Everyone,
>>
>> I am seeing these messages in pgpool log
>>
>> 2025-01-29 06:52:46.705: sr_check_worker pid 404220: FATAL:  unable to
>> read data from DB node 0
>> 2025-01-29 06:52:46.705: sr_check_worker pid 404220: DETAIL:  EOF
>> encountered with backend
>> 2025-01-29 06:52:46.705: main pid 3598: LOG:  reaper handler
>> 2025-01-29 06:52:46.706: main pid 3598: LOG:  reaper handler: exiting
>> normally
>> 2025-01-29 06:52:46.706: sr_check_worker pid 404222: LOG:  process started
>> 2025-01-29 06:52:46.709: sr_check_worker pid 404222: FATAL:  unable to
>> read data from DB node 0
>> 2025-01-29 06:52:46.709: sr_check_worker pid 404222: DETAIL:  EOF
>> encountered with backend
>> 2025-01-29 06:52:46.709: main pid 3598: LOG:  reaper handler
>> 2025-01-29 06:52:46.709: main pid 3598: LOG:  reaper handler: exiting
>> normally
>> 2025-01-29 06:52:46.709: sr_check_worker pid 404225: LOG:  process started
>> 2025-01-29 06:52:46.712: sr_check_worker pid 404225: FATAL:  unable to
>> read data from DB node 0
>> 2025-01-29 06:52:46.712: sr_check_worker pid 404225: DETAIL:  EOF
>> encountered with backend
>>
>>
>> what exactly this error means?
>>
>> in pgpool.conf we set this below,
>>
>> *sr_check_period = 10sr_check_user = 'monitor'*
>>
>>
>> at the same time im seeing this in postgresql log file :
>>
>> 2025-01-29 06:52:43.987 UTC [404104] FATAL:  remaining connection slots
>> are reserved for non-replication superuser connections
>> 2025-01-29 06:52:46.557 UTC [404108] FATAL:  remaining connection slots
>> are reserved for non-replication superuser connections
>> 2025-01-29 06:52:46.562 UTC [404110] FATAL:  remaining connection slots
>> are reserved for non-replication superuser connections
>> 2025-01-29 06:52:46.566 UTC [404112] FATAL:  remaining connection slots
>> are reserved for non-replication superuser connections
>> 2025-01-29 06:52:46.569 UTC [404114] FATAL:  remaining connection slots
>> are reserved for non-replication superuser connections
>>
>>
>> We haven't set any idle_session_timeout
>> or idle_in_transaction_session_timeout parameters in postgresql.conf file.
>>
>> Can someone help me understanding on this issue.
>>
>> Regards
>> Mukesh Tanuku
>>