Обсуждение: [ADMIN] Failover but master is running
Hello everyone in the list, testing pgpool a failover is triggered but the master is running. The pgpool version is 3.5.5. This is the log reported by pgpool.
Could anybody help me to clarify this error?
The reason could be postgresql server has reached max_connections?
Regards.
2017-02-23 14:18:58: pid 5303: ERROR: unable to read data
2017-02-23 14:18:58: pid 5303: DETAIL: child connection forced to terminate due to client_idle_limit:300 is reached
2017-02-23 14:25:31: pid 5479: LOG: pool_read_kind: error message from 1 th backend:sorry, too many clients already
2017-02-23 14:25:31: pid 5479: ERROR: unable to read message kind
2017-02-23 14:25:31: pid 5479: DETAIL: kind does not match between master(52) slot[1] (45)
2017-02-23 14:25:36: pid 5287: ERROR: failed to authenticate
2017-02-23 14:25:36: pid 5287: DETAIL: invalid authentication message response type, Expecting 'R' and received 'E'
2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate
2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication message response type, Expecting 'R' and received 'E'
2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate
2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication message response type, Expecting 'R' and received 'E'
2017-02-23 14:25:41: pid 55651: LOG: setting backend node 1 status to NODE DOWN
2017-02-23 14:25:41: pid 55651: LOG: received degenerate backend request for node_id: 1 from pid [55651]
2017-02-23 14:25:41: pid 55651: LOG: starting degeneration. shutdown host 192.168.5.19(5432)
Yes, it seems process 5479 hit the max_connections of PostgreSQL while connecting to it. Best regards, -- Tatsuo Ishii SRA OSS, Inc. Japan English: http://www.sraoss.co.jp/index_en.php Japanese:http://www.sraoss.co.jp > Hello everyone in the list, testing pgpool a failover is triggered but the > master is running. The pgpool version is 3.5.5. This is the log reported by > pgpool. > > > > Could anybody help me to clarify this error? > > > > The reason could be postgresql server has reached max_connections? > > > > Regards. > > 2017-02-23 14:18:58: pid 5303: ERROR: unable to read data > > 2017-02-23 14:18:58: pid 5303: DETAIL: child connection forced to terminate > due to client_idle_limit:300 is reached > > 2017-02-23 14:25:31: pid 5479: LOG: pool_read_kind: error message from 1 th > backend:sorry, too many clients already > > 2017-02-23 14:25:31: pid 5479: ERROR: unable to read message kind > > 2017-02-23 14:25:31: pid 5479: DETAIL: kind does not match between > master(52) slot[1] (45) > > 2017-02-23 14:25:36: pid 5287: ERROR: failed to authenticate > > 2017-02-23 14:25:36: pid 5287: DETAIL: invalid authentication message > response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate > > 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication message > response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate > > 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication message > response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: LOG: setting backend node 1 status to NODE > DOWN > > 2017-02-23 14:25:41: pid 55651: LOG: received degenerate backend request > for node_id: 1 from pid [55651] > > 2017-02-23 14:25:41: pid 55651: LOG: starting degeneration. shutdown host > 192.168.5.19(5432) >
Thank you very much for the answer. Regards. -----Mensaje original----- De: Tatsuo Ishii [mailto:ishii@sraoss.co.jp] Enviado el: jueves, 23 de febrero de 2017 05:14 p. m. Para: lazaro3487@gmail.com CC: pgpool-general@pgpool.net; pgsql-admin@postgresql.org Asunto: Re: [pgpool-general: 5345] Failover but master is running Yes, it seems process 5479 hit the max_connections of PostgreSQL while connecting to it. Best regards, -- Tatsuo Ishii SRA OSS, Inc. Japan English: http://www.sraoss.co.jp/index_en.php Japanese:http://www.sraoss.co.jp > Hello everyone in the list, testing pgpool a failover is triggered but > the master is running. The pgpool version is 3.5.5. This is the log > reported by pgpool. > > > > Could anybody help me to clarify this error? > > > > The reason could be postgresql server has reached max_connections? > > > > Regards. > > 2017-02-23 14:18:58: pid 5303: ERROR: unable to read data > > 2017-02-23 14:18:58: pid 5303: DETAIL: child connection forced to > terminate due to client_idle_limit:300 is reached > > 2017-02-23 14:25:31: pid 5479: LOG: pool_read_kind: error message > from 1 th backend:sorry, too many clients already > > 2017-02-23 14:25:31: pid 5479: ERROR: unable to read message kind > > 2017-02-23 14:25:31: pid 5479: DETAIL: kind does not match between > master(52) slot[1] (45) > > 2017-02-23 14:25:36: pid 5287: ERROR: failed to authenticate > > 2017-02-23 14:25:36: pid 5287: DETAIL: invalid authentication message > response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate > > 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication > message response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate > > 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication > message response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: LOG: setting backend node 1 status to > NODE DOWN > > 2017-02-23 14:25:41: pid 55651: LOG: received degenerate backend > request for node_id: 1 from pid [55651] > > 2017-02-23 14:25:41: pid 55651: LOG: starting degeneration. shutdown > host > 192.168.5.19(5432) >
Could be possible to use retries upon Streaming Replication Check for preventing failover when connections to master fails? Regards. -----Mensaje original----- De: Tatsuo Ishii [mailto:ishii@sraoss.co.jp] Enviado el: jueves, 23 de febrero de 2017 05:14 p. m. Para: lazaro3487@gmail.com CC: pgpool-general@pgpool.net; pgsql-admin@postgresql.org Asunto: Re: [pgpool-general: 5345] Failover but master is running Yes, it seems process 5479 hit the max_connections of PostgreSQL while connecting to it. Best regards, -- Tatsuo Ishii SRA OSS, Inc. Japan English: http://www.sraoss.co.jp/index_en.php Japanese:http://www.sraoss.co.jp > Hello everyone in the list, testing pgpool a failover is triggered but > the master is running. The pgpool version is 3.5.5. This is the log > reported by pgpool. > > > > Could anybody help me to clarify this error? > > > > The reason could be postgresql server has reached max_connections? > > > > Regards. > > 2017-02-23 14:18:58: pid 5303: ERROR: unable to read data > > 2017-02-23 14:18:58: pid 5303: DETAIL: child connection forced to > terminate due to client_idle_limit:300 is reached > > 2017-02-23 14:25:31: pid 5479: LOG: pool_read_kind: error message > from 1 th backend:sorry, too many clients already > > 2017-02-23 14:25:31: pid 5479: ERROR: unable to read message kind > > 2017-02-23 14:25:31: pid 5479: DETAIL: kind does not match between > master(52) slot[1] (45) > > 2017-02-23 14:25:36: pid 5287: ERROR: failed to authenticate > > 2017-02-23 14:25:36: pid 5287: DETAIL: invalid authentication message > response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate > > 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication > message response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate > > 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication > message response type, Expecting 'R' and received 'E' > > 2017-02-23 14:25:41: pid 55651: LOG: setting backend node 1 status to > NODE DOWN > > 2017-02-23 14:25:41: pid 55651: LOG: received degenerate backend > request for node_id: 1 from pid [55651] > > 2017-02-23 14:25:41: pid 55651: LOG: starting degeneration. shutdown > host > 192.168.5.19(5432) >
Retrying is not possible, but by setting fail_over_on_backend_error to off, you can avoid failover. Best regards, -- Tatsuo Ishii SRA OSS, Inc. Japan English: http://www.sraoss.co.jp/index_en.php Japanese:http://www.sraoss.co.jp > Could be possible to use retries upon Streaming Replication Check for > preventing failover when connections to master fails? > > Regards. > > -----Mensaje original----- > De: Tatsuo Ishii [mailto:ishii@sraoss.co.jp] > Enviado el: jueves, 23 de febrero de 2017 05:14 p. m. > Para: lazaro3487@gmail.com > CC: pgpool-general@pgpool.net; pgsql-admin@postgresql.org > Asunto: Re: [pgpool-general: 5345] Failover but master is running > > Yes, it seems process 5479 hit the max_connections of PostgreSQL while > connecting to it. > > Best regards, > -- > Tatsuo Ishii > SRA OSS, Inc. Japan > English: http://www.sraoss.co.jp/index_en.php > Japanese:http://www.sraoss.co.jp > >> Hello everyone in the list, testing pgpool a failover is triggered but >> the master is running. The pgpool version is 3.5.5. This is the log >> reported by pgpool. >> >> >> >> Could anybody help me to clarify this error? >> >> >> >> The reason could be postgresql server has reached max_connections? >> >> >> >> Regards. >> >> 2017-02-23 14:18:58: pid 5303: ERROR: unable to read data >> >> 2017-02-23 14:18:58: pid 5303: DETAIL: child connection forced to >> terminate due to client_idle_limit:300 is reached >> >> 2017-02-23 14:25:31: pid 5479: LOG: pool_read_kind: error message >> from 1 th backend:sorry, too many clients already >> >> 2017-02-23 14:25:31: pid 5479: ERROR: unable to read message kind >> >> 2017-02-23 14:25:31: pid 5479: DETAIL: kind does not match between >> master(52) slot[1] (45) >> >> 2017-02-23 14:25:36: pid 5287: ERROR: failed to authenticate >> >> 2017-02-23 14:25:36: pid 5287: DETAIL: invalid authentication message >> response type, Expecting 'R' and received 'E' >> >> 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate >> >> 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication >> message response type, Expecting 'R' and received 'E' >> >> 2017-02-23 14:25:41: pid 55651: ERROR: failed to authenticate >> >> 2017-02-23 14:25:41: pid 55651: DETAIL: invalid authentication >> message response type, Expecting 'R' and received 'E' >> >> 2017-02-23 14:25:41: pid 55651: LOG: setting backend node 1 status to >> NODE DOWN >> >> 2017-02-23 14:25:41: pid 55651: LOG: received degenerate backend >> request for node_id: 1 from pid [55651] >> >> 2017-02-23 14:25:41: pid 55651: LOG: starting degeneration. shutdown >> host >> 192.168.5.19(5432) >> >