Senhores, em uma replicação nativa master/slave apareceu o erro:
"record with incorrect prev-link B/BBD4FF30 at B/C0D4FF78" no slave.
Apenas um master e um slave

Log do Server (SLAVE):

2017-01-06 13:06:45 BRST [1883]: [1-1] user=, db= LOG:  database system was
interrupted while in recovery at log time 2017-01-06 12:39:03 BRST
2017-01-06 13:06:45 BRST [1883]: [2-1] user=, db= HINT:  If this has
occurred more than once some data might be corrupted and you might need to
choose an earlier recovery target.
2017-01-06 13:06:46 BRST [1883]: [3-1] user=, db= LOG:  entering standby
mode
2017-01-06 13:06:46 BRST [1883]: [4-1] user=, db= LOG:  redo starts at
B/C0CED7C0
2017-01-06 13:06:46 BRST [1883]: [5-1] user=, db= LOG:  consistent recovery
state reached at B/C0D3B030
2017-01-06 13:06:46 BRST [1227]: [3-1] user=, db= LOG:  database system is
ready to accept read only connections
2017-01-06 13:06:46 BRST [1883]: [6-1] user=, db= LOG:  record with
incorrect prev-link B/BBD4FF30 at B/C0D4FF78
2017-01-06 13:06:46 BRST [2132]: [1-1] user=, db= LOG:  started streaming
WAL from primary at B/C0000000 on timeline 1
2017-01-06 18:51:03 BRST [2132]: [2-1] user=, db= FATAL:  terminating
walreceiver due to timeout
2017-01-06 18:51:03 BRST [1883]: [7-1] user=, db= LOG:  record with zero
length at B/C1AE0F48
2017-01-06 18:51:18 BRST [2331]: [1-1] user=, db= LOG:  started streaming
WAL from primary at B/C1000000 on timeline 1
2017-01-06 18:52:18 BRST [2331]: [2-1] user=, db= FATAL:  terminating
walreceiver due to timeout
2017-01-06 18:52:33 BRST [2333]: [1-1] user=, db= LOG:  started streaming
WAL from primary at B/C1000000 on timeline 1
2017-01-06 18:53:33 BRST [2333]: [2-1] user=, db= FATAL:  terminating
walreceiver due to timeout
2017-01-06 18:53:48 BRST [2334]: [1-1] user=, db= LOG:  started streaming
WAL from primary at B/C1000000 on timeline 1
2017-01-06 18:54:48 BRST [2334]: [2-1] user=, db= FATAL:  terminating
walreceiver due to timeout
2017-01-06 18:55:03 BRST [2335]: [1-1] user=, db= LOG:  started streaming
WAL from primary at B/C1000000 on timeline 1

Tentei achar uma documentação sobre o assunto mais não foi possível.
Então:
1 - gostaria de saber se este erro implica em ter que refazer o standby
(slave)?
2 - os erros posteriores de timeout tem alguma coisa haver com esse erro?
3 - Existe uma forma de verificar a integridade do master/slave?

Verifiquei a quantidade de registros em várias tabelas e estes estavam
consistentes em ambos server's!


Uso linux nos dois server's: CentOS Linux release 7.1.1503 (Core);
Postgresql Versão 9.4.10 no master e slave

Agradecimentos antecipados.
_______________________________________________
pgbr-geral mailing list
[email protected]
https://listas.postgresql.org.br/cgi-bin/mailman/listinfo/pgbr-geral

Responder a