Обсуждение: WAL archive messages in streaming standby server's log
For some reason, after testing failover successfully and then failing back to restore the original primary server, my standbyserver is acting a little strange. I have the following log messages repeating every 5 minutes: < 2014-05-06 13:41:17.705 EDT <@> % >LOG: restored log file "000000020000000000000046" from archive < 2014-05-06 13:41:17.709 EDT <@> % >LOG: record with zero length at 0/46000090 < 2014-05-06 13:41:17.710 EDT <@> % >LOG: restarted WAL streaming at 0/46000000 on timeline 2 < 2014-05-06 13:41:17.710 EDT <@> % >LOG: replication terminated by primary server < 2014-05-06 13:41:17.710 EDT <@> % >DETAIL: End of WAL reached on timeline 2 at 0/46000090. Note that these are not shown as error messages, but it doesn't appear that WAL processing is continuing as there are atleast a dozen newer WAL files from the primary which are not being copied into the standby server's pg_xlog directory.I suspect the problem lies with the the second line indicating a record with zero length, but don't know how toclear this out so it goes away - short of doing a pgbasebackup on the primary and rebuilding the standby from that. Anyone have any ideas? Jay
On Tue, May 6, 2014 at 2:41 PM, Scalia, John <john.scalia@verizon.com> wrote:
< 2014-05-06 13:41:17.710 EDT <@> % >DETAIL: End of WAL reached on timeline 2 at 0/46000090.
Looks like you did a promotion or a complete restore and changed the timeline of your primary server.
If can add the following line on your recovery.conf file to make it follows the new timeline:
recovery_target_timeline = latest
recovery_target_timeline = latest
Best regards,
--
Matheus de Oliveira
Analista de Banco de Dados
Dextra Sistemas - MPS.Br nível F!
www.dextra.com.br/postgres
--
Matheus de Oliveira
Analista de Banco de Dados
Dextra Sistemas - MPS.Br nível F!
www.dextra.com.br/postgres