Postgresql链式奴隶和奴隶提升

Question

所以我有一个类似-> Slave1 -> Slave2的设置。Slave1从主人那里复制，Slave2从Slave1复制。如果我把Slave1推广到掌握中，那么Slave2是否有可能继续从Slave1复制。这是我在Slave2上看到的日志，我将Slave1提升为master之后：

2014-06-13 04:28:11 UTC LOG:  replication terminated by primary server
2014-06-13 04:28:11 UTC DETAIL:  End of WAL reached on timeline 3 at 3E/62B2C758.
2014-06-13 04:28:11 UTC LOG:  fetching timeline history file for timeline 4 from primary server
2014-06-13 04:28:11 UTC LOG:  invalid record length at 3E/62B2C758
2014-06-13 04:28:11 UTC LOG:  restarted WAL streaming at 3E/62000000 on timeline 3
2014-06-13 04:28:11 UTC LOG:  replication terminated by primary server
2014-06-13 04:28:11 UTC DETAIL:  End of WAL reached on timeline 3 at 3E/62B2C758.
2014-06-13 04:28:16 UTC LOG:  restarted WAL streaming at 3E/62000000 on timeline 3
2014-06-13 04:28:16 UTC LOG:  replication terminated by primary server

Answer 1

一旦您提升了Standby1，您将发现并应该将0000000x.history (其中x是时间线)文件复制到standby2的pg_xlog dir中。如果在recovery.conf中使用restore命令，这将“自动”发生。