Re: Streaming replication slave crash

Quentin Hartman <qhartman(at)direwolfdigital(dot)com> writes:
> Yesterday morning, one of my streaming replication slaves running 9.2.3
> crashed with the following in the log file:

What process did you use for setting up the slave?

There's a fix awaiting release in 9.2.4 that might explain data
corruption on a slave, depending on how it was set up:

* Ensure we do crash recovery before entering archive recovery,
if the database was not stopped cleanly and a recovery.conf file
is present (Heikki Linnakangas, Kyotaro Horiguchi, Mitsumasa Kondo)

This is needed to ensure that the database is consistent in certain
scenarios, such as initializing a standby server with a filesystem
snapshot from a running server.

This theory would be more probable if it's a relatively new slave, since
any corruption would have been there in the slave's initial state, just
waiting for the replay to run into it.

regards, tom lane