"David F. Skoll" <dfs@roaringpenguin.com> writes:
> I have one PostgreSQL 8.3 machine feeding WAL files to another PostgreSQL 8.3
> machine that's running in recovery mode. However, fairly often (every
> few days), the standby machine breaks out of recovery mode with log messages
> like:
> 2009-10-23 21:47:40 EDT LOG: incorrect resource manager data checksum in record at 11/4E5AA18C
> I'm pretty sure the WAL is shipped correctly (it's shipped via SCP).
> The only thing I can think of is that the shipping version is 8.3.7 and
> the consuming version is 8.3.8. Would this be enough to cause the
> "incorrect checksum" message?
No; there's no WAL change between 8.3.7 and 8.3.8. What seems more
likely is that you're somehow shipping the WAL files before they're
quite finished.
regards, tom lane
