Re: [PATCH] Revert default wal_sync_method to fdatasync on Linux 2.6.33+

On 12/06/2010 08:38 PM, Tom Lane wrote:
> Josh Berkus<josh(at)agliodbs(dot)com> writes:
>> Making it support O_DIRECT would be possible but more complex; I don't
>> see the point unless we think we're going to have open_sync_with_odirect
>> as a seperate option.
> Whether it's complex or not isn't really the issue. The issue is that
> what test_fsync is testing had better match what the backend does, or
> people will be making choices based on not-comparable test results.
> I think we should have test_fsync just automatically fold in O_DIRECT
> the same way the backend does.
>
>

Indeed. We were quite confused for a while when we were dealing with
this about a week ago, and my handwritten test program failed as
expected but test_fsync didn't. Anything other than behaving just as the
backend does violates POLA, in my view.

cheers

andrew