While I agree with your reply in general and am working that angle and more, I'm hoping to add to my personal tool kit and gain more insight into methods to test fsync and prove without a doubt that it is functioning properly on any given system no matter what type of database I'm running.
T
On Tue, Aug 7, 2012 at 4:01 PM, Simon Riggs <simon@xxxxxxxxxxxxxxx> wrote:
On 7 August 2012 23:23, Terry Schmitt <tschmitt@xxxxxxxxxxxxxxxx> wrote:I think its important to say that as soon as someone patches/forks
> I have a pretty strange issue that I'm looking for ideas on.
> I'm using Postgres Plus Advanced Server 9.1, but I believe this problem is
> relevant to Postgres Community. It is certainly possible to be a EDB bug and
> I am already working with them on this.
Postgres code then the onus is really on whoever did that to show the
problem wasn't created by them when they did that. That is very much
the reason why my company and many others don't market a forked
product and stick closely to the community distributed binaries.
I have much respect for your vendors ability to diagnose faults and
recreate them on stock Postgres if appropriate, so I think we should
wait for that to happen rather than attempt to answer this just yet.
Or alternatively, please submit a test case that fails on Postgres if
you doubt that.
--
Simon Riggs http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
