Speaking of fulltextindex...

I just noticed this in some testing..

When I use my PHP application to insert text into the field that's used in
the full text index it takes 9 full seconds, when investigating resource
usage using 'top' I see this :

Development from a completely idle start up :

PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND

682 postgres 2 0 124M 2328K select 0 0:00 0.00% 0.00% postgres

Production server -- this one is being used :

PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU CPU COMMAND

96825 postgres 2 0 38380K 35464K sbwait 0:04 2.10% 2.10% postgres

The backend is started exactly the same way with the same version (7.0.2) on
these two servers with the options -B 4096 -o '-S 16384' -- can anyone
think of a reason why would one initially grow to 124 megs?

I'm waiting to see about that before I continue investigating the sudden
exponential increase in my INSERT speed - hopefully it's related (because I
sure see why the transaction suddenly take ten times longer to complete than
it did!)..

Thanks!!

-Mitch