On 1 October 2010 14:12, Dave Page <dpage@pgadmin.org> wrote:
> On Fri, Oct 1, 2010 at 2:00 PM, Thom Brown <thom@linux.com> wrote:
>> I can't recreate this, so it may not be much help, but I left PgAdmin
>> running with no query windows open at all, and eventually I noticed my
>> PC slowing right down as memory started to swap. I've attached a
>> screenshot to show its memory usage.
>
> VM size of a gig? Thats small for Windows isn't it? :-p
>
>> Actions I had performed earlier are attempting to select all rows from
>> a massive table, and closed that result window a few seconds into it
>> attempting to fetch results.
>
> That's where my suspicions lie. Can you try to reproduce it by doing that again?
No, I've attempted that with several large tables, but it doesn't make
any noticeable difference. Where it's actually started fetching
results and displaying them, closing the window causes pgAdmin to
correctly returns the memory. I've also checked what's happening
server-side when closing the window, and the query is cancelled. Is
it theorically possible for pgAdmin to fail to cancel the query when
closing the results window?
Note: Whenever I say results window, I actually mean the view/edit data window.
--
Thom Brown
Twitter: @darkixion
IRC (freenode): dark_ixion
Registered Linux user: #516935