Hi,
Is there a way to find out how much work_mem my server(*) is actually
using on a running system in production?
* - session or cluster. I don't mean explain analyze.
What i am really interested in is: how does postgres utilize the
memory in the server? How much goes where, in reality.
I want to sound an alarm when the server starts swapping (can do) and
be able find out if work_mem is too high.
I also want to see if work_mem gets in the way of shared_buffers or
the OS file cache. On munin for example.
So, is there a way to find out the amount of work_mem that has been allocated?
Cheers,
WBL
--
"Patriotism is the conviction that your country is superior to all
others because you were born in it." -- George Bernard Shaw