Re: Is it a memory leak in PostgreSQL 7.4beta?

Hans,

You are right about the startup memory - here is the top line for a few 
seconds after startup :
PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  
COMMAND          
10116 postgres  15   0  3816 3816 3180 R 33.8  1.0   0:01.03 postmaster  

seems that VIRT, RES, SHR all get the increase counted against them as 
time goes on (as Tom suggested, I guess its to do with how top does its 
accounting on this platform).

Hans-Jürgen Schönig wrote:

>
> I can hardly imagine that the backend started working with 9mb of 
> memory. what did you do that PostgreSQL needed so much memory from the 
> beginning??? are you using the default settings? usually the 
> postmaster does not need more than 3mb at startup (in this scenario).
>
>