On Thu, 1 Oct 1998, Jim Peters wrote:

> Here is a free report for everyone BTW, you'll enjoy this.  I admit it's a
> weird situation.  I'm trying newer kernels as I write this.
> 
>              total       used       free     shared    buffers     cached
> Mem:        257176     238112      19064      69184      91992      69844
> -/+ buffers/cache:      76276     180900
> Swap:       130748          0     130748
> 
>              total       used       free     shared    buffers     cached
> Mem:        515848     482776      33072      76616     176136     210868
> -/+ buffers/cache:      95772     420076
> Swap:       261496          0     261496

Uhhh, what's wrong with this?  All it means is that you have 128 MB of
swap (or 256 MB of swap in the second case) and that you haven't used
any yet.  I have a bunch of systems like that.  It's not like you WANT
the systems to use swap, it's more like you want them to be ABLE to use
swap if they need to.  Crank up a couple of 100 MB jobs and you should
see them swap.

A system with NO swap looks like this (just happen to have one handy:-):

rgb@ganesh|T:249>b6 free
             total used free shared buffers cached
Mem:  514368 36132 478236 11600 12676 15244
-/+ buffers:             8212     506156
Swap:            0          0          0

A system with swap that hasn't used it yet looks like this:

rgb@ganesh|T:250>b1 free
             total used free shared buffers cached
Mem:  514368 60780 453588 11972 35716 13860
-/+ buffers:            11204     503164
Swap:       522992          0     522992

And a system with swap that has just barely dipped into it during a peak
looks like this (a the beauties of having a rack of identical systems
except for b6 that is waiting to be reinstalled:-):

rgb@ganesh|T:252>b4 free
             total used free shared buffers cached
Mem:  514300 505952 8348 4004 13860 175604
-/+ buffers:           316488     197812
Swap:       522992      19052     503940


So, you just haven't made your systems work hard/long enough.

   rgb

Robert G. Brown                        http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:[EMAIL PROTECTED]


Reply via email to