> On 1998-08-16T02:37:44,
> linker@z.ml.org said:
>
> > While creeping swap would be a problem, I think that there should be a
> > daemon to add a little emergency swap. Not more then 32megs or so.. and
> > ofcourse it should shrink it once it's not in use..
>
> Thats nonsense. You could just have added the swap right from the start. We
> are talking about the point where all the available swap has been allocated.
>
11:35am up 2 days, 16:48, 6 users, load average: 0.00, 0.00, 0.00
total used free shared buffers cached
Mem: 63308 54100 9208 8004 19664 23660
-/+ buffers: 10776 52532
Swap: 72288 300 71988
^my personal machine.
I only have swap for that rare situation that something has gone out of
control. If it weren't for that fact that it's saved me a few times, it
would be 16mb or less...
Disk space is cheap right now but I'm almost out on this box (I've gotta
make clean my gimp source tree b4 I can do anything else)... I'd love to
be able to have only 16megs and have a daemon increase it in an emergency,
at least then I could make tempoary use of that space..
It's not that important on this system.. But I have some 486es with 200meg
hdds that it would be moure useful (though, as long as they at least
reboot when running OOM then I dont care much)...
Ram is so cheap today that many people dont need swap unless it's an
emergency..
Some other examples:
My glfs computer:
11:33am up 2 days, 16:44, 2 users, load average: 0.12, 0.04, 0.01
total used free shared buffers cached
Mem: 63432 59976 3456 38388 9160 32516
-/+ buffers/cache: 18300 45132
Swap: 66492 1068 65424
A server:
11:32am up 86 days, 22:12, 2 users, load average: 0.22, 0.31, 0.36
total used free shared buffers cached
Mem: 63160 62484 676 23424 16648 14960
-/+ buffers/cache: 30876 32284
Swap: 72288 3964 68324
Another server:
11:32am up 34 days, 11:09, 3 users, load average: 0.35, 0.44, 0.39
total used free shared buffers cached
Mem: 160116 157664 2452 31580 56612 40520
-/+ buffers/cache: 60532 99584
Swap: 124956 0 124956
See, swap in unnessassary on all of these computers.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html