Re: Getting OLD messages

Kjetil Torgrim Homme (kjetilho@ifi.uio.no)
20 Feb 1996 17:14:24 +0100


[Ryan O'Connell]

| Yup - I think the European server is having problems.

Not really the European server, but it's the default server for
Europe. I find it strange that noone has volunteered to be an exploder
for the Netherlands or the UK.

Ifi.uio.no is an exploder for Sweden and Denmark, and I'm very
impressed with how smoothly it is running. I never seem to notice the
posts from linux-kernel in the mail queue, despite the hundred or so
messages which are sent every day to 50-60 recipients. MaFrom owner-linux-kernel-outgoing@vger.rutgers.edu Wed Feb 21 09:04:47 1996
Received: from orchard.washtenaw.cc.mi.us (orchard.washtenaw.cc.mi.us [198.111.176.4]) by herbie.ucs.indiana.edu (8.6.12/8.6.12) with ESMTP id JAA28963 for <linuxkernel@herbie.ucs.indiana.edu>; Wed, 21 Feb 1996 09:04:47 -0500
Received: from vger.rutgers.edu by orchard.washtenaw.cc.mi.us (8.6.10/2.3)
with ESMTP id IAA17738; Wed, 21 Feb 1996 08:38:29 -0500
Received: by vger.rutgers.edu id <115954-7625>; Wed, 21 Feb 1996 07:57:30 -0500
Date: Tue, 20 Feb 1996 13:40:01 +0100 (MET)
From: Hans-Helmut Buehmann <hans@expmech.ing.tu-bs.de>
To: linux-kernel@vger.rutgers.edu
Subject: memory counted wrong ?!
Message-ID: <Pine.LNX.3.91.960220133802.4743A-100000@golum.expmech.ing.tu-bs.de>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-kernel@vger.rutgers.edu
Precedence: bulk

Hello!

With my Linux-system (version 1.3.66) "cat /proc/meminfo"
gives, if the system has high load, for exaple the following result:

total: used: free: shared: buffers: cached:
Mem: 31940608 22433792 9506816 22142976 1044480 8998912
Swap: 33640448 0 33640448

With version 1.3.59 a few weeks ago it gives:

total: used: free: shared: buffers: cached:
Mem: 15138816 14675968 462848 15314944 262144 4608000
Swap: 33640448 2252800 31387648

total: used: free: shared: buffers: cached:
Mem: 15138816 14704640 434176 14376960 180224 4370432
Swap: 33640448 2441216 31199232

total: used: free: shared: buffers: cached:
Mem: 15073280 14745600 327680 16306176 364544 4546560
Swap: 33640448 2392064 31248384

total: used: free: shared: buffers: cached:
Mem: 15073280 14888960 184320 16666624 233472 4849664
Swap: 33640448 2281472 31358976

Why is cached + shared > used ?
Or why is shared > total ?

Interrups the kswapd-process (high priority) the kernel-process
while counting the memory ?

I wrote a program (patch) for procps to display the memory, but
sometimes the graphics is wrong.

The used computer:
i486, 66 MHz
16 MB ram / kernel v1.3.59
32 MB ram / kernel v1.3.66
NCR 810 SCSI
GCC 2.7.2 libc.so.5.2.18

Is this a bug or a feature ?

Greetings
Hans-Helmut Buehmann

hans@expmech.ing.tu-bs.de