Re: Page faults with physical I/O on Linux

From: Ulrich Herbst <Ulrich.Herbst@dont-contact.us>
Date: Tue, 07 Apr 1998 08:02:38 +0200

betka@tri0332.urh.uiuc.edu schrieb:
>
> Okay, I've been lurking on this list for long enough, time to ask a
> question :)
>
> I've read a good portion of the docs (FAQ, Users Guide, and Release Notes)
> and I think I've decided that the squid box I've set up is paging an
> abnormal amount.
>
> Machine details:
> RedHat 5.0
> Kernel 2.0.33 with FDs patched to 4096
> 128 MB of RAM
> SCSI Disks
> 3600 MB of cache space on a single drive (about 1/3 full right now)
> OS and swap on seperate drive on same controller

My setup is similar, I have about 2000-3000 connections per hour with a
linux-machine, P-133,2.0.33,FDs patched to 3000,80MB RAM, 4 GB Cache.

I use 1.NOVM.19 (same with former versions) and I got the ratio
connections:Page faults is about 1:2.

On the other hand, this system doesn't use any swapspace (as top says).

I think, "Page faults with physical i/o" has to be interpreted somewhat
different on linux then on other systems.

Uli

-- 
-------------------------------------------------------------
Ulrich Herbst
Ulrich.Herbst@ccs-sw.debis.de
debis Systemhaus CCS
Daimlerstr. 1
76742 Woerth
Tel. ++49 7271 71 5752
Fax. ++49 7271 71 5556
Received on Mon Apr 06 1998 - 23:08:14 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:39:37 MST