Re: [SQU] Page faults & mem usage

From: Henrik Nordstrom <[email protected]>
Date: Wed, 14 Feb 2001 21:09:06 +0100

10216 KB is nothing for a box with 328megs. Do you at all have a
configured cache on this Squid?

Please uignore the % value there. For a start it is a % of the current
Squid process size, not the amount of memory in your server, and
secondly it is calculated wrongly (should be 100%, as 10217 KB of 10235
KB is in use inside the Squid process).

Regards
Henrik Nordstrom
Squid hacker

David Wilson wrote:
>
> Hi guys,
>
> I'm running a Linux 2.2.18 box with +- 328megs of RAM, it is a transparent
> proxy for about 1000 dialup users and 40 leased line clients.
> I seem to be getting way too many page faults, i.e. page faults increase at
> about 70 a minute... ?
> Please look at this output from the cache manager and tell me what you
> think:
> Page faults with physical i/o: 1167
> Memory usage for squid via mallinfo(): Total space in arena: 10059 KB
> Ordinary blocks: 10040 KB 8 blks
> Small blocks: 0 KB 0 blks
> Holding blocks: 176 KB
> 1 blks Free Small blocks: 0 KB Free
> Ordinary blocks: 18 KB
> Total in use: 10216 KB 102%
> Total free: 18 KB 0%
> My cache_mem is set to: 15Megs
> I'm particularly interested in the Total in use: 10216 KB 102%
> This is the cause of all the page faults?
> Any guidance would be most appreciated.
> Thanks.
>
>
> Kindest regards
> David Wilson
> The S.A Internet
>
> ---------------------------------------------------------------
>
> Name: winmail.dat
> Part 1.2 Type: application/ms-tnef
> Encoding: base64

--
To unsubscribe, see http://www.squid-cache.org/mailing-lists.html
Received on Wed Feb 14 2001 - 13:58:59 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:58:00 MST