high values in cachemgr.

From: Marcelo M. Sosa Lugones <marcelo@dont-contact.us>
Date: Thu, 12 Aug 1999 14:00:25 -0300 (ART)

Hello,

i think i have some high values in the cache manager, can someone tell me
if it is normal or not?

Just for info:
Connection information for squid:
        Number of clients accessing cache: 766
        Number of HTTP requests received: 680686
        Number of ICP messages received: 775909
        Number of ICP messages sent: 776440
        HTTP requests per minute: 81.1
        ICP messages per minute: 185.0
        Select loop called: 1331090825 times, 0.378 ms avg

The times are a bit high, aren't they? i notice some delay (about 2-3
seconds) before starting to download the page from any client browser and
then the page comes very fast. i am using delay pools, can it be part of
the problem? also the dns lookup is very high, i think.
Median Service Times (seconds) 5 min 60 min:
        HTTP Requests (All): 0.94847 0.00000
        Cache Misses: 1.91442 0.94847
        Cache Hits: 0.00000 0.00000
        Near Hits: 0.94847 0.94847
        Not-Modified Replies: 0.00000 0.00000
        DNS Lookups: 0.96589 0.00000
        ICP Queries: 0.00000 0.00000

Too much cpu usage? it is a P100 box.
The page fault are right? or too many?
Resource usage for squid:
        UP Time: 503388.240 seconds
        CPU Time: 155156.450 seconds
        CPU Usage: 30.82%
        CPU Usage, 5 minute avg: 33.46%
        CPU Usage, 60 minute avg: 31.59%
        Page faults with physical i/o: 204448

Also this one, i have 2 siblings in the lan, the 2 boxs are linux 2.2.10,
like this one, same config, and i think that avg rtt is very high, they
are all of them in a fast ethernet switch...
Sibling : squid1.sosa.com.ar/3128/3130
AVG RTT : 951 msec
Sibling : squid3.sosa.com.ar/3128/3130
AVG RTT : 869 msec

Thanks,
-=marcelo=-
Marcelo M. Sosa Lugones
Received on Thu Aug 12 1999 - 10:35:44 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:48:01 MST