Hi
> The reason?
> Good question.
> I just felt that since we support only 200 users and we average
> approximately 500 connections per hour, response should be a little more
> faster than it is currently. I am not complaining about the performance we
> are getting now, but I thought that I would throw this out to see if maybe
> I set something up wrong by using the wrong OS. My main concern is the
> delay in retrieving a page. If we have a couple of users on the Squid,
> requests are gathered quite fast. Once we begin adding another user and
> then another, delays begin to last a good minute. To me, it sounds like the
Something is wrong with your setup if that is the case...
We run some machines here with linux-2.0.30+filehandle-patch and squid-NOVM
getting >500 connections a minute... they have no problems like this
Firstly - when the machine is loaded - try running "vmstat 1"
procs memory swap io system cpu
r b w swpd free buff cache si so bi bo in cs us sy id
is the machine swapping at all?
Is the CPU completely used
What % is spent in system time?
If you look at the cachemgr output under "dns server statistics" is the
last dns server every being used? If so it's probably the cause for lag...
Also - do your ACL's use "acl src domain.com"? Don't use this as it
does an extra dns lookup for every request...
Hope that this helps
Oskar
Received on Thu Jun 19 1997 - 08:13:01 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:35:33 MST