Re: [squid-users] 4 second Cache Miss Service Times

Ben Hathaway wrote:

>Dear All,
>
> I have recently set up a Squid cache using WCCP and a cisco router.
>I am getting very impressive performance for my cache hits, but my cache
>misses sometimes take as long as 4 seconds! We are at the end of a high
>bandwidth, high latency satellite link with a normal latency (for example:
>to google.com) of around 600ms round-trip. So why the 4 second delay?
>Sometimes it comes down to 1.3sec but even that is a lot slower than I would
>expect when my cache hits are being pumped out in 0.017s
>
> Any ideas?
>
> Here's my cachemanager stats (I've marked the relevant line with an
>asterisk) :
>
>Squid Object Cache: Version 2.5.STABLE9
>Start Time: Thu, 15 Jun 2006 08:43:36 GMT
>Current Time: Fri, 16 Jun 2006 07:14:09 GMT
>
>
SNIP

>Median Service Times (seconds) 5 min 60 min:
> HTTP Requests (All): 1.62803 1.71839
>** Cache Misses: 3.11263 2.79397
> Cache Hits: 0.01745 0.01745
> Near Hits: 1.38447 1.24267
> Not-Modified Replies: 0.01164 0.01164
> DNS Lookups: 0.01535 0.02033
> ICP Queries: 0.00000 0.00000
>
>
>
SNIP

>
>Regards,
>
>Ben Hathaway
>
>
For what it's worth, both your misses and Near Hits are double mine
(similar situation). Your DNS requests on the other hand are much lower
(my Squid server runs a caching DNS server as well, and DNS requests
show about 600ms). What is the output of "time wget -q -Y off -O
/dev/null http://www.google.com/intl/en/images/logo.gif"? This tells
wget to fetch the content quietly, without using a proxy, and to save
the file to /dev/null. Try it a few times to get a good sample set.

Perhaps the time required to make DNS queries is being included in the
Cache Miss and Near Hits time due to using WCCP (my clients are set
explicitly to use the proxy).

Chris
Received on Fri Jun 16 2006 - 14:43:37 MDT