Squid Weekly Summary Interpretation

I have been using Squid for roughly 5 years, perhaps a little longer. Late
last week, a customer asked how to interpret the information presented in
the Squid Weekly Summary.

To ensure that I didn't mislead the customer, I went to WWW.SQUID-CACHE.ORG
to verify the accuracy of what I was going to tell him. Unfortunately, I
have either forgotten where the discussion of the Squid Weekly Summary was
located or it was removed when the SNMP capabilities were added.

Are the following statements reasonably accurate?

COUNTS

(1) The value in the Counts column is the total number of client requests
for the listed object class or type.

(2) The value in the %all column is the percentage of all client requests
for all object classes or types that the value in the Counts column
represents.

(3) The value in the %hit column is the percentage of the client requests
displayed in the Counts column that were satisfied from cache.

MBYTES

(1) The value in the MByte column is the total number of bytes transferred
to a client for the listed object class or type.

(2) The value in the %all column is the percentage of the total number of
bytes transferred to clients regardless of object class or type that the
value in the MByte column represents.

(3) The value in the %hit column is the percentage of the value displayed in
the MByte column that were transferred from cache.

If the client sends an "if-modified-since" request what is the impact on the
above, i.e. is the request quietly ignored if the cache has a "fresh" copy
of the object? Or, would the request count be incremented and the request
be treated as a "hit"?

I have some client objects that have 60% or higher "hit" rates on their
request counts. Can an impatient user "artificially" inflate the %hit value
by frequently refreshing the display? Or, does this really indicate that
the user is frequently accessing a finite set of objects that change
infrequently?

Merton Campbell Crockett
Received on Sat Jun 24 2000 - 11:42:45 MDT