[squid-users] refresh_pattern explanation wanted

7Hello,

I'm trying to understand refresh_pattern a bit more.

Reading comments config file didn't help much, and the SQUID configuration
manual at visolve.com is 1. buggy (unsecaped '>' and '<' chars makes it
unreadable, reported already) and 2. doesn't contain more informations.

I currently miss the 'percent' number:

# usage: refresh_pattern [-i] regex min percent max [options]
[...]
# 'Percent' is a percentage of the objects age (time since last
# modification age) an object without explicit expiry time
# will be considered fresh.

percent of what time? percens of "max" time? or does it mean thar lm-factor
thing below? (should be mentioned in the default manual imho)

and also:

# Basically a cached object is:
#
# FRESH if expires < now, else STALE
                         ^^^^^^^^^^^^^
shouldn't that be "expires > now"? an object is fresh, if it will expire in
the future, not if it already expired, right? (a bug in the doc?)

another strange thing: the lm-factor is explained on
http://www.squid-cache.org/Doc/FAQ/FAQ-12.html#ss12.20

    # OBJ_DATE is the time when the object was given out by the origin server.
    This is taken from the HTTP Date reply header.
    # OBJ_LASTMOD is the time when the object was last modified, given by the
    HTTP Last-Modified reply header.

    # OBJ_AGE is how much the object has aged since it was retrieved:

        OBJ_AGE = NOW - OBJ_DATE

- It it really calculated from current local date, and Date: from object
header? Does squid mix local Date and remote servers' Date or is this part
of configuration incorrect and squid counts current local date and locatl
date when the object was fetched?

    # LM_AGE is how old the object was when it was retrieved:

        LM_AGE = OBJ_DATE - OBJ_LASTMOD

    # LM_FACTOR is the ratio of OBJ_AGE to LM_AGE:

        LM_FACTOR = OBJ_AGE / LM_AGE

- the same as above.

-- 
Matus UHLAR - fantomas, uhlar@fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
If Barbie is so popular, why do you have to buy her friends?