Hi,
I have squid 2.6.STABLE12 on a Gentoo boox (2.6.19-gentoo-r9).
My swap space is configured as follows:
cache_dir ufs /var/cache/squid 3000 16 256
cache_dir ufs /var/cache/squid2 20000 16 256
Those directories map on two logical devices which belong to two
different volume groups. The logical devices reside on two phisically
different raid-1 chains.
df -h gives:
/dev/mapper/vg-dati2
4,0G 2,2G 1,9G 55% /var/cache/squid
/dev/mapper/vg2-squidcache
30G 8,3G 22G 28% /var/cache/squid2
As you can see, the first cache_dir is 4G in size, and squid is
configured to use about 3G of it, but only 2.2G are used.
The other cache_dir is 30G in size, and squid has a size setting of
approximately 20G, but the cache usage is only 8.3G.
Those figures have been more or less constant for some weeks now.
The high amount of free space seems odd to me, and a waste also :-) I
think I could get a better byte hit ratio if squid would make use of
more swap space...
Other useful bits from my config file are:
# cache_swap_low 90
# cache_swap_high 95
maximum_object_size 810000 KB
cache_mem 128 MB
Thanks in advance for any tips.
--
Marcello Romani
Responsabile IT
Ottotecnica s.r.l.
http://www.ottotecnica.com
