We have a fast connection and a cache server running gentoo linux with
dual 2.8gig xeon processors and 3 gigs of ram. FS is reiserfs, config
file is below
The cache works well on regular small to medium size. However, when we
try to download a large file, it starts out at about 1.7MB/sec, and
slowly eventually works its way down to 200KB/sec (as the dl speed
drops on the client, the processor usage on the server approaches 100%
of 1 processor) . When we bypass the caching server completely we
generally get a sustained 2.4MB/sec.
Any Ideas why large requests act this way?
Thanks,
--jeremy
Southgate Schools
http_port 3128
http_port 80
hierarchy_stoplist cgi-bin ? s.js
acl QUERY urlpath_regex cgi-bin \? s.js t.aspx
acl EDPERF dstdomain test.edperformance.com
cache_mem 1500 MB
cache_swap_low 96
cache_swap_high 98
maximum_object_size_in_memory 200 MB
cache_dir diskd /squidcache/cache 18000 48 256
acl all src 10.81.0.0-10.254.0.0/255.255.0.0
acl manager proto cache_object
acl localhost src 127.0.0.1/255.255.255.255
acl SSL_ports port 443 563
acl Safe_ports port 80-81
acl Safe_ports port 21
acl Safe_ports port 443 563
acl Safe_ports port 70
acl Safe_ports port 210
acl Safe_ports port 1025-65535
acl Safe_ports port 280
acl Safe_ports port 488
acl Safe_ports port 591
acl Safe_ports port 777
acl CONNECT method CONNECT
http_access allow manager localhost
http_access allow manager
http_access allow Safe_ports
http_access deny CONNECT !SSL_ports
http_access allow localhost
http_access deny all
icp_access allow all
acl edperformance dstdomain test.edperformance.com
no_cache deny QUERY EDPERF
always_direct allow edperformance
visible_hostname squid.sgate.k12.mi.us
maximum_object_size 350 MB
cache_effective_user squid
cache_effective_group squid
dns_nameservers 127.0.0.1
redirect_program /usr/bin/squidGuard -c /etc/squid/squidGuard.conf
redirect_children 10
Received on Mon Oct 11 2004 - 14:04:42 MDT
This archive was generated by hypermail pre-2.1.9 : Mon Nov 01 2004 - 12:00:01 MST