[squid-users] Squid gets really slow using SquidGuard filtering

From: Ismael Milach da Silveira <ismael@dont-contact.us>
Date: Tue, 19 Dec 2006 16:28:46 -0200

Hi guys,

I'm having a problem here, the subject says it all, when I activate the
filtering rules, the cache performance goes down...

When I use:
####################
default {
                pass all allowed !porn
            }
###################

It works OK... of course, squid won't filter anything...

using..
###########################
default {
                pass allowed !porn all
            }
############################

The filtering goes up but the response time gets REALLY slow, impossible to
use the web, most of the time only half the page requested opens.

I have squid+squidGuard running here on dozens of servers with the same
config without any problem... My first guess is that It's something related
with hardware performance, but top doesn't show anything abnormal, and I
tried to get some files from the server using FTP without any problem at
~100Mbps.

I've also ran badblocks and fsck without any errors.

########################################
press:/etc/squid# cat /etc/debian_version
3.1

press:/etc/squid# uname -r
2.6.12-doctor-p2p

press:/etc/squid# squid -v
Squid Cache: Version 2.5.STABLE9
configure
options: --prefix=/usr --exec_prefix=/usr --bindir=/usr/sbin --sbindir=/usr/sbin
 --libexecdir=/usr/lib/squid --sysconfdir=/etc/squid --localstatedir=/var/spool/squid
 --datadir=/usr/share/squid --enable-async-io --with-pthreads --enable-storeio=ufs,aufs,diskd,null
 --enable-linux-netfilter --enable-arp-acl --enable-removal-policies=lru,heap
 --enable-snmp --enable-delay-pools --enable-htcp --enable-poll --enable-cache-digests
 --enable-underscores --enable-referer-log --enable-useragent-log --enable-auth=basic,digest,ntlm
 --enable-carp --with-large-files i386-debian-linux

>>> top - 16:24:15 up 12 days, 21:38, 13 users, load average: 0.13, 0.21,
>>> 0.25

press:/home/doctor# free -m
             total used free shared buffers cached
Mem: 885 836 48 0 42 482
-/+ buffers/cache: 311 573
Swap: 2651 0 2651

press:/home/doctor# cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 6
model : 7
model name : AMD Duron(tm) Processor
stepping : 1
cpu MHz : 1303.036
cache size : 64 KB

<<snipped>>>

press:/home/doctor# cat /etc/squid/squid.conf
hierarchy_stoplist cgi-bin ?
acl QUERY urlpath_regex cgi-bin \?
no_cache deny QUERY
cache_mem 128 MB
cache_dir ufs /var/spool/squid 300 16 256
#cache_dir /dev/null
logfile_rotate 1
<<snipped>>
########################################

Again, it's basically the same overall config/environment I use everywhere
here without any problem.

The ACL's are short ones, less than 100 lines each...

cache.log and squidguard.log don't show any errors...

Any ideas?

Thanks!
Ismael
Received on Tue Dec 19 2006 - 11:28:59 MST

This archive was generated by hypermail pre-2.1.9 : Mon Jan 01 2007 - 12:00:01 MST