Topic: Squid Proxy Server - Slows down over the Day

[shawnbishop]

Good Day

We have configured a Squid Proxy Server that should be proxing internet connections for a number of users, say about 200. Over a period of time the server slows down until it comes to a standstill, and I can only reset it by rebooting the server, this happens on a daily basis.

I decided to install SystemMonitor to see if the hardware wasnt coping etc..

I then saw the following was happening on my external NIC..see attached image

http://i916.photobucket.com/albums/ad10/penguinsolutions/Eth0-Report.png

The following is my squid.conf file

hierarchy_stoplist cgi-bin ?
acl QUERY urlpath_regex cgi-bin \?
no_cache deny QUERY
cache_mem 2048 MB
maximum_object_size 1280096 KB
cache_dir ufs /var/spool/squid 3000 16 256

# caching windows update various windows versions
refresh_pattern http://.*\.windowsupdate\.microsoft\.com/ 0 80% 20160 reload-into-ims
refresh_pattern http://.*\.update\.microsoft\.com/ 0 80% 20160 reload-into-ims
refresh_pattern http://download\.microsoft\.com/ 0 80% 20160 reload-into-ims
refresh_pattern http://windowsupdate\.microsoft\.com/ 0 80% 20160 reload-into-ims
refresh_pattern http://office\.microsoft\.com/ 0 80% 20160 reload-into-ims

# cache ubuntu updates [check logs use COUNTRY SPECIFIC first line or generic below]
#refresh_pattern http://.*\.archive\.ubuntu\.com  0 80% 20160 reload-into-ims 
#refresh_pattern http://archive\.ubuntu\.com  0 80% 20160 reload-into-ims

http_port 192.168.253.100:3128
http_port 127.0.0.1:3128
udp_incoming_address 192.168.253.100
udp_outgoing_address 0.0.0.0

acl all src 0.0.0.0/0.0.0.0
acl manager proto cache_object
acl localsrc src 127.0.0.1 192.168.253.0/255.255.255.0

acl SSL_ports port 443 563
acl Safe_ports port 21 70 80 81 119 210 443 563 980 1024-65535
acl CONNECT method CONNECT
acl webdav method PROPFIND TRACE PURGE PROPPATCH MKCOL COPY MOVE LOCK UNLOCK

pid_filename /var/log/squid/squid.pid
logfile_rotate 0
append_domain .wifidirect.co.za
cache_mgr admin@wifidirect.co.za
ftp_user nobody@wifidirect.co.za
extension_methods REPORT MERGE MKACTIVITY CHECKOUT
http_access allow manager localsrc
http_access deny manager


http_access deny CONNECT !SSL_ports
http_access allow localsrc
http_access deny all

redirect_program /usr/bin/squidGuard
redirect_children 500



httpd_accel_host virtual
httpd_accel_with_proxy on
httpd_accel_uses_host_header on
icp_access allow all
miss_access allow all
store_avg_object_size 64 KB
cache_dir ufs /var/spool/squid 4096 32 512
always_direct allow webdav
always_direct allow all

I have also looked through the /var/log/messages and other squid log files with no explanation or error??

Anybody got any ideas??

Shawn

[byte]

What is the hardware you are running (including RAM) ?

[shawnbishop]

Good day

Intel Quad Core CPU
4GB RAM
Gb NIC
200GB 7200 HDD

[byte]

What's your memory load like just before it stalls ? Next time it appears to stall open a putty window and type:

ps fax

then report back the results.

Also what version of SME Server are you using ? any custom-templates you've added ?

[Stefano]

is there anything interesting in the logs?

[byte]

If you have 200 users connected at the same time then you might find redirect_children 500 is not enough to handle the requests.

[shawnbishop]

I have looked at all the log files and no errors, I am going to remove the SquidGuard and see if that changes anything

[shawnbishop]

Hi

I removed squidguard and it seems to be stable now..

Now to find out why the SquidGuard seems to have caused it, now custom templates installed either