Hi, we use squid as shipped with SuSE Linux Enterprise 9 (squid-2.5.STABLE5-42.41). The throughput is slower, when I get the data from cache and is faster, when I use the -r option to get the data from the webserver. --snip-- squid06:~ # time squidclient -p 8080 -r http://mueller.datevnet.de/Richter.pdf >/dev/null real 0m0.702s user 0m0.004s sys 0m0.015s squid06:~ # time squidclient -p 8080 http://mueller.datevnet.de/Richter.pdf >/dev/null real 0m6.085s user 0m0.001s sys 0m0.002s squid06:~ # time squidclient -p 8080 -r http://mueller.datevnet.de/Richter.pdf >/dev/null real 0m0.881s user 0m0.003s sys 0m0.012s squid06:~ # time squidclient -p 8080 http://mueller.datevnet.de/Richter.pdf >/dev/null real 0m5.785s user 0m0.001s sys 0m0.003s squid06:~ # time squidclient -p 8080 -r http://mueller.datevnet.de/Richter.pdf >/dev/null real 0m0.742s user 0m0.005s sys 0m0.012s squid06:~ # time squidclient -p 8080 http://mueller.datevnet.de/Richter.pdf >/dev/null real 0m5.890s user 0m0.001s sys 0m0.002s squid06:~ # --snip-- here the logfile entries: --snip-- 1153747089.142 698 127.0.0.1 TCP_CLIENT_REFRESH_MISS/200 2274238 GET http://mueller.datevnet.de/Richter.pdf - FIRST_UP_PARENT/127.0.0.1 application/pdf 1153747098.419 6081 127.0.0.1 TCP_HIT/200 2274245 GET http://mueller.datevnet.de/Richter.pdf - NONE/- application/pdf 1153747102.320 878 127.0.0.1 TCP_CLIENT_REFRESH_MISS/200 2274238 GET http://mueller.datevnet.de/Richter.pdf - FIRST_UP_PARENT/127.0.0.1 application/pdf 1153747109.546 5782 127.0.0.1 TCP_HIT/200 2274245 GET http://mueller.datevnet.de/Richter.pdf - NONE/- application/pdf 1153747111.639 739 127.0.0.1 TCP_CLIENT_REFRESH_MISS/200 2274238 GET http://mueller.datevnet.de/Richter.pdf - FIRST_UP_PARENT/127.0.0.1 application/pdf 1153747118.797 5887 127.0.0.1 TCP_HIT/200 2274245 GET http://mueller.datevnet.de/Richter.pdf - NONE/- application/pdf --snip-- This is my squid.conf --snip-- squid06:~ # grep -v "^#" /etc/squid/squid.conf | grep -v "^$" http_port 10.252.104.20:8080 http_port 10.252.104.80:8080 http_port 127.0.0.1:8080 icp_port 0 cache_peer 127.0.0.1 parent 8280 0 no-query no-digest no-netdb-exchange hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin \? no_cache deny QUERY cache_mem 512 MB maximum_object_size 20480 KB cache_dir diskd /var/cache/squid 15360 16 256 cache_store_log none ftp_user datevnet@ ftp_list_width 50 auth_param basic children 15 auth_param basic realm DATEVnet Proxy-Server auth_param basic credentialsttl 2 hours auth_param basic program /usr/sbin/squid_ldap_auth -b ou=Accounts,dc=datevnet,dc=de -R -f (&(uid=%s)(dvPerm=WEB)) -s sub -h haldap.services.datevnet.de -p 389 -D cn=admin,ou=proxy,ou=systems,dc=datevnet,dc=de -w proxy refresh_pattern ^ftp: 1440 20% 10080 refresh_pattern ^gopher: 1440 0% 1440 refresh_pattern . 0 20% 4320 half_closed_clients off shutdown_lifetime 5 seconds acl all src 0.0.0.0/0.0.0.0 acl manager proto cache_object acl localhost src 127.0.0.1/255.255.255.255 acl to_localhost dst 127.0.0.0/8 acl SSL_ports port 443 563 acl Safe_ports port 80 # http acl Safe_ports port 21 # ftp acl Safe_ports port 443 563 # https, snews acl Safe_ports port 70 # gopher acl Safe_ports port 210 # wais acl Safe_ports port 1025-65535 # unregistered ports acl Safe_ports port 280 # http-mgmt acl Safe_ports port 488 # gss-http acl Safe_ports port 591 # filemaker acl Safe_ports port 777 # multiling http acl CONNECT method CONNECT acl Admin_ports port 1812 8445 # Viruswall-GUI acl PURGE method PURGE acl snmpread snmp_community public acl mrtg_host src 10.252.0.88/255.255.255.248 acl adminclients src 10.252.16.0/255.255.255.0 acl adminusers proxy_auth A00000330020 acl password proxy_auth REQUIRED acl transon-server dst 212.114.203.97/255.255.255.255 acl frustcenter dst 193.27.49.0/255.255.255.0 acl www_crl_esecure_de dst 193.27.50.195/255.255.255.255 acl nocacheservers dst 193.27.50.178/32 193.27.50.179/32 193.27.50.135/32 193.27.50.137/32 acl blockhostsip dst "/etc/squid/blockhosts.ip" acl blockhostsdomain dstdomain "/etc/squid/blockhosts.domain" acl cdbservers url_regex "/etc/squid/squid.cdbservers" http_access allow manager localhost http_access deny manager http_access deny !Safe_ports http_access deny CONNECT !SSL_ports http_access deny Admin_ports http_access allow PURGE localhost http_access deny PURGE http_access allow localhost http_access deny blockhostsip http_access deny blockhostsdomain http_access allow transon-server http_access allow frustcenter http_access allow www_crl_esecure_de http_access allow cdbservers http_access allow password http_access allow adminusers http_access deny all http_reply_access allow all icp_access deny all cache_effective_user squid visible_hostname squid06.services.datevnet.de unique_hostname squid06.services.datevnet.de deny_info ERR_CUSTOM_BLOCKHOSTS blockhostsip deny_info ERR_CUSTOM_BLOCKHOSTS blockhostsdomain never_direct allow all !CONNECT error_directory /usr/share/squid/errors/German_Datev snmp_access allow snmpread mrtg_host snmp_access deny all snmp_incoming_address 10.252.104.80 coredump_dir /var/cache/squid client_persistent_connections off --snip-- I'm alone on this server, which has 2G Ram and 2 Xeon 3.4 GHz CPUs. The cache_dir is a hardware raid1 with 36 GByte Space. Does anybody have an idea why the throughput is lower, when I get the data from cache instead from the webserver ? -- Gruß Dieter -- I do not get viruses because I do not use MS software. If you use Outlook then please do not put my email address in your address-book so that WHEN you get a virus it won't use my address in the From field.
Attachment:
pgp5QeshJylK4.pgp
Description: PGP signature
