[squid-users] Extreme slow loading of one web site

Hello

We are in the progress to replace our old Squid 2.4 caches with new
ones. We testet on Ubuntu and Gentoo with Squid 2.7.x. Our setup
includes a local Squid and a parent Squid in the DMZ where the local
Squid connects to the parent with ICP. The parent Squid also serves the
clients of ist subnet diretly. Everything is working fine with the new
Squid, except one web site. When connecting to

http://www.buero-schoch.ch

the site starts loading, then the loading get stalled and after some
minutes the site is loaded completly. In the logs, I couldn't find
anything special. I used WGET to retrieve the headers:

On the old Squid 2.4:
HTTP request sent, awaiting response...
 1 HTTP/1.1 200 OK
 2 Content-Type: text/html;charset=iso-8859-1
 3 Set-Cookie: fe_typo_user=fcdc25dec3; path=/
 4 Server: Oracle-Application-Server-10g/9.0.4.0.0 Oracle-HTTP-Server
OracleAS-Web-Cache-10g/9.0.4.0.0 (N)
 5 Connection: Close
 6 Date: Fri, 27 Mar 2009 17:01:13 GMT
 7 X-Powered-By: PHP/4.3.10
 8 X-Pad: avoid browser bug
200 OK

On the new Squid 2.7:
HTTP request sent, awaiting response...
  HTTP/1.1 200 OK
  Content-Type: text/html;charset=iso-8859-1
  Set-Cookie: fe_typo_user=99d58268d6; path=/
  Server: Oracle-Application-Server-10g/9.0.4.0.0 Oracle-HTTP-Server
OracleAS-Web-Cache-10g/9.0.4.0.0 (N)
  Connection: Close
  Date: Mon, 30 Mar 2009 05:40:16 GMT
  X-Powered-By: PHP/4.3.10
Length: unspecified [text/html]

Here is a difference for "X-Pad: avoid browser bug", but Squid shouldn't
be a Netscape Navigator...

We have this behavior with both Squids of version 2.7 (Ubuntu and
Gentoo). The problem seems to be with the version 2.7. It does not
depend if I connect over the child -> parent squid or make a direct
connect to the parent Squid from its subnet.

Here is the configuration of the parent Squid, where the localnet can
use it directly too.

On Gentoo with
net-proxy/squid-2.7.6 USE="logrotate pam ssl"

------------------------------------------------------------
/etc/squid/squid.conf:
----------------------
acl all src all
acl manager proto cache_object
acl localhost src 127.0.0.1/32
acl to_localhost dst 127.0.0.0/8
acl localnet src <LOCAL_NET>/27
acl icpchilds src <CHILD_1>/32
acl icpchilds src <CHILD_2>/32
acl SSL_ports port 443
acl Safe_ports port 80 # http
acl Safe_ports port 21 # ftp
acl Safe_ports port 443 # https
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 Safe_ports port 901 # SWAT
acl purge method PURGE
acl CONNECT method CONNECT
http_access allow manager localhost
http_access deny manager
http_access allow purge localhost
http_access deny purge
http_access deny !Safe_ports
http_access deny CONNECT !SSL_ports
http_access allow localnet
http_access allow icpchilds
http_access allow localhost
http_access deny all
icp_access allow icpchilds
icp_access deny all
http_port <LOCAL_IP>:8080
hierarchy_stoplist cgi-bin ?
cache_mem 128 MB
cache_dir aufs /var/cache/squid 1024 16 256 access_log
/var/log/squid/access.log squid
refresh_pattern ^ftp: 1440 20% 10080
refresh_pattern ^gopher: 1440 0% 1440
refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
refresh_pattern . 0 20% 4320
acl shoutcast rep_header X-HTTP09-First-Line ^ICY.[0-9]
upgrade_http0.9 deny shoutcast
acl apache rep_header Server ^Apache
broken_vary_encoding allow apache
visible_hostname prx02
icp_port 3130
udp_incoming_address <LOCAL_IP>
forwarded_for off
coredump_dir /var/cache/squid
------------------------------------------------------------

Any ideas where the problem is ? Anyone out there who encountered the
same problem ?

Sincerly
Andreas Bollhalder-Bello
Received on Mon Mar 30 2009 - 05:57:00 MDT