[squid-users] Squid restarts when accessing some webpages

Hey There,

We are running Squid (squid-2.5.STABLE6-3.4E.9) on Centos 4.1. When
accessing some web-pages the Squid restarts itself. I can reproduce this
problem on different machines (PIII, P4, SMP) but cannot really find
anything that helps me in the logs. Here the config we use: (I turn off
the squidguard and the authentication for this testing -> it happens with
and without!)

debug_options ALL,2
#redirect_program /usr/bin/squidguard -c /etc/squid/squidguard.conf
#redirect_children 4

cache_effective_user squid
visible_hostname fw.ch.gemue.intern
http_port 8080
tcp_outgoing_address 10.10.111.203
maximum_object_size 200 MB
pipeline_prefetch on
request_body_max_size 10 MB
cache_access_log /var/log/squid/access.log
cache_log /var/log/squid/cache.log
cache_dir aufs /var/spool/squid/ 9000 16 256
cache_store_log none
cache_mem 64 MB
#auth_param ntlm program /usr/bin/ntlm_auth
--helper-protocol=squid-2.5-ntlmssp -d 4
#auth_param ntlm children 100
#auth_param ntlm max_challenge_reuses 0
#auth_param ntlm max_challenge_lifetime 30 minutes
#auth_param ntlm use_ntlm_negotiate on
#auth_param basic program /usr/bin/ntlm_auth
--helper-protocol=squid-2.5-basic
#auth_param basic children 5
#auth_param basic realm Squid proxy-caching web server
#auth_param basic credentialsttl 2 hours
#external_acl_type NT_global_group %LOGIN /usr/lib/squid/wbinfo_group.pl
#cache_mgr rr@d-con.com
cachemgr_passwd hugo all
coredump_dir /var/spool/squid
acl UBS dstdomain .ubs.com
acl ADOBE dstdomain .adobe.com
acl SUVA dstdomain .suva.ch
#acl FullAccess external NT_global_group InternetFull
#acl SurfAccess external NT_global_group InternetWWW
acl SurfAccess src 10.10.1.0/255.255.255.0
#acl AuthorizedUsers proxy_auth REQUIRED
#acl Block_Attachment url_regex -i "/etc/squid/block_attachments.cfg" #acl
Skype dst "/etc/squid/skype.ips"
acl CONNECT method CONNECT
acl QUERY urlpath_regex cgi-bin \?
acl SSL_ports port 443 563 44443
acl Safe_ports port 21 # ftp
acl Safe_ports port 443 563 44443 # https, snews
acl Safe_ports port 488 # gss-http
acl Safe_ports port 70 # gopher
acl Safe_ports port 777 # multiling http
acl Safe_ports port 80 # http
acl Messenger dstdomain .msn.com
acl Messenger dstdomain .messenger.hotmail.com
acl Messenger dstdomain .google.com
acl all src 0.0.0.0/0.0.0.0
acl localhost src 127.0.0.1/255.255.255.255
acl manager proto cache_object
hierarchy_stoplist cgi-bin ?
no_cache deny QUERY
http_access allow manager localhost
http_access deny !Safe_ports
http_access deny CONNECT !SSL_ports
http_access allow UBS
http_access allow ADOBE
http_access allow Messenger
http_access allow SUVA
http_access allow CONNECT UBS
#http_access deny !AuthorizedUsers
#http_access allow FullAccess
#http_access deny Block_Attachment
http_access allow SurfAccess
http_access deny all
http_access deny manager

Attached I have the cache.log and access.log files tared'n'zipped of just
the one request the crashes the proxy with debug=ALL,2. For testing I made
this with http://bild.de (tested with IE and Mozilla -> here it is with
Mozilla on Linux).... Har anyone experienced similar behaviour and found
some fix??? If there is more log-details needed I can easily provide it!!!

cheers.Roman