After rebuilding the rpm without the with-maxfd=16384 and installing it in two very different servers the are "32768 file descriptors available" for Squid in each server. No idea why there are no more file descriptors available. The SO config seems to be correct with regards to file descriptors available in both servers. Here is an example of one of them: [root@anything ]# cat /proc/sys/fs/file-max 100451 [root@anything ]# ulimit -Hn 65535 [root@anything ]# ulimit -Sn 65535 [root@anything ]# cat /proc/sys/fs/file-max 100451 [root@anything ]# sysctl fs.file-max fs.file-max = 100451 [root@anything ]# service squid stop Stopping squid: . [ OK ] [root@anything ]# su - squid This account is currently not available. [root@anything ]# service squid start Starting squid: . [ OK ] [root@anything ]# sysctl fs.file-nr fs.file-nr = 1152 0 100451 Also max_filedesc 98304 is set at the end of squid.conf which clearly works because when removed there are only 1026 file descriptors or so. ulimit -HSn 98304 is also added to the beginning of vi /etc/init.d/squid As you can see now there is no with-maxfd=16384 when squid -v is used: Squid Cache: Version 2.6.STABLE21 configure options: '--host=x86_64-redhat-linux-gnu' '--build=x86_64-redhat-linux-gnu' '--target=x86_64-redhat-linux' '--program-prefix=' '--prefix=/usr' '--exec-prefix=/usr' '--bindir=/usr/bin' '--sbindir=/usr/sbin' '--sysconfdir=/etc' '--includedir=/usr/include' '--libdir=/usr/lib64' '--libexecdir=/usr/libexec' '--sharedstatedir=/usr/com' '--mandir=/usr/share/man' '--infodir=/usr/share/info' '--exec_prefix=/usr' '--bindir=/usr/sbin' '--libexecdir=/usr/lib64/squid' '--localstatedir=/var' '--datadir=/usr/share' '--sysconfdir=/etc/squid' '--enable-arp-acl' '--enable-epoll' '--enable-snmp' '--enable-removal-policies=heap,lru' '--enable-storeio=aufs,coss,diskd,null,ufs' '--enable-ssl' '--with-openssl=/usr/kerberos' '--enable-delay-pools' '--enable-linux-netfilter' '--with-pthreads' '--enable-ntlm-auth-helpers=SMB,fakeauth' '--enable-external-acl-helpers=ip_user,ldap_group,unix_group,wbinfo_group' '--enable-auth=basic,digest,ntlm,negotiate' '--enable-negotiate-auth-helpers=squid_kerb_auth' '--enable-digest-auth-helpers=password' '--with-winbind-auth-challenge' '--enable-useragent-log' '--enable-referer-log' '--disable-dependency-tracking' '--enable-cachemgr-hostname=localhost' '--enable-underscores' '--enable-basic-auth-helpers=LDAP,MSNT,NCSA,PAM,SMB,YP,getpwnam,multi-domain-NTLM,SASL' '--enable-cache-digests' '--enable-ident-lookups' '--enable-follow-x-forwarded-for' '--enable-wccpv2' '--enable-fd-config' 'build_alias=x86_64-redhat-linux-gnu' 'host_alias=x86_64-redhat-linux-gnu' 'target_alias=x86_64-redhat-linux' 'CFLAGS=-D_FORTIFY_SOURCE=2 -fPIE -Os -g -pipe -fsigned-char' 'LDFLAGS=-pie' The only thing I can think of to try is to rebuild again the rpm but with with-maxfd=98304 (instead of simply removing with-maxfd=16384). Also I will probably try soon with a more recent version of Squid (because of the better performance and in order to see whether that limit of 32768 disappears or not). Any ideas? Thank you in advance -- View this message in context: http://squid-web-proxy-cache.1019090.n4.nabble.com/About-bottlenecks-Max-number-of-connections-etc-tp4658650p4658732.html Sent from the Squid - Users mailing list archive at Nabble.com.
