Thank you, Amos. Taking into account your and Rafael's recommendations, I configured HAProxy and Squid to use the PROXY protocol instead of reformatting the messages. At the moment I disabled authentication, due to internal requirements. I had a hard time dealing with the HAProxy health checks, but I was able to fix it. However, by configuring Squid in this way, I had a last problem that I didn't expect: Squid reports the client's IP to my internet gateway instead of their own IP. It's true, I want to see my clients' IPs in Squid's log; but I want Squid's IP to reach my gateway and not my clients'. This way I can make my clients browse internet only through my proxies.
I'm really not sure if continuing with this configuration I'm doing will be possible to achieve it. I appreciate in advance any indication you can give me. haproxy.cfg global log /dev/log local0 log /dev/log local1 notice chroot /var/lib/haproxy stats socket /run/haproxy/admin.sock mode 660 level admin expose-fd listeners stats timeout 30s user haproxy group haproxy daemon maxconn 4000 ca-base /etc/ssl/certs crt-base /etc/ssl/private ssl-default-bind-ciphers ECDH+AESGCM:DH+AESGCM:ECDH+AES256:DH+AES256: ECDH+AES128:DH+AES:RSA+AESGCM:RSA+AES:!aNULL:!MD5:!DSS ssl-default-bind-options no-sslv3 defaults log global mode tcp option tcplog option dontlognull timeout connect 5000 timeout client 50000 timeout server 50000 frontend squid_in bind *:3128 default_backend squid_pool backend squid_pool balance source mode tcp server px1 x.x.x.1:3128 check port 8181 send-proxy inter 2000 rise 2 fall 3 server px2 x.x.x.2:3128 check port 8181 send-proxy inter 2000 rise 2 fall 3 squid.conf acl localnet src 192.168.12.1-192.168.13.254 # my clients IP's acl localnet src 192.168.11.80 # haproxy IP 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 8181 # haproxy health checks port acl CONNECT method CONNECT # list of allowed domains acl LS_whitedomains dstdomain "/etc/squid/acl/whitedomains.txt" # list of bloqued domains acl LS_blackdomains dstdomain "/etc/squid/acl/blackdomains.txt" acl LS_malicius dstdomain "/etc/squid/acl/malicius.txt" acl LS_porn dstdomain "/etc/squid/acl/porn.txt" # Deny requests to certain unsafe ports http_access deny !Safe_ports # Deny CONNECT to other than secure SSL ports http_access deny CONNECT !SSL_ports # Only allow cachemgr access from localhost http_access allow localhost manager http_access deny manager # We strongly recommend the following be uncommented to protect innocent web applications running on the proxy server who think the only one who can access services on "localhost" is a local user #http_access deny to_localhost # # INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS # # implementation of access list policies http_access deny LS_blackdomains http_access allow LS_whitedomains http_access deny LS_malicius http_access deny LS_porn # limit downloads to 10 Mb/s to the localnet network delay_pools 1 delay_class 1 2 delay_parameters 1 1310720/1966080 917504/1310720 delay_access 1 allow localnet # implementation of core access policies proxy_protocol_access allow localnet http_access allow localnet http_access allow localhost # And finally deny all other access to this proxy http_access deny all # Squid normally listens to port 3128 http_port 3128 require-proxy-header # port used only for haproxy health checks http_port 8181 # Leave coredumps in the first cache dir coredump_dir /var/spool/squid # Add any of your own refresh_pattern entries above these. 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 # squid customization settings quick_abort_min 0 KB quick_abort_max 0 KB read_timeout 5 minutes request_timeout 3 minutes shutdown_lifetime 0 seconds ipcache_size 2048 fqdncache_size 4096 cache_mgr m...@tsa.net visible_hostname px1 httpd_suppress_version_string on forwarded_for off Best regards Gabriel El vie., 9 de ago. de 2019 a la(s) 03:45, Amos Jeffries ( squ...@treenet.co.nz) escribió: > On 9/08/19 1:44 am, neok wrote: > > Hi, I finally did the configuration differently. It's working very well > for > > me. > > > > What you are doing is polluting every HTTP message with two new headers. > > The way Rafael suggested is more efficient since the PROXY protocol > details are only delivered at the start of each TCP connection, and HTTP > messages do not need to be reformatted at the LB as they pass. > > Note that Squid will be extending that XFF header itself anyway. But > with your way the LB IP address will be broadcast to the origin(s) in > the XFF header along with the client address. With PROXY protocol Squid > adds only the client address there. > > Amos > _______________________________________________ > squid-users mailing list > firstname.lastname@example.org > http://lists.squid-cache.org/listinfo/squid-users >
_______________________________________________ squid-users mailing list email@example.com http://lists.squid-cache.org/listinfo/squid-users