Netgate Discussion Forum
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Search
    • Register
    • Login

    [SOLVED] HaProxy forward client IP

    Scheduled Pinned Locked Moved Cache/Proxy
    11 Posts 4 Posters 15.5k Views
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • manjotscM
      manjotsc
      last edited by manjotsc

      Hi,

      I am running haproxy -> (192.168.80.90) on pfsense -> (192.168.80.1), In frontend I have checked the “Use “forwardfor” option.” But instead haproxy forwards the 192.168.80.1 address, instead of the clients ip.

      Look for frontend FrontEndProxy

      Thanks,

      # Automaticaly generated, dont edit manually.
      # Generated on: 2020-12-31 02:42
      global
      	maxconn			1000
      	log			/var/run/log	local0	info
      	stats socket /tmp/haproxy.socket level admin  expose-fd listeners
      	uid			80
      	gid			80
      	nbproc			1
      	nbthread			1
      	hard-stop-after		15m
      	chroot				/tmp/haproxy_chroot
      	daemon
      	tune.ssl.default-dh-param	2048
      	server-state-file /tmp/haproxy_server_state
      
      listen HAProxyLocalStats
      	bind 127.0.0.1:2200 name localstats
      	mode http
      	stats enable
      	stats admin if TRUE
      	stats show-legends
      	stats uri /haproxy/haproxy_stats.php?haproxystats=1
      	timeout client 5000
      	timeout connect 5000
      	timeout server 5000
      
      frontend FrontEndProxy
      	bind			192.168.80.90:443 name 192.168.80.90:443   ssl crt-list /var/etc/haproxy/FrontEndProxy.crt_list  
      	mode			http
      	log			global
      	option			socket-stats
      	option			http-keep-alive
      	option			forwardfor
      	acl https ssl_fc
      	http-request set-header		X-Forwarded-Proto http if !https
      	http-request set-header		X-Forwarded-Proto https if https
      	timeout client		30000
      	acl			speed	var(txn.txnhost) -m str -i speed.manjot.net
      	acl			auth	var(txn.txnhost) -m str -i auth.manjot.net
      	acl			jellyfin	var(txn.txnhost) -m str -i jellyfin.manjot.net
      	acl			aclcrt_FrontEndProxy	var(txn.txnhost) -m reg -i ^speed\.manjot\.net(:([0-9]){1,5})?$
      	acl			aclcrt_FrontEndProxy	var(txn.txnhost) -m reg -i ^auth\.manjot\.net(:([0-9]){1,5})?$
      	acl			aclcrt_FrontEndProxy	var(txn.txnhost) -m reg -i ^jellyfin\.manjot\.net(:([0-9]){1,5})?$
      	http-request set-var(txn.txnhost) hdr(host)
      	use_backend Speed_ipvANY  if  speed aclcrt_FrontEndProxy
      	use_backend Auth_ipvANY  if  auth aclcrt_FrontEndProxy
      	use_backend Jellyfin_ipvANY  if  jellyfin aclcrt_FrontEndProxy
      
      frontend HTTP-TO-HTTPS
      	bind			192.168.80.90:80 name 192.168.80.90:80   
      	mode			http
      	log			global
      	option			http-keep-alive
      	option			forwardfor
      	acl https ssl_fc
      	http-request set-header		X-Forwarded-Proto http if !https
      	http-request set-header		X-Forwarded-Proto https if https
      	timeout client		30000
      	http-request redirect scheme https 
      
      frontend GMFrontEndProxy
      	bind			192.168.80.39:443 name 192.168.80.39:443   ssl crt-list /var/etc/haproxy/GMFrontEndProxy.crt_list  
      	mode			http
      	log			global
      	option			http-keep-alive
      	timeout client		30000
      	acl			gm	var(txn.txnhost) -m str -i gm.manjot.net
      	acl			aclcrt_GMFrontEndProxy	var(txn.txnhost) -m reg -i ^gm\.manjot\.net(:([0-9]){1,5})?$
      	http-request set-var(txn.txnhost) hdr(host)
      	use_backend GM_ipvANY  if  gm aclcrt_GMFrontEndProxy
      
      frontend GM-HTTP-TO-HTTPS
      	bind			192.168.80.39:80 name 192.168.80.39:80   
      	mode			http
      	log			global
      	option			http-keep-alive
      	timeout client		30000
      	http-request redirect scheme https 
      
      backend Speed_ipvANY
      	mode			http
      	id			100
      	log			global
      	timeout connect		30000
      	timeout server		30000
      	retries			3
      	option			httpchk OPTIONS / 
      	server			speed 192.168.80.74:443 id 101 ssl check-ssl check inter 1000  verify none crt /var/etc/haproxy/server_clientcert_5f5d1c960d78a.pem 
      
      backend Auth_ipvANY
      	mode			http
      	id			104
      	log			global
      	timeout connect		30000
      	timeout server		30000
      	retries			3
      	option			httpchk OPTIONS / 
      	server			Auth 192.168.80.73:443 id 105 ssl check-ssl check inter 1000  verify none crt /var/etc/haproxy/server_clientcert_5f5f10b443a8f.pem 
      
      backend Jellyfin_ipvANY
      	mode			http
      	id			106
      	log			global
      	timeout connect		30000
      	timeout server		30000
      	retries			3
      	server			jellyfin 192.168.80.18:8920 id 107 ssl check-ssl check inter 1000  verify none crt /var/etc/haproxy/server_clientcert_5f8a7e8154947.pem 
      
      backend GM_ipvANY
      	mode			http
      	id			102
      	log			global
      	timeout connect		30000
      	timeout server		30000
      	retries			3
      	option			httpchk OPTIONS /si/home.do 
      	server			GM 192.168.80.38:9001 id 103 check inter 1000
      

      Vendor: HP
      Version: P01 Ver. 02.50
      Release Date: Wed Jul 17 2024
      Boot Method: UEFI
      24.11-RELEASE (amd64)
      FreeBSD 15.0-CURRENT
      CPU Type: Intel(R) Core(TM) i5-7500 CPU @ 3.40GHz
      Current: 3606 MHz, Max: 3400 MHz
      4 CPUs : 1 package(s) x 4 core(s)

      manjotscM 1 Reply Last reply Reply Quote 0
      • manjotscM
        manjotsc @manjotsc
        last edited by manjotsc

        Solved

        Guide : https://www.digitalocean.com/community/questions/get-client-public-ip-on-apache-server-used-behind-load-balancer

        First you need to set option forwardfor in each backend to have in HaProxy, under Advanced Settings then in Backend pass thru box put option forwardfor and hit save, Don't set forwardfor in frontend, then you need to enable "sudo a2enmod remoteip" in your apache2 server and edit config /etc/apache2/apache2.conf and add the following line at bottom of config, RemoteIPHeader X-Forwarded-For and in that same config locate the log file line : LogFormat "%h %l %u %t "%r" %>s %O "%{Referer}i" "%{User-Agent}i"" combined and change %h to %a to see the ip in logs, like this LogFormat "%a %l %u %t "%r" %>s %O "%{Referer}i" "%{User-Agent}i"" combined and restart your apache2 server and you are done and check your apache access logs and you should see public ip of clients.

        Screenshot 2021-01-02 114223.png

        [root@cPanel ~]# a2enmod remoteip
        Enabling module remoteip.
        To activate the new configuration, you need to run:
          systemctl restart apache2
        [root@cPanel ~]# systemctl restart apache2
        [root@cPanel ~]#
        [root@cPanel ~]# nano /etc/apache2/apache2.conf 
        
        LogFormat "%a %l %u %t \"%r\" %>s %O \"%{Referer}i\" \"%{User-Agent}i\"" combined
        
        # HaProxy Forward for Enable
        
        RemoteIPHeader X-Forwarded-For
        
        [root@cPanel ~]# systemctl restart apache2
        
        

        Screenshot 2021-01-02 114051.png

        ------------Before--------------- 
        
        192.168.80.1 - - [02/Jan/2021:11:10:00 -0500] "GET / HTTP/1.1" 200 623 "-" "Mozilla/5.0 (Linux; Android 10; SM-G975W) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.101 Mobile Safari/537.36"
        192.168.80.1 - - [02/Jan/2021:11:10:00 -0500] "GET /style.css HTTP/1.1" 200 277 "https://speed.manjot.net/" "Mozilla/5.0 (Linux; Android 10; SM-G975W) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.101 Mobile Safari/537.36"
        192.168.80.1 - - [02/Jan/2021:11:10:02 -0500] "GET /favicon.ico HTTP/1.1" 200 112448 "https://speed.manjot.net/" "Mozilla/5.0 (Linux; Android 10; SM-G975W) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.101 Mobile Safari/537.36"
        
        ---------- After --------------  
        
        204.48.93.246 - - [02/Jan/2021:11:11:05 -0500] "GET / HTTP/1.1" 200 623 "-" "Mozilla/5.0 (Linux; Android 10; SM-G975W) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.101 Mobile Safari/537.36"
        204.48.93.246 - - [02/Jan/2021:11:11:06 -0500] "GET /favicon.ico HTTP/1.1" 200 112448 "https://speed.manjot.net/" "Mozilla/5.0 (Linux; Android 10; SM-G975W) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.101 Mobile Safari/537.36"
        

        Vendor: HP
        Version: P01 Ver. 02.50
        Release Date: Wed Jul 17 2024
        Boot Method: UEFI
        24.11-RELEASE (amd64)
        FreeBSD 15.0-CURRENT
        CPU Type: Intel(R) Core(TM) i5-7500 CPU @ 3.40GHz
        Current: 3606 MHz, Max: 3400 MHz
        4 CPUs : 1 package(s) x 4 core(s)

        G 1 Reply Last reply Reply Quote 1
        • G
          GreenAnt @manjotsc
          last edited by

          @manjotsc

          some further notes on this:

          • the forwardfor option in the HAProxy front-end config works fine for forwarding the client address
          • you can use the backend option but I don't think it's necessary.

          For users using nginx. A similar logging can be achieved by:

          add this to /etc/nginx/nginx.conf in the http { block:

              ##
              # Logging Settings
              ##
          
              access_log /var/log/nginx/access.log;
              error_log /var/log/nginx/error.log;
          
              log_format combined_realip '$http_x_forwarded_for - $remote_user [$time_local] '
                              '"$request" $status $body_bytes_sent '
                              '"$http_referer" "$http_user_agent"';
          
          

          then in your virtualhost config:

          access_log      /var/log/nginx/access.log combined;
          access_log      /var/log/nginx/access_realip.log combined_realip;
          

          the access_realip.log will then include the client address. You can then use that to verify the HAProxy is passing through the variable properly.

          see this guide for reference: https://djangocas.dev/blog/nginx/nginx-access-log-with-real-x-forwarded-for-ip-instead-of-proxy-ip/

          G 1 Reply Last reply Reply Quote 0
          • G
            GreenAnt @GreenAnt
            last edited by

            @greenant

            if you have the http_realip_module available in nginx, then the config is even simpler, just add to nginx.conf:

            real_ip_header X-Forwarded-For;
            real_ip_recursive on;
            set_real_ip_from <your proxy/CDN IP CIDR here>;
            
            1 Reply Last reply Reply Quote 1
            • B
              braunerroei
              last edited by braunerroei

              Hello,

              Didn't work for me (both ways [frontend / backend]).

              I have Synology NAS behind my pfSense/HaProxy, and the Synology shows the pfSense IP and not the real client IP.

              -Roei

              B B 2 Replies Last reply Reply Quote 0
              • B
                braunerroei @braunerroei
                last edited by

                for synology:
                just configure "Trusted Proxies" - Control Panel> Security > Trusted Proxies

                1 Reply Last reply Reply Quote 0
                • B
                  bigtfromaz @braunerroei
                  last edited by

                  @braunerroei I curious, how did you get HAProxy in front of Synology to work? Is there an article somewhere that lays it out? Frontend-backend relation seems to be correct. It's working for our private Docker registries but when I set up https offload to http backend Synology, I just get 503s. "Use forwardfor" is checked and the front-end and backend are green in the stats.

                  B 1 Reply Last reply Reply Quote 0
                  • B
                    braunerroei @bigtfromaz
                    last edited by

                    @bigtfromaz
                    Hi

                    I am not pfSense / Haproxy expert, but it seems to be very easy.

                    Why https offload to http backend?

                    I configured backend with port 5001, and 2 frontends, one (shared) with 443 port and one with 5001 port, both use the same backend.

                    -Roei

                    B 1 Reply Last reply Reply Quote 0
                    • B
                      bigtfromaz @braunerroei
                      last edited by

                      @braunerroei

                      I assume you are storing your certificate on the Synology's box. If not, how are you securing your connection?

                      I have the HAProxy frontend handling SSL negotiation using a certificate stored on the pfSense router. The certificate is acquired using the ACME package, configured to use Let's Encrypt DNS-01 protocol. This way we don't have to expose any ports when renewing our certificate. In this scenario, SSL offload is not from pfSense to Synology. It's the other way around. Our HAProxy frontend is listening on a custom port, negotiating the encrypted connection and forward traffic to the Synology backend which listening, unsecured, on a 5xxx port.

                      I have the exact same configuration working (using different ports) for our Docker repository, without checking the forward for box. It works fine. In that case the frontend port is 443 and the backend port is in the 5xxx range.

                      It appears I need to open a ticket with Synology to see if they can help me find logs. I'm also having trouble getting detailed logs from HAProxy on pfSense.

                      B 1 Reply Last reply Reply Quote 0
                      • B
                        braunerroei @bigtfromaz
                        last edited by

                        @bigtfromaz

                        In the matter af fact, I use the certificate stored on the pfSense, that was created by ACME.

                        -Roei

                        B 1 Reply Last reply Reply Quote 0
                        • B
                          bigtfromaz @braunerroei
                          last edited by

                          @braunerroei Then your frontend config looks like this?

                          196c1e01-1e74-49f5-87c8-4d22eb7bf590-image.png

                          That's the SSL Offloading I was talking about. If you don't check that box, then pfSense won't negotiate SSL. I was worried that you might be processing unencrypted.

                          In any event, I resolved my 503 problems. I'm not using the default port 443 for this new connection. Therefore, the value of the "Host Matches" ACL entry needed to be my.host.com:6407. I had used my.host.com with no port.

                          I had assumed that HAProxy would tack the port number on to the value because the port number is specified in the external address table. I see now it can't do that. The External Address table may contain multiple entries. It follows that the ACL match routine has no way to know your intentions unless you specify the port number in the ACL.

                          Thanks for the help. Your answers got me questioning my own configuration which turned out to be in error.

                          1 Reply Last reply Reply Quote 0
                          • L louis2 referenced this topic on
                          • First post
                            Last post
                          Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.