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

    (SOLVED) Load balancing gateways on a single Interface (WAN)

    Routing and Multi WAN
    2
    5
    4.0k
    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.
    • M
      mv202
      last edited by

      Hi Everyone,  ;D

      I am having an issue with Load balancing on a single Interface.

      I found this message and followed the advice but I am still having problems.

      http://forum.pfsense.org/index.php/topic,9422.msg53290.html#msg53290

      I have edited my config.xml by going to Diagnostics / edit file.

      I edit the /cf/conf/config.xml file to add my two gateways and two
      monitor IP's. Here is a copy of my config.

      <load_balancer><lbpool><type>gateway</type>
                  <behaviour>balance</behaviour>
                  <monitorip>55.210.120.5</monitorip>
                  <name>WAN_Pool</name>
                  <desc>2 wan ip's used for load balancing</desc>

      <servers>192.168.1.254| 44.223.8.113</servers>
                  <servers>192.168.1.252| 55.210.120.5</servers></lbpool></load_balancer>

      Problem is the 192.168.1.254 is also set as the default gateway for
      the WAN interface. When I setup FW rules using the WAN_Pool traffic
      just uses the default gateway (1.254) and never uses the 2nd IP listed
      (1.252). In the load balance status 1.252 shows up as "offline". I can
      ping the 1.252 monitor IP from the FW and it works but it still shows up as offline.
      Is there something else I need to do in order to have load balancing on a single NIC (WAN).

      Thanks in advance for your help and suggestions.

      -mv202

      1 Reply Last reply Reply Quote 0
      • GruensFroeschliG
        GruensFroeschli
        last edited by

        I'm not sure if the automatic adding of a static route to the second gateway works if you add it this way.
        Can you verify that you can ping the monitor IP of the 1.252 gateway if you connect a client directly and have a single default gateway the 1.252 router?
        Could you also look at the routing-table and check if a static route is in place?
        If not you could try to add manually a static route for the monitor ip to the 1.252 router.

        btw: i cannot ping 44.223.8.113 nor 55.210.120.5 from here.
        Does your ISP allow pings only from their customers?
        You could also try to change the monitor IP's to 81.221.252.10 and 81.221.250.10 (only as a test, not as solution) since i know 100% that these IP's are pingable from everywhere.

        We do what we must, because we can.

        Asking questions the smart way: http://www.catb.org/esr/faqs/smart-questions.html

        1 Reply Last reply Reply Quote 0
        • M
          mv202
          last edited by

          Thanks for the response.

          I can ping the monitor IP from a client using the 1.252 router as its default gateway.  I did not supply the real IP's of my ISP for security reasons but the real ip's are pingable from the Internet.  I just tested pinging them from www.nwtools.com and it was successfull.

          I tried adding a route using 1.252 as the gateway for the monitor IP and the status still says "offline no changes found in log file."  Although the monitor IP's are pingable I did try the two IP's you suggested and even added a route to them via the 1.252 and it still says offline.

          I looked into the file  /var/etc/slbd.conf and found the following. Not sure if this helps.

          WAN_Pool|2 wan ip's used for load balancing:
          :poolname=WAN_Pool:
          :vip=127.0.0.1:
          :vip-port=666:
          :sitedown=127.0.0.1:
          :sitedown-port=666:
          :service-port=666:
          :method=round-robin:
          :services=1:
          :0=44.223.8.113:
          :ping:

          This is what I see in the load balancer log:

          Aug 15 09:53:12 kernel: arpresolve: can't allocate route for 192.168.1.252
          Aug 15 09:53:12 kernel: arplookup 192.168.1.252 failed: host is not on local network
          Aug 15 09:53:07 kernel: arpresolve: can't allocate route for 192.168.1.252
          Aug 15 09:53:07 kernel: arplookup 192.168.1.252 failed: host is not on local network
          Aug 15 09:53:07 kernel: arpresolve: can't allocate route for 192.168.1.252
          Aug 15 09:53:07 kernel: arplookup 192.168.1.252 failed: host is not on local network

          I am using traffic shaping and I turned it off to see if that had any adverse effects to multiwan but it did not help.  I am willing to try anything else anyone can suggest.

          1 Reply Last reply Reply Quote 0
          • M
            mv202
            last edited by

            The load balancer status now says Online! which is great.  I took the default gateway 1.254 out of the loadbalance pool and just left 1.252 to see if I can force traffic for certain FW rules to go out this gateway but I cannot get ANY traffic to go through the 1.252 gateway.

            I am still seeing these messages popup in the loadbalance log when I try to use that gateway via FW rules.

            Aug 15 09:53:12 kernel: arpresolve: can't allocate route for 192.168.1.252
            Aug 15 09:53:12 kernel: arplookup 192.168.1.252 failed: host is not on local network
            Aug 15 09:53:07 kernel: arpresolve: can't allocate route for 192.168.1.252
            Aug 15 09:53:07 kernel: arplookup 192.168.1.252 failed: host is not on local network
            Aug 15 09:53:07 kernel: arpresolve: can't allocate route for 192.168.1.252
            Aug 15 09:53:07 kernel: arplookup 192.168.1.252 failed: host is not on local network

            1 Reply Last reply Reply Quote 0
            • M
              mv202
              last edited by

              This problem is solved.  I rebooted the FW and now everything is working fine.  I never reboot and for once it seemed to help.

              1 Reply Last reply Reply Quote 0
              • First post
                Last post
              Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.