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

    WAN Gateway Group balancing more than expected

    Routing and Multi WAN
    2
    2
    244
    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.
    • T
      talaverde
      last edited by

      Multi WAN Gateway Group, Gigabit Fiber WAN1 is Tier1, 200 Mbit Cable is Tier 2. Trigger Level is 'Member Down'.

      When I first set this up a while back, all traffic would go through WAN1, unless is was truely DOWN, then it would flip over to WAN2. It seemed more like one or the other (which is what I thought 'member down' was supposed to do). Exactly as expected.

      Recently (with same configuration), it seems more like gateway group is balancing the two WAN connections. WAN1 is processing twice as much data, for sure, but it's also 5X+ faster. WAN2 shouldn't be doing anything at all, unless WAN1 goes down, completely.

      IMO, it shouldn't be this way. If I wanted it more balanced, I'd have the group based on packet loss / latency. Are there any thoughts why it's balancing more than it should be? I don't want WAN2 to work at all, if WAN1 is available.

      Thanks

      1 Reply Last reply Reply Quote 0
      • J
        jkamal
        last edited by

        Hi talaverde,

        I am new at this, so please take whatever I say wit ha grain of salt or two, but you may want to examine the settings used to determine "Member down".

        The testing is usually based on a monitored IP. It is quite possible that transient instability is causing the rule to trigger your tier2 connection. Below are my settings from System -> Routing -> Gateways -> Edit.
        8a256521-9131-4277-9323-b3194b2a5788-image.png

        If you are seeing more hits on your tier2 connection, you may need to adjust these values to account for any transient events that are triggering member-down.

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