Skip to content

14. MultiWAN & Load Balancing

MultiWAN lets NTPSense Pro use more than one internet connection at once — for automatic failover (switch to a backup ISP if the primary goes down), load balancing (split traffic across multiple links), or both. This is a Pro-exclusive feature.

14.1 Prerequisites

  • At least one additional internet connection beyond your primary WAN1 (a second ISP, a 4G/5G modem, etc.)
  • A spare interface for it — typically one of your OPT ports, physically connected to that second internet connection
  • The gateway/router IP that connection's modem or ISP hands out

14.2 Adding a Gateway

Open Network > MultiWAN > Gateways and click Add Gateway.

  1. Name — a label for this link (e.g. ISP1, ISP2, Backup-4G).
  2. Interface — the physical port this connection uses.
  3. Gateway IP — the upstream router/modem's IP address for this connection.
  4. Monitor IP — a reliable, always-on internet address used to test this link's health (e.g. a public DNS server like 8.8.8.8 or 1.1.1.1). This should be different from your other gateways' monitor IPs, so a health-check result for one link is never confused with another.
  5. Link type — leave as Dedicated unless you're specifically using this link for Site Mesh VPN traffic preference (Shared).
  6. Click Save.

Note

You don't need to configure any firewall rules yourself for the health check to work — NTPSense automatically manages the minimal firewall access this specific link needs to reach its Monitor IP, the moment you save the gateway.

14.3 Health Check Settings

Open Network > MultiWAN > Settings to adjust how aggressively links are monitored:

Setting Meaning
Check interval How often (in seconds) each gateway is tested. Default 5 seconds.
Fail threshold How many consecutive failed checks before a link is marked Down. Default 3 (i.e. ~15 seconds of failure at default interval before failover triggers).
Recover threshold How many consecutive successful checks before a recovered link is trusted again. Default 3.

Lower thresholds react faster but are more sensitive to brief, harmless blips; higher thresholds are steadier but slower to fail over.

14.4 Checking Gateway Status

Network > MultiWAN > Gateways shows live status for every configured link:

Column Meaning
Status Up / Down, based on the health check.
Latency Round-trip time to the Monitor IP.
Jitter Variation in latency between checks — high jitter can affect real-time traffic like VoIP even when the link is technically "Up".
Loss Percentage of recent health-check probes that failed.

If a link shows Down with 100% loss but you believe the connection is genuinely working, see the troubleshooting note below before assuming there's a real outage.

14.5 Groups — Failover and Load Balancing

A Group decides how traffic is distributed across your gateways. Open Network > MultiWAN > Groups and click Add Group.

  • Failover — traffic always uses the highest-priority Up gateway; others are only used if it goes down.
  • Load Balance — traffic is split across all Up gateways in the group, according to each one's assigned weight.

Add each gateway you want in the group, set its priority (for Failover) or weight (for Load Balance), and save.

14.6 Setting the System Default

One group must be marked as the System Default — this is the group used for general outbound traffic that isn't otherwise matched by a specific routing rule. Set this from Network > MultiWAN > Groups, using Set as Default next to the group you want.

14.7 Troubleshooting

A gateway shows Down / 100% loss, but the connection seems fine when tested manually:

  1. Confirm the Monitor IP is genuinely reachable — try pinging it manually from the gateway's console using that specific link's route.
  2. Check Network > MultiWAN > Gateways to confirm the Monitor IP is entered correctly (a typo here is the most common cause).
  3. If you're using an OPT-type interface for this link (rather than a dedicated WAN2 port), confirm the interface itself has a valid IP and link status under Dashboard > Interfaces first — MultiWAN health-checking depends on the interface being properly connected regardless of its role.

Failover doesn't seem to happen even though a gateway shows Down:

  • Confirm the group containing that gateway is genuinely set as your System Default, or that specific routing rules are pointing traffic at that group.
  • Check that Fail threshold hasn't been set unreasonably high for your needs.