Every router’s DHCP service can only hand out addresses from a fixed range, called the DHCP pool. When the number of devices asking for an address gets close to or exceeds that range, the router can start failing to renew existing leases or refusing new requests, and a device that had a perfectly working connection can drop offline, not because anything is wrong with it, but because the pool ran out of room. The giveaway is timing: you add a new device, and a completely unrelated older one goes dark soon after. Checking your router’s configured pool size against your actual device count confirms whether this is what’s happening.
What a DHCP pool actually is
Your router’s DHCP server hands each device on your network an IP address from a defined range, not from the whole subnet automatically. A common home router might reserve a portion of its subnet, say the last two hundred or so addresses, for this pool, though the exact default range varies by manufacturer and model and is worth checking rather than assuming. Each address is handed out on a lease: a device holds it for a set period, then renews, usually getting the same one back if nothing has changed.
That pool has a ceiling. If you have, say, forty smart plugs, bulbs, sensors, a couple of laptops, three phones, a smart TV, and a few guest devices that never fully disconnect, you can be closer to that ceiling than you think. Smart home households cross this threshold faster than most, because a handful of switches, sensors, and plugs each count as a separate device the same as a laptop does.
Why adding one device can knock a different one offline
When the pool is genuinely full, two different things can happen depending on the router’s firmware, and neither is a dramatic failure message.
The router may simply be unable to grant an address to the new device until an old lease expires and frees one up, in which case the new device is the one that struggles to connect, not an existing one. But the more confusing case is when an existing device’s lease comes up for renewal at a moment the pool is tight: the renewal can fail or get delayed, and that device drops off even though nothing about it changed. It looks random because it is not tied to that device doing anything wrong. It is tied to the pool’s arithmetic at that moment.
This is why the symptom people report is so specific: “I added the 51st device, and three older ones fell off.” The new device isn’t causing a conflict with the old ones directly. It is just the addition that finally pushed total demand past what the pool can comfortably serve.
Confirming this is actually your problem
Before assuming this is the cause, check two things in your router’s settings, usually found under a section labeled something like “DHCP Server,” “LAN Setup,” or “Network Settings,” though the exact label and location depend on your router’s brand and firmware:
- The configured pool range (a start and end address) and how large it is.
- The current count of connected or leased devices, often shown in the same section or in a separate “Connected Devices” or “Client List” screen.
If your device count is at or near the pool size, or if the timing of a device dropping off lines up closely with when you added a new one, this is a reasonable diagnosis. If your device count is well under the pool size and nothing was recently added, look elsewhere before spending time here. The device, radio, router or cloud test is the right starting point if you are not sure which category this falls into at all.
What actually fixes it
Widen the pool range, if your router’s settings allow specifying a custom start and end address within your subnet. This is the most direct fix and doesn’t require touching individual devices.
Extend the lease time, if your router exposes that setting, so addresses churn less often and fewer renewals compete for room in the pool at once. This helps more with a pool that’s borderline than one that’s genuinely too small.
Reserve addresses for the devices you care most about staying reachable, so they are guaranteed a spot regardless of how crowded the general pool gets. That is a separate setting from widening the pool itself; DHCP reservations covers how that works and when it’s worth doing, and this article deliberately doesn’t repeat those steps here, since the two problems are related but not the same.
If your router’s firmware caps the pool size and you’re genuinely out of room, the longer-term options are a router with a larger configurable range or splitting devices across a second network, which is a bigger change worth its own research before committing to it.
What this isn’t
A full pool is a capacity problem: too many devices for the room available. It is not the same as a device that keeps losing its specific address after a reboot even though the network has plenty of spare capacity, which is what a DHCP reservation actually solves. It is also not the same as a device that’s technically getting an address fine but can’t be found by an app or another device on the same network, which points toward AP isolation or mDNS and multicast traffic being blocked instead. All three can look like “the device disappeared,” and telling them apart starts with checking whether the pool is actually the constraint.
FAQ
How do I know if my devices are hitting a DHCP limit and not something else?
Check whether the pool size in your router’s DHCP settings is close to your actual connected device count, and whether the timing of a device dropping off lines up with recently adding a new one. If both line up, this is a reasonable first diagnosis.
Can I just expand the DHCP pool range myself?
On most routers that expose a custom start and end address for the pool, yes, as long as the new range stays within your subnet and doesn’t overlap addresses you’ve already reserved for other devices.
Does every router show me the pool size?
No. Some consumer routers hide this entirely and only expose a device count or connected-clients list, without a visible pool range. If yours doesn’t show a range, a rough device count against common sense (dozens of IoT devices plus your regular devices) is the best you can do without more advanced tools.
Will restarting my router temporarily fix this?
It can, because a reboot can clear stale leases and momentarily free up room in the pool, but it does not fix the underlying capacity problem. Expect the same symptom to return once the device count climbs back to where it was.
Is this the same problem as my mesh system’s device limit?
No. Some mesh systems and access points advertise a maximum number of simultaneously associated wireless clients, which is a separate ceiling from the DHCP pool’s address range. Both can cause similar-looking symptoms, but they are configured, and fixed, in different places.
Sources referenced in this article
- LazyAdmin, What is a DHCP Server? Everything you need to know.
https://lazyadmin.nl/network/what-is-a-dhcp-server/(accessed 2026-07-19). Source of the lease/renewal mechanism and general pool-exhaustion behavior described above. - Cisco, IP Addressing: DHCP Configuration Guide.
https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/ipaddr_dhcp/configuration/xe-16/dhcp-xe-16-book/config-dhcp-addr-pm-xe.html(accessed 2026-07-19). Source for how a DHCP address pool is defined as a bounded range rather than the full subnet.
Review interval: 12 months (router firmware and default pool behavior vary by vendor and can change).