I appreciate the willingness!
Our ISP has recently been changed to Starlink. We’re located in the middle of nowhere with only one other ISP, who stinks, so Mr. Musk was very welcome.
With the faster speeds, however, comes a price: we’ve discovered something is seriously wrong in our network. Before we simply chalked it up to bad ISP (they only guaranteed 25 down!)
Anyway, we have only Linksys routers (of differing models) with the Starlink router in bypass mode. We have 1 main router which is dishing out iPs to connected devices. All other routers (5) are in ‘bridge mode.’
(As far as I understand it, when I connect to a network which has a router in bridge mode, all the router is doing is sending me back to the main router to connect to the internet. That may be wrong, I’m not an expert.)
Here’s the problem:
When connected via Ethernet at any of the routers (main, or any router in bridge mode) speed is fast and consistent (around 80 down. Not as good as direct from Starlink, but ours is not the best router). When connected to Wi-Fi at any network (including the main) there are not only slower speeds (which I somewhat expect) but also inconsistent signal. Goes anywhere from 60 down to less than 1 in less than 10 seconds. Devices stay connected to the network but there is hardly any speed.
I don’t understand why Wi-Fi is so inconsistent at each router, yet very consistent Ethernet at all routers.
If you need me to to draw a diagram, I can.
I think I understand the problem space, but feel free to diagram. It is not at all unexpected for you to see significantly higher speeds when connected via Ethernet. Ethernet speed tests are your baseline for what should more or less be coming out of the actual router issuing IPs. If when performing the first test in the list of wired considerations below you find that you're getting less than 80% of what you're paying for on your bill, call Starlink and know that they're obligated to get it straightened out.
As for the trouble you're having, you're going to want to try to isolate the problem router(s), if possible.
Here are a few things you're going to want to consider:
Wired Considerations:
1. Run a speed test via Ethernet while connected to the primary router (issuing IPs via DHCP) and compare that to a speed test a bit further out (say, through a wall jack, or hardwired to a router running in bridge mode). This stuff can get pretty confusing even if you know what you're doing, so maybe take notes as you test different equipment and different scenarios. If nothing else it'll help me help you.
2. Are the routers running in bridge mode hardwired to a switch which is connected to the primary router, or are all of these routers connected wirelessly to each other? If they're connected wirelessly, connecting them to the primary router via Ethernet will help considerably.
Wireless Considerations:
1. A Wi-Fi Analyzer is going to be a must. I don't have any particular recommendations as far as apps go, however. You want something that'll show you a signal-to-noise ratio (SNR) and ideally you want it to be in the 24-40 dB range. 40 is better than 24.
2. Another thing that will help troubleshoot and isolate the issues will be if you can give the routers in bypass mode unique SSIDs. I'm not sure if that's possible or not with your deployment. If it's not possible but if you can deal with a bit of an outage in order to troubleshoot, then you can shut down the other routers that are in bridge mode. This will enable you to look at RFI using the app with respect to a particular router.
3. Where the endpoints are located matters. When I say endpoints, I'm talking about laptops, cell phones, etc. Anything you're connecting to the network with. If you're very close to a router in bridge mode but on the other side of drywall with electrical and metal panels in the wall, you could be dealing with isolated radio frequency interference resulting in signal loss.
4. I'm not familiar with the specific routers you're using in bridge mode, but I would be interested to know if any of the endpoints are mobile. That is to say, will a person using a laptop start connected to one router in bridge mode and then move some distance away (say, for a meeting) with their laptop. This could cause the signal to degrade. Depending on the model and the firmware, the routers in bridge mode may preferentially transfer a mobile endpoint to a router that is closer in proximity, whereas others may hold on to that connection.
5. You'll also want to look at SSID congestion. This sounds like it shouldn't be an issue if you're in the middle of no where. Depending on the range of the endpoints that are connecting to WiFi, you might also consider using 5 GHz instead of 2.4 GHz. It doesn't go as far but will offer greater bandwidth to the endpoints.
Last but not least, if you're running older hardware with older network interface cards (NICs), bear in mind that this can also be a bottleneck. Truthfully, there's bottlenecks all the way through from the primary router to the endpoint at the furthest bridged router. Depending on the specific manner in which the routers are bridged, you may be multiplying the number of packets needed to get to the primary router from an endpoint also, which even at 2x will significantly degrade your bandwidth. Also note that if the ISP isn't offering you some sort of synchronous connection where your upload and download speeds are the same, depending on what type of traffic your endpoints are sending across the wire, this can lead to issues as well. Many routers have a QoS feature which will allow you to prioritize certain traffic or all traffic from certain endpoints.
As a general heuristic, you want to begin by eliminating as many of the variables as possible (e.g. hardware to include additional routers in bridge mode and assorted endpoints), start with what you know (Ethernet speedtests taken while connected to the primary router, which will be your baseline), and then begin to work your way out from there (testing wired patch panels and Ethernet jacks, comparing speedtests from them to your baseline). From there, add one bridged router and one wired endpoint, repeating your tests with one wireless endpoint. Turn off that bridged router and turn on another, repeating your tests with the same wireless endpoint. For any significant variations, use the WiFi Analyzer. Then slowly and methodically add bridged routers and endpoints back onto the network, repeating your tests from each and monitor closely for when the issue presents itself.
Hopefully this helps, or if nothing else gives you some things to think about. If you do feel inclined to produce a diagram, I may be able to offer some more specific advice. Pray in all things and holler if you need anything.
) We will also be upgrading g the main router soon to increase Ethernet speeds (current router has 100 mb ports only)
