NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
Detached satellite
After noticing a low speed at my wife's Mac at the far end of the house, I checked orbilogin and found the satellite was not listed as connected. What happened? I'm not sure how to make them reconne...
ArtLee wrote:Also note the following--Mac OSX devices will NOT roam from AP to AP unless the signal drops way, way, down. It has to go down below -75 dBm.
Did not know that. Thank you...
And, from my testing this morning, I see the sat's message "No Connection" means no device has connected to it. But, it is connected to the router.
I'm testing a new location to see it the flakiness between router and sat gets fixed.
It may be confusing but the Orbi system considers the connection of client "devices" different from the router's connection to the satellites. That's why the satelite is shown in a different table on the router. The satellite is not just anohter cleint "device," it's a special unit connected on the separate 5 GHz dedicated backhaul channel.
If you look closely at the diagram on the satellite status page there are three connections shown there:
1) Router to satellite dedicated backhaul. This is shown on the left of the digram, between the router and sat.
2) 2.4 GHz client connections. This is on the right top of the diagram.
3) 5 GHz client connections. This is on the right bottom of the diagram.
The router-satellite connection is separate from the two client connections and the messages you noted.
Also, unless you have some sort of hardware problem, the flakiness is due to poor signal between the router and the satellite. It's up to you to find a geomtery that provides an adequate signal. If you have plaster, cinderblock, or stone walls, the signal will not travel far. Unfortunately, Orbi provides almost zero information to assist you, other than the blue light.
I'd suggest starting with the router and satellite closer together than you'd perfer, to demonstrate they can maintain a stable connection. Then you can move the satellite farther away in small steps.
Thanks. I think you've hit on it. Unfortunately, there is no way to measure signal strength for the 5GHz backhaul. Unless you've got a secret. Thanks for your help.
ArtLee wrote:
Thanks. I think you've hit on it. Unfortunately, there is no way to measure signal strength for the 5GHz backhaul. Unless you've got a secret. Thanks for your help.
the way i test the backhaul connection is to bring the orbi router and sat claoser together say 15 feet apart so you know the backhaul will be connected at max sync , connect a comp via ethernet to the orbi router and say a laptop connected by ethernet to the sat , then do a file transfer from the comp to kappy and back again , record those speeds as they are your benchmark
now move the sat and lappy further away and do the same and record , and keep moving further away
for the most part i have found that about 10 to 15 meters or 30 to 50 feet is about the butter zone for distance between the router and sat and after that the backhaul wont sync as well and the recorded numbers start to drop
doing the above will give you a specific location where you achieve the best backhaul sync and thus the best throughput and connectivity to all the clients connecting through the sat
I can do that! Thanks. Good tip.
btw when checking whats actually connected to the sat access ip gui via its ip address and it will show you the clients actually connected to the sat
the orbi router now does not show whats actually connected to the sat just whats connected to the whole system
I've done the same sort of testing, but I'm guessing most people wouldn't want to do it.
If you are up for manually testing the backhaul throughput, I would suggest you use iperf3. You can download the progam for Mac and PC. https://iperf.fr/iperf-download.php
On a computer on one side of the backhaul run the command in server mode "iperf3 -s"
On the other side, run the command in client mode "iperf3 -c <IP OF SERVER> -t 30"
Repeat the client command after each change in the setup.
Oh. Now you're getting complicated. Maybe I'll pass on this...
ArtLee wrote:
Oh. Now you're getting complicated. Maybe I'll pass on this...
i would agree , a simple file transfer will give you the answers you need to get the placement right
iperf3 is good and lets you get down to nickles and dime type info but is a bit overkill for the one of testing you prob need to achieve
I find iperf3 far simpler and faster to use than setting up a file transfer. To each his own though.
Okay, I'll try it tonight.