NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Daisy-chained RBS750 always goes to "poor" backhaul status if "Connected Orbi" briefly loses power
I could not find my model in the dropdown selector. I have a RBR750 (let's call this R) with 4x satellites (RBS750) (let's call them A, B, C, D). Satellites A, B, and C connect directly to R. But D c...
Re: distances. I did some rough measuring.
A->R = 20ft (this is A's chosen connection path)
B->R = 30ft (this is B's chosen connection path)
C->R = 30ft (this is C's chosen connection path)
D->R = 30ft (D chooses against this path for some reason)
D->A = 40ft (this is D's chosen connection path)
Note that D chooses to daisy chain through A even though R is closer. This may because there is slightly more wall interference with D->R compared to D->A, I'm not sure. But note that D->R is completely adequate (and is actually the way the system used to work, since A was a late addition). Once I added A, D started choosing it. Which is completely fine with me: D will work great with either of D->R or D->A. The issue is that A briefly loses power sometimes, and D refuses to recover gracefully from the A power cycle unless I also manually power cycle D.
One workaround solution I'd be happy with is if I could force D to choose the D->R path. Then the A power loss/cycle would no longer be an issue, because the daisy chain would no longer exist. But... I don't think this is possible? The system chooses its own connection topology, I think?
The111 wrote:
Re: distances. I did some rough measuring.
A->R = 20ft (this is A's chosen connection path) <Distance is too close here. 30 feet or more maybe need here. Possible that the D RBS is confusing A with the RBRs signal when it should be connecting to next nearest RBS...C?
B->R = 30ft (this is B's chosen connection path)
C->R = 30ft (this is C's chosen connection path)
D->R = 30ft (D chooses against this path for some reason) <As it should and should connect to next nearest RBS. You may need to put more distance between the #D RBS and the RBR to help it ensure that it connects to the nearest RBS instead of the RBR or A RBS...Is that nearest RBS #A or #C
D->A = 40ft (this is D's chosen connection path)
Note that D chooses to daisy chain through A even though R is closer. This may because there is slightly more wall interference with D->R compared to D->A, I'm not sure. But note that D->R is completely adequate (and is actually the way the system used to work, since A was a late addition). Once I added A, D started choosing it. Which is completely fine with me: D will work great with either of D->R or D->A. The issue is that A briefly loses power sometimes, and D refuses to recover gracefully from the A power cycle unless I also manually power cycle D.
One workaround solution I'd be happy with is if I could force D to choose the D->R path. Then the A power loss/cycle would no longer be an issue, because the daisy chain would no longer exist. But... I don't think this is possible? The system chooses its own connection topology, I think? Depends on placement and how the RBS are turned on. If placement is out in a liner line from the RBR, then all thats needed is to turn ON the first nearest RBS to the RBR. Once it's ready, then turn ON the next RBS out from the 1st RBS and wait for ready, and so on and so on for the other RBS. This sets the system for Daisy chain.
I think distances and placement are a factor here for all RBS. I would play around with the distance of A RBS and get more distance between it and the RBR. Also you might test with Transmit power on the RBR and see if 50% changes the behavior your seeing.
What is the SQ Feet of your home?
Possible that you may have too much wifi signal around as well.
Thanks for the suggestion. I may be able to move A a bit further away from R, which would presumably force D to connect directly to R instead, and thus solve my issue in a roundabout way. Out of curiosity, where do you see the 30ft or more rule? I don't see it in the page you linked or in any of the Netgear documentation.
And FWIW, I consider this ultimately an issue that could be resolved in software/firmware, since again the system works fine until a power cycle of a single unit messes up the topology, after which the system knows there is an issue but refuses to try to fix it. High level solutions to this problem would be (1) allow hardcoding the topology as a configuration option (2) design the system to attempt to fix poor backhaul status through at least a single reboot. Seems worth reporting to Netgear if there is a path here to do that.
The house is 3200 sqft living space, 2 stories. But one satellite (B) is in the attic (i.e. "3rd story") immediately next to a smart hub for our central air, since that hub has horrible reception and will not maintain connected status any other way. And another satellite (A) is in the garage. Which even though it is relatively close to R, there are enough walls and floors being passed through that the garage will not have a reliable signal without A being there, currently at the 20ft distance you say is too short. I will experiment with other locations in the garage if possible. I don't think I'm overprovisioned, because the signal is not full strength everywhere in the house. But it is usable everywhere, with this configuration, and full strength in most places.
The111 wrote:
Thanks for the suggestion. I may be able to move A a bit further away from R, which would presumably force D to connect directly to R instead, and thus solve my issue in a roundabout way. Out of curiosity, where do you see the 30ft or more rule? I don't see it in the page you linked or in any of the Netgear documentation. Ya it used to be in a KB article but that got removed for unknown reasons. Anyways, the Orbi systems wifi signal is pretty strong and can go a pretty good distance before it starts to degrade the signal strength. Been a long time user experiences seen over the years that 30 feet has been a good starting point for RBS from the RBR. Some users have reported seeing good operaton with RBS out at 50-60 feet. However again, this depends on building materials. Standard wood and sheet rock homes tend to not impact wifi signals as much as concrete and steel. Basements are not recommended places for the RBR. RBS yes. RBS being too close is just something like two people yelling at each other when talking about something instead of a normal tone if that help make sense. Also having too many people all talking at the same time is like having too much wifi signals in the home. Gets noisy and causes problems for devices trying to pick and chose which person is best to listen too and talk too.
And FWIW, I consider this ultimately an issue that could be resolved in software/firmware, since again the system works fine until a power cycle of a single unit messes up the topology, after which the system knows there is an issue but refuses to try to fix it. High level solutions to this problem would be (1) allow hardcoding the topology as a configuration option (2) design the system to attempt to fix poor backhaul status through at least a single reboot. Seems worth reporting to Netgear if there is a path here to do that. You might start with NG support first.
The house is 3200 sqft living space, 2 stories. But one satellite (B) is in the attic (i.e. "3rd story") immediately next to a smart hub for our central air, since that hub has horrible reception and will not maintain connected status any other way. And another satellite (A) is in the garage. Which even though it is relatively close to R, there are enough walls and floors being passed through that the garage will not have a reliable signal without A being there, currently at the 20ft distance you say is too short. I will experiment with other locations in the garage if possible. I don't think I'm overprovisioned, because the signal is not full strength everywhere in the house. But it is usable everywhere, with this configuration, and full strength in most places. Well for 3200 sq ft, and you having 4 RBS, this maybe over kill for this size of home. Like I mentioned before, having too much wifi signals maybe causing problems. I have a 5000k sq ft home 2 levels and a basement. I run two RBSs, one 1st level dining area that I can get far as I can from the RBR which is about 30 feet. 1 RBS down in the basement opposite end from the RBR thats on 2nd level. I have use just 1 RBS before and have had great coverage through out the home. Garage area included which is attached. Since I run two RBS with the RBR, I've turned down the power output of the RBR as one of my NEST controller sits 4 feet away from the RBR and have had problems with it and Orbi AX in the past. V4 seems to be working now however I don't want to burn out the poor controller being that close. Get great coverage and operation for this configuration and placement. Has worked well.
So for your size of home, you may need to go back to the drawing board and take a look at your situtuation, placement of RBS and how many is really needed. You may need only two RBS instead of four.
Is the drawing you made accurate to placement in the home or can you draw up a top down flat drawing and indicate where exactly the RBR and RBS are currently placed. Maybe we find a better placement for your home and see if we can help you get better operation and wifi coverage for what you need.
R is near the center of the 2nd floor. It needs to be here (only coax drop) and it also seems like a very reasonable location for it.
C and D are on opposing corners of the 1st floor. These were the only locations that gives the whole house good coverage.
A is in the garage. Before adding it, the garage got very inconsistent coverage.
B is in the attic. Before adding it, the furnace smart hub got very inconsistent coverage.
Again, the configuration works completely fine. The distances are clearly ok for wireless signal transmission. The actual issue is with topology resolution after a single unit loses power. There is no other way to arrange this physical topology to meet all my constraints, and even if there was, achieving that would probably be a bigger PITA than my current workaround which is just to manually cycle the power on D when the hiccup happens. It would just be really nice if the system could do it for me. I'm skeptical about NG support, but I'll check if there is a path there to report the issue, thanks.