@ITV 

The orange is simply the existing background 'noise' on your selected channel.  It is possible your RF neighbourhood has changed, and you may need to rebalance or otherwise re-engineer to adjust.

I run a number of the 235-walls on 3.1.99 in an urban environment (noisy) and have not experienced the same behaviour you describe.

Here are a few of mine:

  @d0ugmac1 

Thank you for the quick response!

If it where just these orange area's then yes - I would agree.

But there is more - the most questionable one is that the EAP is not seen in the top interference list - as if it doesn't exist.

How does this look on your end?

Another part is the dropped rate - how does this look on your end?

On my end it runs above 30% - from time-to-time even above 50%

What controller version are you using?

I'm running 5.13.30 (in a docker container).

=====

Over night, I noticed that the EAP225-Outdoor is experiencing simular issues.

However, by far not as much as the EAP230-Wall.

The only solution seems to be installing an EAP245 or running the EAP230-Wall in standalone mode.

Just noticed the 3.2 beta - which seems to be compatible with controller version 5.9 and higher.

Will give that a try.

=====

Just changed the topic name as the questions are related to the EAP230-Wall-EU - not the EAP235-Wall-US.

Sorry for the "noise"/confusion.

=====

Hi@ITV 

For the top interference list, please try to upgrade the firmware to 3.2 and see whether it will show on the list.

And for the retired rate/dropped rate, may I confirm whether the performance of the EAP230-Wall will be affected? Have you encountered any unstable issue on this EAP?

  @Hank21 - I upgraded the firmware - no real improvements (see attached screenshots):

- still a significant # of retries (compared to the EAP450)

- still a lot of packet drops => a few 1000 every 5 minutes

- the EAP230 still doesn't show in the interference list

Any suggestions?

Based on the controller log it looks like connections are stable - no more WPA authentication failures.

What remains is high utilization @ the 2.4-Ghz band with 10-plus connected clients.

The EAP-225-outdoor and the EAP245 shows a utilization breakdown with rx-frames, tx-frames, interference and free.

While with the EAP230 it just shows utilization with busy and free.

It looks like the Omada monitoring stats for this EAP230 is a mess compared to the other two EAP models I tried.

Also because in standalone mode, the retries and drops are by far not as high as in controller mode.

While the CPU and memory utilization are very simular.

=====

Hi@ITV 

What about the actual performance of the EAP230-Wall? Will you encounter any unstable connection with it? We need to locate the issue that is related to the performance of EAP, or it is related to the compatible issue between the controller and the EAP. 

For the high retired rate/drooped rate, it might happen when most of the connected devices are using 2.4GHz WiFi. However, this situation may not necessarily affect the actual usage.

  @Hank21 

I just finished a few performance tests comparing the EAP230-Wall with the EAP245.

In terms of throughput: the EAP230-Wall runs in the range of 190-210 Mbps. While the EAP245 runs in the range of 200-240 Mbps.
What is your view here?

During these few tests, it looks as if the connection was stable - no session drops or serious delays.

In terms of dropped packet on the radio's:

The Omada controller shows about 80-90% packet drops on the 2.4-Ghz radio. And 20-50% on the 5-Ghz radio.

In the Omada dashboards there is no breakdown on RX or TX (like for the EAP245).
In addition the EAP230 doesn't show in the top interference list.

That is it for now (I guess).

Cheers - Will

ITV wrote

  @Hank21 

 

I just finished a few performance tests comparing the EAP230-Wall with the EAP245.

 

In terms of throughput: the EAP230-Wall runs in the range of 190-210 Mbps. While the EAP245 runs in the range of 200-240 Mbps.
What is your view here?

 

During these few tests, it looks as if the connection was stable - no session drops or serious delays.

 

In terms of dropped packet on the radio's:

The Omada controller shows about 80-90% packet drops on the 2.4-Ghz radio. And 20-50% on the 5-Ghz radio.

In the Omada dashboards there is no breakdown on RX or TX (like for the EAP245).
In addition the EAP230 doesn't show in the top interference list.

 

That is it for now (I guess).

 

Cheers - Will

 

 

 

 

Hi @ITV 

Thanks for update.

The performance on the EAP230-Wall looks fair as long as you haven't encountered the unstable issue with it.

We need to consult with the department and see whether there are any limitations with EAP-Wall series when we are using the widgets or checking the details.

You are highly welcome to share further about your finds with us.

  @Hank21 

Just finished a series of tests with the EAP650-Wall (replacing the EAP230-Wall).

What I noticed is that the WiFi dropped packets percentage of the EAP230-Wall is more or less equal to the WiFi retry percentage of the EAP650-Wall.

What I also noticed is that the 5-Ghz SNR value of the 2 Android devices in that room is 25-30 with the EAP230-Wall. While 60-70 with the EAP650-Wall.

Anyone an idea how to "read" this? Does this make any sense at all?

With warm regards - Will

ITV wrote

  @Hank21 

Just finished a series of tests with the EAP650-Wall (replacing the EAP230-Wall).

What I noticed is that the WiFi dropped packets percentage of the EAP230-Wall is more or less equal to the WiFi retry percentage of the EAP650-Wall.

What I also noticed is that the 5-Ghz SNR value of the 2 Android devices in that room is 25-30 with the EAP230-Wall. While 60-70 with the EAP650-Wall.

 

Anyone an idea how to "read" this? Does this make any sense at all?

 

 

With warm regards - Will

 

Hi @ITV 

Thanks for your updating.

The Statistics of EAP230-Wall and EAP650-Wall might be different in some way since they are not the same generation. And when we want to monitor the performance of them, we need to combine the actual performance like internet speed and the statistics info on controller.

For the SNR, signal-to-noise ratio is a measure used in science and engineering that compares the level of a desired Signal to the level of background Noise. It is defined as the ratio of signal power to the noise power. A ratio higher than 1:1 (or 0dB) indicates more signal than noise. A higher SNR value indicates a stronger and more reliable signal.