What's new

[ZenWifi XT9] Random issues with latency and dropped packets

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!


New Around Here

I recently purchased 2x ZenWifi XT9 units and have these configured as follows with FTTP:

FTTP ONT -> 1st XT9 (Main router) -> 2nd XT9 (Upstairs)

ONT is connected to the main XT9 via CAT 7 Ethernet cable. 2nd XT9 is connected via a wireless backhaul on 5 GHz-2 (160 MHz, fixed channel 100).

I may be being pedantic, but after checking network stability via ping tests, I'm seeing some odd results.
Randomly, ping will spike to between 180-220ms for 1 ICMP packet. The next response is within normal range (8-12ms). This happens with multiple remote hosts and even when pinging the main router from the 2nd XT9.

I have an Ethernet cable from the 2nd XT9 plugged into my machine, and I've also had the same result with WiFi connection to the 2nd XT9.

Ping plotter (0.5s) - top is main XT9, 2nd is, 3rd is, 4th is google.com:


As well as random ping spikes, I'm also experiencing random packet loss. It is really only the odd packet and so likely to not have any real effect on the network, but it still isn't as stable as I'd like it to be for the price and it's bugging me! The ping spikes are much more annoying then the odd lost packet. I have also lost packets pinging the main XT9 from my computer (ethernet connected to 2nd XT9).

Wireless backhaul (5 GHz-2) info:
noise: -89 dBm Channel: 100/160
RSSI: Fluctuates between -65 and -68
TX rate: 1297M
RX rate: 1945.6M
Chanspec: 5GHz channel 114 160MHz (0xe872)
Primary channel: 100 (Fixed)
Disabled: Enable IGMP Snooping, Enable TX Bursting, Enable WMM No-Acknowledgement, Enable WMM APSD, Optimize AMPDU aggregation, Airtime Fairness, Universal Beamforming

Wireless fronthaul (5 GHz-1) info:
noise: -88 dBm Channel: 60/80
Chanspec: 5GHz channel 58 80MHz (0xe23a) (Disabled 160 MHz for testing on the fronthaul but this has made no difference)
Primary channel: 60 (Fixed)

Smart connect and Roaming assistant are both enabled as I'd like the mesh functionality to work correctly - not possible to split the bands out
Pinging direct from the router sees a very occasional ping spike up to 80ms or so but seems way less frequent than what I see on a device using the wireless backhaul.
QoS enabled and AIProtection also enabled (not sure if these could be impacting?)

I know ICMP is not a prioritised packet but I didn't expect to see the results I've seen. Is this sort of result expected when using a wireless backhaul? I thought DFS channel checks may be the culprit but they happen too frequently, and ping spikes don't always line up with eachother when running multiple at the same time.

Would it be worth trying 5 GHz-2 on 80MHz instead of 160 to see if that appears more stable? There is probably 7-8m distance between the two XT9s and a few walls (though plaster, not brick walls)

I've tried disabling beamforming etc for the 5 GHz-2 band but still this is no good. I work from home and game a lot all from upstairs so I'm hoping there is a way I can get this more stable. Annoyingly, even though UK permits so, 5.9GHz channels do not appear on my XT9s and so I can't try a non-DFS channel for 160mhz.

Some CMD ping results:
Ping statistics for
Packets: Sent = 1773, Received = 1772, Lost = 1 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 5ms, Maximum = 193ms, Average = 9ms

Ping statistics for
Packets: Sent = 1822, Received = 1822, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 7ms, Maximum = 195ms, Average = 11ms

Ping statistics for (XT9 main router):
Packets: Sent = 1823, Received = 1822, Lost = 1 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 3ms, Maximum = 222ms, Average = 5ms

Appreciate your time and sorry for the bombardment of info - I just wanted to share everything I thought could be relevant. Thanks so much for reading!
Welcome to the forums @Graxly.

I would start by testing different Control Channels, systematically, and one by one (do not test by changing both backhaul and fronthaul at the same time). I would extend this testing to the 2.4GHz band too.

Do not use a WiFi analyzer (they're useless as they can't see/show non-WiFi interference). Particularly, do not have a WiFi analyzer app running while doing your tests. Just having it running will affect your observations.

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!