Unreasonable(?) amounts of deauth events occurring on AX-86U

[Burrito]

So I have a few Echoes in my house. I didn't buy 'em. This is a sort of double help request here, because I'm foraying into territory I'm unfamiliar with.


Jan 14 12:30:20 wlceventd: wlceventd_proc_event(494): eth7: Deauth_ind 24:4C:E3:C4:70:B6, status: 0, reason: Unspecified reason (1), rssi:0

This is one of the Echoes in the house. Between this and another, there are dozens of deauth/auth events happening over a five-minute period of two MAC addresses. They all seem to be Echo devices, and if I unplug problem devices then others start popping up in the log. I'm starting to wonder if this is affecting my wireless network because my phone is occasionally telling me it's connected without internet. I realize that by default on stock firmware I'm not supposed to be seeing these events, but I'm still seeing a shirt-ton of them and I can't help but think there's something wrong here. Are Echo devices awful on WiFi? I started noticing I had a problem when one of my full-size Echoes started telling me it couldn't connect to the internet. Restarting it would fix it for a few hours, then it would come right back. Sometimes the Echo would say the network ID changed, even though it didn't. I noticed later on as well that if I tried to use it and it gave me the troubled connection message, my phone would lose its WAN connection.

This isn't indicative of a failing router, is it? Damn thing's only six months old.

 

[bbunge]

So I have a few Echoes in my house. I didn't buy 'em. This is a sort of double help request here, because I'm foraying into territory I'm unfamiliar with.


Jan 14 12:30:20 wlceventd: wlceventd_proc_event(494): eth7: Deauth_ind 24:4C:E3:C4:70:B6, status: 0, reason: Unspecified reason (1), rssi:0

This is one of the Echoes in the house. Between this and another, there are dozens of deauth/auth events happening over a five-minute period of two MAC addresses. They all seem to be Echo devices, and if I unplug problem devices then others start popping up in the log. I'm starting to wonder if this is affecting my wireless network because my phone is occasionally telling me it's connected without internet. I realize that by default on stock firmware I'm not supposed to be seeing these events, but I'm still seeing a shirt-ton of them and I can't help but think there's something wrong here. Are Echo devices awful on WiFi? I started noticing I had a problem when one of my full-size Echoes started telling me it couldn't connect to the internet. Restarting it would fix it for a few hours, then it would come right back. Sometimes the Echo would say the network ID changed, even though it didn't. I noticed later on as well that if I tried to use it and it gave me the troubled connection message, my phone would lose its WAN connection.

This isn't indicative of a failing router, is it? Damn thing's only six months old.

No, nothing failing. Asus has the log level turned up so you will see a lot of almost worthless entries.
You have two choices:
1. ignore the log
2. turn the log level down. https://www.snbforums.com/threads/ax86u-asuswrt-fw-log-level-settings-via-ssh.74207/

 

[Burrito]

No, nothing failing. Asus has the log level turned up so you will see a lot of almost worthless entries.
You have two choices:
1. ignore the log
2. turn the log level down. https://www.snbforums.com/threads/ax86u-asuswrt-fw-log-level-settings-via-ssh.74207/

That's fair, I guess. But if deauths are supposed to indicate disconnections and there's hundreds, if not a thousand of them a day, how are they worthless?

I still have the problem of at least one Echo device flat out refusing to connect after being powered on for several hours and possibly punting my phone off the WAN when it's bugging out on use.

 

[tgl]

That's fair, I guess. But if deauths are supposed to indicate disconnections and there's hundreds, if not a thousand of them a day, how are they worthless?

FWIW, I agree that this log data suggests a problem. Deauth events might indicate an intentional disconnection by the device, for instance if it goes to sleep or roams to another AP. But you've only got the one AP (right?) and even if you have more, non-mobile devices like Echoes shouldn't be roaming too much.

Maybe these Echoes are on the ragged edge of being too far from the AP? (Try to compare the log data for signal strength between the devices having problems and those not.) Maybe you are having interference problems? (If these are on the 2.4GHz band, see if the disconnects correlate with use of a nearby microwave oven...)

 

[AlbumMan]

That's fair, I guess. But if deauths are supposed to indicate disconnections and there's hundreds, if not a thousand of them a day, how are they worthless?

I still have the problem of at least one Echo device flat out refusing to connect after being powered on for several hours and possibly punting my phone off the WAN when it's bugging out on use.

I had a similar problem with my 2.4G Ambient Weather Station while the Honeywell thermostat was rock solid. Turns out after doing a factory reset, the problem went away and has not dropped since.

 

[Burrito]

So as it turns out the "repeatedly disconnecting" issue was solved by widening my channel bandwidth. It was set to an automatic 20/40/80 instead of a straight 80MHz, since I use a Quest 2 for wireless PCVR, but it's weird because I know I manually set it before. That issue's fixed.

Though I do still have issues with Amazon Echoes and unholy amounts of deauths. There's several hundred events' worth of deauths and reconnects from a single Echo right now and it's the one closest to the router. I wasn't having an issue with this singular Echo like this yesterday.

Can a ridiculous number of drops and reconnects cause the router to lose connection to the WAN at all?

 

[OzarkEdge]

So as it turns out the "repeatedly disconnecting" issue was solved by widening my channel bandwidth. It was set to an automatic 20/40/80 instead of a straight 80MHz, since I use a Quest 2 for wireless PCVR, but it's weird because I know I manually set it before. That issue's fixed.

I suspect not. Clients will connect at their max bandwidth permitted by conditions, including the router max bandwidth setting... both 80MHz and 20/40/80MHz max bandwidth setting will permit clients to connect at 20, 40, or 80MHz bandwidth. You can verify this by inspecting client connection details in the Wireless Log.

OE

 

[Burrito]

Then I really don't know what to do. I don't know if this is a question that Support can help me with and Asus' NA support is apparently designed to be as inconvenient as possible. Even the sketchy translated site for the rest of the world would let me do this via email.

The issues summarized: I have not had any issues with the router until around December, when I noticed one of my Amazon Echoes randomly being unable to connect to the internet until I reset it, and this was the only one that was having problems. Now when I see a big batch of deaths it's multiple devices and they're almost all Echoes, but not on the same scale after permanently disconnecting the problem Echo, which was on the 2.4GHz band.

Then I'd notice my Quest 2 getting massive latency spikes and disconnecting from my PC. I check the logs and there are dozens of deaths all in the same second in multiple batches. The Quest 2 is on the 5GHz band.

My phone will randomly get a drop in WAN connection for about ten to fifteen seconds, and during this time the web interface for the router will be unresponsive. Also on the 5GHz band.

 

[Burrito]

Did a factory settings reset. Now another echo device is constantly deauthing and authing like it do. In the span of two minutes there are 1116 total events regarding this specific Echo. Over a thousand.

What the absolute frack is happening right now?

 

[degrub]

did this occur after a firmware update in the echo ?

What signal level are your devices seeing ?

turn off all of your wireless users. Only the wireless router or AP on.
you may need a wireless survey app. if an IOS device, download Airport utility and run a WiFi scan ( should be in the upper right corner of the app page. If not, you have to enable in settings for the app. see apple support if no obvious). 15-20 seconds should be enough. On 2.4 GHz, an RSSI of -65dB or greater should be ok, but depends on interference.


Go around to each location where the device is located. What is the measured RSSI on your SSID ?
Realize that the radios you have in the IOS device are not the same as those in the Echo device and may be measuring different signal level. This is just an approximation to look for issues.
Are there any equal or stronger radios, perhaps from neighbors on close channels ?

assuming no wireless issues obvious,
turn off all of your wireless users.
try a hard factory reset of the router. manually configure the minimum settings. do not use a saved config file. This may also be the right time to update firmware in the router before manual configuration to one that is known least problems. power up each of the wireless devices one by one looking for issues.

If nothing is obvious, it could be a power supply issue to/on the wireless router . replacing the wall wart power supply sometimes solves issues.

 

[bbunge]

Try the Asus test firmware for the AX86U.

 

[camaro322hp]

@Burrito I have noticed quite a few Deauth events in my AX86U log as well, although not thousands as you are seeing. I'm running stock ASUS 22068. I checked a few of the MAC addresses and most seem to be Google Homes or Chromecasts. I haven't noticed any adverse effects from these events.

 

[Burrito]

Actually recently unplugged one of the Echoes. It was the one downstairs in the basement and I've stopped getting the deauth spam. Might be a faulty device, so that rules out the spam as the cause for why the router locks up on the Wi-Fi the way it does.