WiFi slows down after 2/3 weeks. Wired connections and access points fine

[#1]

I had the 5GHz WiFi issue and in Oct 2020, @bobpullen installed thee new firmware (4.7.5.1.83.8.273) on my Plusnet Hub One router. This fixed the 5GHz issue (the bandwidth being completely lost). 

My issue, which I've had for several months, is that every 2/3 weeks, the WiFi slows down from 70/18 to 21/16 (download/upload speeds in Mbps). When this happens, wired connections are running at full speed. A have a PC wired to the router; and there is a switch connected to the router, which has two access points wired to it. Wireless devices connection to either access point continue working at full speed when the Hub One has the WiFi issue. I'm close to the router, so the issue is probably the 5GHz network. The 5GHz network is not sync'd with the 2.5 one.

Is this a known issue and, if so, what is the fix? Any suggestion would be appreciated.

Thanks

Praful

[#2]

D/load a wifi analyser app to see which wifi channels are being used in your locale. Then manually
set in the router those not being used/ least used.

[#3]

Thanks @Gel

[#4]

Yes, it's a known issue.

I've pushed an update to your router that should fix it. You'll need to reboot to apply it.

[#5]

Thanks @bobpullen .

I've just seen this message. I thought I'd check the forum because it just happened again. As suggested, I'd set the channel to a channel that, when it happened, was not being used by any surrounding router. My two access points continued working at full speed.

I'll reboot and monitor.

Thanks again.

Praful

[#6]

@bobpullen I've had two instances with the latest firmware where the router has stopped accessing the internet on both wi-fi and wired connections. For the most recent instance this morning (about 13 May 9am), I could access the router via a wired connection and restart it.

Although the 5Ghz issue hasn't recurred, this firmware seems less stable, which is worse since it cuts off the internet completely. I've now wired up key machines so I would rather have the previous firmware and lose 5Ghz every now and then than lose the internet completely, which is essential when working from home and having meetings.

The boot log looks odd with two October entries appearing between reboots:

Time and date	Message
09:13:39, 13 May.	( 31.940000) System up, firmware version: 4.7.5.1.83.8.289
09:13:25, 13 May.	( 17.750000) System start from GUI
01:00:32, 29 Oct.	( 33.630000) System up, firmware version: 4.7.5.1.83.8.289
01:00:18, 29 Oct.	( 19.740000) System start
01:00:18, 29 Oct.	( 19.740000) Boot reason: watchdog reset (cause: 0x0)
04:51:24, 11 May.	( 32.630000) System up, firmware version: 4.7.5.1.83.8.289
04:51:11, 11 May.	( 19.570000) System start
04:51:11, 11 May.	( 19.570000) Boot reason: watchdog reset (cause: 0x0)
01:00:31, 29 Oct.	( 32.720000) System up, firmware version: 4.7.5.1.83.8.289
01:00:18, 29 Oct.	( 19.790000) System start
23:49:04, 05 May.	( 31.660000) System up, firmware version: 4.7.5.1.83.8.289
23:48:50, 05 May.	( 17.720000) System start from GUI
10:52:13, 05 May.	( 32.320000) System up, firmware version: 4.7.5.1.83.8.289

Please can you look at the logs to see if you can see anything that may be causing the issue.

As an aside, it would be handy to have a feature on the router than allows you to reboot it automatically at a given time eg reboot weekly on Sunday at 3am. The same feature exists on Android phones and it's a good way to clear any memory leak or other bug that manifests itself over a few days or weeks.

Thanks

Praful

[#7]

There's nothing I'm aware of that would cause loss of Internet connectivity, however those watchdog resets do not look healthy. There's something I'd like to try, however I have a few questions/suggestions...

1. Is your router physically rebooting itself without any interaction from you?

2. Can you confirm that you are not running multiple DHCP servers on the network e.g. you have DHCP disabled on your additional access points?

3. Do you have something running on your network that is exposing port 22 to the Internet? I'm struggling to access your device to get any worthwhile information from it and this could be because you're running a service on the port I'm trying to use.

4. If the issue occurs again, please can you try running some DNS lookups from an affected machine, e.g. from a command line run each of the following in turn and note the output:-

nslookup bbc.co.uk 192.168.1.254
nslookup bbc.co.uk 212.159.6.9
nslookup bbc.co.uk

[#8]

@bobpullen thanks for replying.

Time and date	Message
09:13:39, 13 May.	( 31.940000) System up, firmware version: 4.7.5.1.83.8.289
09:13:25, 13 May.	( 17.750000) System start from GUI
01:00:32, 29 Oct.	( 33.630000) System up, firmware version: 4.7.5.1.83.8.289
01:00:18, 29 Oct.	( 19.740000) System start
01:00:18, 29 Oct.	( 19.740000) Boot reason: watchdog reset (cause: 0x0)
04:51:24, 11 May.	( 32.630000) System up, firmware version: 4.7.5.1.83.8.289
04:51:11, 11 May.	( 19.570000) System start
04:51:11, 11 May.	( 19.570000) Boot reason: watchdog reset (cause: 0x0)
01:00:31, 29 Oct.	( 32.720000) System up, firmware version: 4.7.5.1.83.8.289
01:00:18, 29 Oct.	( 19.790000) System start
23:49:04, 05 May.	( 31.660000) System up, firmware version: 4.7.5.1.83.8.289
23:48:50, 05 May.	( 17.720000) System start from GUI
10:52:13, 05 May.	( 32.320000) System up, firmware version: 4.7.5.1.83.8.289
10:51:59, 05 May.	( 18.370000) System start
10:47:48, 05 May.	( 33.000000) System up, firmware version: 4.7.5.1.83.8.289
* 10:47:34, 05 May.	( 19.030000) System start
20:10:05, 04 May.	( 35.530000) System up, firmware version: 4.7.5.1.83.8.289
20:09:49, 04 May.	( 19.740000) System start

1. In the boot log above, my reboots are in blue. The first (4 May) was after the firmware update. I'm not sure if I rebooted on 5 May, 10:47 but have highlighted it. The most recent two reboots were because I had no internet. For the 5 May 23:48 reboot, I couldn't ping an IP address or a URL. For today's reboot, I had to get on a conference call, so didn't have time for diagnostics. So it would seem some reboots are happening without my involvement. Looking at the boot log, I see that the watchdog resets started from when the latest firmware was installed (4 May). They do not occur for the rest of the log, which goes back a year (to May 2020).

2. The access points have static IP addresses and no DHCP server. However, I just checked and DHCP was set to "auto" - which I assume would be a DHCP client if I had not set a static IP address. I have now disabled DHCP on both APs just in case. The auto setting is the default and was the case before the firmware update.

3. I'm not knowingly using port 22. Nothing has changed as far as I know since last week when you installed the update. So if you connected to the router then it should be the same. I have a number of devices on my network. An IP camera, smart plugs and Google Home speakers. I don't know if they're using port 22. But as I said, these were all there last week prior to the update. Is there something I can do to confirm which ports are being used on the router?

4. I'll run the nslookup commands next time.

Thanks

Praful

[#9]

Correction to my previous post:

2. I've been using the mobile app to manage the access points and decided to check out the web interface. It turns out there is a DHCP server that can be configured via the web interface but not on the mobile app. As mentioned, this was set to "auto" before the firmware upgrade. I've changed to "disabled" now.

[#10]

@praful thanks.

Can you reboot the hub when you next get the chance so I can see if that allows me to then access it?

[#11]

@bobpullen I've just rebooted

[#12]

@praful thanks.

I've pushed a dot version firmware to your device. The next time you reboot you should find it's running a version ending .1.1

This is actually the build we're preparing to deploy en masse.

One notable difference is the fact that it addresses a crash scenario where there are secondary access points on the network. The symptoms don't quite match what you've described but I'm hopeful it might help.

Once you've rebooted, keep an eye on things and let me know how you get on. If there's a genuine problem with this dot build (that did not exist in earlier firmwares), then we need to identify it pretty sharpish

[#13]

@bobpullen thanks again for the quick response.
I've rebooted (before the day begins in earnest) and will monitor.
Praful

[#14]

Hi @praful - am I right in assuming you've since swapped the hub out for a BT device? Did you see any crashes following the most recent Hub one firmware update?

[#15]

Hi @bobpullen - I've just seen this (are we supposed to get an email when someone posts?).

I had ordered a BT Business Smart Hub 6 as contingency after the second firmware update gave me problems. I've been trialling the BT router for the past week.

After the last (third) firmware update to the Hub One, there were no crashes for the three days I ran it. The first update sometimes ran for three weeks before I got the WiFi slowdown (for which I raised this issue). So it's not a thorough test but it's better than the second update.

Sorry for bailing on you but Hub One has been somewhat unreliable for over a year for me to now try alternatives. It could be I switched just when it became stable! If so others will benefit and I may revert back to it if the BT router isn't stable for me.

I appreciate your responsiveness and the help you have provided.

Praful