Intermittent connection issues loading/changing site- secure connection failed/site can't be reached

[EDIflyer]

Is anyone else occasionally getting a 'connection down' error from Chrome (or 'Secure Connection Failed' from Firefox) when changing page - if I hit refresh it works fine. At first I wondered if it was the site I was using but have noticed it on multiple different major websites over the past couple of months and not noticed it from other locations where I don't use Three Broadband so does seem to be related to the router/connection (this is via wired Ethernet too, so not a WiFi problem).  I've tried changing DNS server in case that helps but no difference.

It mainly seems to happen when trying to first load a site or (annoyingly) at checkout when a different site is being loaded as part of the checkout process.  It certainly doesn't happen everytime but does happen with reasonable frequency.  I've also noticed I often get it when trying to pull/push from/to Github too and have to do so a number of times for it to work (browsing the Github website works fine).

I'm using a NR5103E with Firmware Version V1.00(ACBJ.0)b14 - I tried rebooting it but to no effect.

[applepieman]

That isn't an answer, an answer would be why the service desk isn't proving you with an update, is it because there isn't one? Do were now have another 4 month wait? 

[yacoboy]

Hi, I've just created an account specifically so I can also add my comments on how this issue is ALSO affecting myself. 

I use a three mobile broadband SIM in a Three branded NR5103E. I've been experiencing the exact same issues as all the other users in this thread. Multiple websites connections failing, "ERR_CONNECTION_CLOSED" github.com, diy.com, even topcashback.com is giving me grief. It's now got to the point where my connection is almost unusable, banking apps are failing requests and I have to click retry 3 or 4 times before it actually does anything. My octopus energy account isn't loading and keeps logging me out. It's absolutely ridiculous.

This all started around 2 months ago. I've been using this as my main internet connection for over 12 months and it was flawless up until that point. These issues were non-existent and nothing in my network setup has changed.

As it's now got so bad, I've had to take drastic action. I've added another router to my network and switched the SIM to it, for info it's the ZTE MU5001 mobile 5G router, and all my problems are now resolved! All websites are loading normally and functioning as expected.

Just to note, I haven't removed the NR5103E from my network, it's still being used to broadcast wifi, I've just disabled DHCP and I'm using the Ethernet WAN port. This proves that it's not an issue with the three 5G network, as I'm still using the same SIM and mobile network, the only thing that's changed is the gateway.

In my eyes it's clear as day that this is a botched firmware update for the NR5103E which has caused these issues. This is not an isolated incident, it's affecting a huge number of customers and is simply unacceptable! Either roll out another firmware update to fix the issue, or roll the firmware back to the last known good version.

I'll be keeping my network setup as it is right now, as it's actually usable for a change. But, I will monitor this thread in the hopes that Three eventually decide to take some responsibility for the issue and address it. My current setup is not ideal, it leads to slower speeds and higher ping as it's a mobile broadband router not designed for long term home broadband use.

Honestly, Three should be ashamed of themselves. Especially @JonathanB, who seems to be brushing the issue under the carpet, dismissing the issue and blaming the "Service Desks", whoever they are! This needs escalating to Zyxel technical support immediately and resolving as a priority.

[hughesceri]

Very interesting that adding a ZTE MU5001 into the mix solves your issue.

While the majority of users on this thread are on the NR5103E, there are others who are using the ZTE MU5001 and still seeing issues. There's one mention of it happening on a Huawei H122-370 and I'm seeing the issue on my Huawei H122-373.

So, while the issue may be related to that version of the firmware on the Zyxel kit, it's strange that it also affects other brands of router.

For info, my H122-373 is on:

Hardware version WL1H122M
Software version 11.0.2.20(H329SP1C21)

Like others in this thread, the issues only affects me when I'm on a 5G network. When I drop to 4G the issue goes away - unfortunately, there's no setting I can find to always force 4G so I have to find increasingly shielded parts of the house to "hide" my router from the 5G signal - it always eventually finds it but I can sometimes get an hour of uninterrupted 4G use.

[yacoboy]

That's very strange indeed.

I've had no issues since changing to use the MU5001, my response time for loading websites is vastly improved and I'm using the 5G network. It's a Vodafone branded device, but unlocked, so works fine with Three.

My Software Version for info is: BD_VDFMU5001V1.0.0B05

Is your H122-373 Three branded or generic?

[MP]

I had this issue for many many months last year and I have the MU5001 with the same firmware version and Vodafone branded (it has never received a firmware update...).

The issue went away for me about a couple of months ago, it was prevalent before that and happened on basically every website I visited and I have a Three Business line.  The only website where I encounter the issue still is on royalmail.com

My contract is ending shortly and I will be thinking about moving away because I need reliablity.

[hughesceri]

It's Three branded

[kris]

Changing the router does not work. I've tried this few times. I used:

HUAWEI CPE PRO (3 different devices, with and without Three branding)

ZYXEL NR5103E (2 different devices)

ZYXEL NR5103EV2

I have access to many routers that's why I could give a go to so many of them.

On every one of them IT WAS DROPPING the connection, behaviour was exactly the same. I ordered LYCA mobile SIM (EE network) and hopefully that will solve the issue.

Three does not give a s**t to solve this problem

[BrummyGit]

I am currently flying back to the UK on a Singapore Airlines A380 and their WiFi service just gave me the exact same issue when hitting GitHub 

what’s the betting that their in-flight service uses Zyxel kit, or it’s related to some form of content filter. 

BTW I see this problem with my second router, but that’s also a Zyxel FWA017 so probably has mostly the same firmware code

[EDIflyer]

@JonathanB more information courtesy of another poster on the ISPReview forums - 

Hi EDIflyer, I signed up just to reply to you. I was running the NR5103Ev2 on Three and ran into the same issues. I am running 100s of API calls a min for work and Three was causing SSL issues with 50% or more of them based on my monitoring logs (even after enabling adult content, limiting the APN to IPV4 and 4G). I was annoyed and bought the Scancom EE long term sim that was on sale at Amazon and popped it into the NR5103Ev2 and changed the APN settings to EEs settings and IPV4. Its been running for 4+ hours now without any issues where before I could see constant issues on Three. The SSL errors are the result of something Three is doing. I would expect whatever DPI they have in place is breaking SSL. I will be dropping Three and keeping EE (I mean I paid for the long term SIM so thats a no brainer at this point for me)

P.S. I tested with my Smarty SIM in my mobile as well and got the same errors which is no surprise since they are essentially Three but figured I would add that so others don't think testing Smarty proves anything.

Example log:

File "XXXXXXXXXXXXXXXl/retry.py", line 592, in increment
raise MaxRetryError(_pool, url, error or ResponseError(cause))
urllib3.exceptions.MaxRetryError: HTTPSConnectionPool(host='XXXXXXXXXXXXXXX.com', port=443): Max retries exceeded with url:XXXXXXXXXXXXXXX
(Caused by SSLError(SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1106)')))

I note they tried many of the things that lots of us have tried too and the symptoms are also very similar.  Have the team properly examined the potential for deep packet inspection to be causing an issue? Activating a VPN seems to resolve the issue which would also seem to be a potential route.

It would be good to know where in the technical team this has reached - given the number of customers involved can you please confirmed this has been highlighted and escalated to the highest levels? I really hope it's not still stuck in a helpdesk somewhere!

[MP]

I do emphasise with the people still affected by this, I was affected by this for many many months and I now consider it to be fixed but I am a business customer.  They must have done something and I hope it can be applied to non-business customers too.