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.

[JonathanB]

Hi @EDIflyer,

The main challenge seems to be that the issue isn't reproducing for the teams we've engaged about the issue. There doesn't seem to be new threads cropping up about this, and to date we've had around 25 unique accounts posting on this thread. 

I'm sure the issue will be more widespread than this, but it doesn't seem to be a universal one, or perhaps as it's a quite random and intermittent issue the frequency of the errors occurring is in a very wide range, so it may well be flying under the radar for a lot of customers, whereas it's obviously causing a lot of disruption for the members posting here.

I've discussed internally how we get this escalated further, and it looks like the best next steps is to gather some fresh data about the frequency of the errors from you all directly. 

I'd suggest that if as many of you as possible can share either by PM to me or in reply to the thread. Gather a list over the next day or two which websites or services are triggering these end of file errors, and date/timestamps of when these are encountered. (Don't share any personal/private info on public responses, I'll get in touch if we need help correlating any info with the actual Three accounts).

This will give fresh and current data to share to the support teams, whilst also helping to explain the severity of the impact this is having on you all. Even better, if anyone is already keeping logs of the errors let me know and I can get this info shared to the teams sooner. If anyone prefers to PM info, you can reach me here: @JonathanB 

Thanks,
Jonathan

[tucu]

> to date we've had around 25 unique accounts posting on this thread

Simply because most of the people that are facing this issue are either
1. not technical enough to notice the issue (they simply refresh the page)
2. don't use the internet for technical stuff like most of the posters in this thread or
3. they don't even know how to report the issue.

The evidence is there @JonathanB - it's incredibly easy to reproduce. There was even a script that tech people can run. Is there evidence of any work / testing done at all on this issue? It's very easy to close an issue with a "can't reproduce" status

[EDIflyer]

Thanks for the post, @JonathanB .  I find it disappointing that the technical team haven't been able to make headway on this when it is so prevalent and also, for those affected, so reproducible.  My recollection was that a number of us previously supplied lots of data on this so I'm a little unsure why yet more data is needed?  As to why more members of the public haven't raised it - I suspect it's because most are less aware of the underlying technical aspects and thus just blame the website they're using rather than the connection.  The way that Github is particularly affected and VPNs resolve the issue will also lead to a more technical group of users being over-represented here. 

Data-wise do you just need date/time and server? If so then perhaps we could just have an agreed bash script to run that would generate the data the technical team are wanting?

[JonathanB]

I'm sure you're right, in that this will be affecting a greater number of users to a lesser extent. And I appreciate asking for more info again is frustrating, but the best way to get more urgency on this issue is for us to display that this is a frequent disruption.

Dates, times and the websites or services which seem to be returning the errors ought to help do this. 

As you've mentioned there's clearly a more technical ability and interest amongst customers on this thread, so if anyone is able to explain scripting which is consistently triggering the errors I can also share this feedback.

Thanks,
Jonathan

[MP]

Just wasted about an hour filling an online form just to encounter:  ERR_CONNECTION_CLOSED
Obviously not ideal that the form didn't auto-save but even worse for it to fault out because of an unreliable network service.  This is now having a negative financial impact so I will have to consider moving away if this is not fixed.

[robjames_1]

I got in touch with three via live chat tonight to cancel my contract following months of this issue. I was offered a replacement hub which I’ve decided to try out. Currently using a huawei h122-373. It’s arriving Wednesday so will update at that point to see if it’s made a difference. If not then I just need to wait for the 17th when my plusnet contract starts!

[AkiTaiyo]

Here's an odd one, might be worth trying if you can..
I moved the antenna (I use an external antenna) yesterday to improve my EE reception.. As a result my 3 signal dropped a bit (router does dual sim).. Now getting about 250/18Mbps (down/up) average compared to the previous 400/35.. But using Three internet today with a lot of Microsoft and Azure services, no 'End of file' errors, even with my original DNS settings..
Could still be coincidence though..

[AlSeek]

I have found setting the router in 4G only mode results in the error appearing less frequently than when in 5G mode, potentially could be related to the reduced speeds of 4G. Who knows ¯\_(ツ)_/¯
It does still happen in 4G mode though.

[AkiTaiyo]

Yep, was purely coincidence that I'd had a day without error.. about an hour after writing that post, back to errors again.

[bytespider]

I took get days where I think all is good and others where every other request gets EOF