Menu
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
264 Views
Message 1 of 21
Flag for a moderator

Derby area - unusual UDP packet/latency issue

Hi all,

For 3-4 weeks now, my Network Time Protocol (NTP) server has become extremely unreliable. When graphed it used to show a perfectly flat, horizontal line whereas now it looks like this:

Screen Shot 2017-03-19 at 4.26.09 pm.png

This is obviously not great. A BQM of my connection shows lots of latency spikes but no packet loss. If you bear in mind that from 12am to 3.30pm in the graph below my internet connection was idle (some background refreshes from an iPhone but nothing more), that seems very pointy:

51c05bf5ae1a797c1a161bd463dcd3ab-19-03-2017

I've checked all my connections, including tightness of the Virgin co-ax cables with their magic plastic spanner, I've rebooted the router multiple times, turned it off and then on again, restored it to factory settings, all with no luck. Below are my downstream and upstream figures as of 16.38, and my Network Log:

Screen Shot 2017-03-19 at 4.37.17 pm.png

Screen Shot 2017-03-19 at 4.37.43 pm.png

Screen Shot 2017-03-19 at 4.38.03 pm.png

My broadband speeds are fine and online gaming is generally okay. Any help/ideas on how to fix this UDP issue would be gratefully received.

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
246 Views
Message 2 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

Here's an updated BQM. I noticed there's a spike in the average latency every 6 hours or so, and there are also large spikes in the maximum latency every six hours or so, but not at the same time. For information purposes, the connection was in use from 4pm to 12am (web surfing, online gaming), but idle before and after that. I would expect to see a difference between the two but they appear, to all intents and purposes, to be identical.

2407dc77813bee9b8a96fd6ad8214614-20-03-2017

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
231 Views
Message 3 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

I should add that this is on a Superhub 2ac. Are the "critical" errors in my Network Log out of the ordinary? Should I request an engineer visit? Should I get back in my box? Can anyone, anyone at all, reply and offer an opinion?

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
224 Views
Message 4 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

To keep bumping this post until someone from Virgin replies, here are three traceroutes, to virginmedia.com, www.pool.ntp.org and to bbc.co.uk:

Screen Shot 2017-03-20 at 15.13.59.png

 

 

Screen Shot 2017-03-20 at 15.14.16.png

 

 

Screen Shot 2017-03-20 at 15.15.12.png

 

 

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
217 Views
Message 5 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

Okay. Last night my Superhub began ignoring settings changes. Port forwarding changes show up in the web interface but are ignored by the router. I attempted to reboot the router, again from the web interface, and the command was ignored. Switching the router off for 60 seconds and then back on has not made any difference. Then I checked the network log. I've had a fair few "No Ranging Response received" messages in the past few weeks, but I'm going to go out on a limb and say what's going on here is not good:Screen Shot 2017-03-21 at 07.56.08.png

I would appreciate a reply from a Virgin Media representative about this. I really, really don't want a Murdoch dish on the side of the house but my incessant whinging has been roundly ignored by the mods here. All I would like is an official response saying someone will look into the issue (followed by it being fixed, obviously).

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
209 Views
Message 6 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

Remarkably, in the last five hours things have got worse:

Screen Shot 2017-03-21 at 11.56.53.png

The service status page automated test reports that everything is fine. Is this a network thing or an impending router failure?

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
200 Views
Message 7 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

My Superhub's logs are now full of the "No Ranging Response received" error occurring dozens of times a second, and they've been like this for at least 45 minutes. There was a brief moment, though, when this happened:

Screen Shot 2017-03-21 at 12.17.30.png

So that was nice.

  • 62
  • 2
  • 10
deamondiamond
Dialled in
193 Views
Message 8 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

Post your upstream power levels again, are they still quite high as one of them was 50 which i'm sure is on VM's borderline "scale" of when it can cause issues, which would explain the T3 time-outs, if they are still high you'd need to get an engineer out to take a look. Reply from forum team is about 2 weeks so would be quicker to ring up. 

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
187 Views
Message 9 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

Thanks for the reply. Here's my current upstream figures:

My Superhub's logs briefly said this:

I know a router that's going to sleep well tonight...I know a router that's going to sleep well tonight...

But then returned to the "No Ranging Response received" errors.

0 Kudos
Reply
  • 23
  • 0
  • 5
ncguk
On our wavelength
173 Views
Message 10 of 21
Flag for a moderator

Re: Derby area - unusual UDP packet/latency issue

Well then. I've just spent nearly 25 minutes on the phone to Virgin Media being treated like the broadband equivalent of a hypochondriac and all I have to show for it is that my upload and download speeds are good and I have good ping. My BQM for today would beg to differ:

69383d53a7ab9b1a0c474f0b86e76fcd-21-03-2017

The red packet loss spikes can be discounted as they're caused by router reboots, but the yellow latency spikes from about 11.30am to 2pm are real and are happening on an idle connection at around the same time I started getting dozens of "No Ranging Response received - T3 time-out" network log errors per second on my Superhub.

In the meantime, when it should be a flat blue line up at around 20 on the right-hand scale, my NTP chart still looks like the world's most terrifying rollercoaster:

Screen Shot 2017-03-21 at 4.40.59 pm.png

I realise the front line phone answerers do their best but once you have a problem that strays from their script they quickly flounder. I asked for the issue to be escalated to the second line team but as my speeds and ping were good, that isn't going to happen. So, here I am.

0 Kudos
Reply