120M New upstream == pings over 500??!?

Rexz · 10-10-2013, 11:27

So I lost connection about 5am this morning according to logs and had to power cycle the modem when I woke up to get a connection back. Upon checking the modem I now have a new upstream channel (now 2). Unfortunately with has brought pings from hell. I am now jumping from 21ms pinging bbc.co.uk to upwards of over 1000ms and back down. Even pfsense is reporting a RTT to the gateway of over 200ms which is usually only 8ms. Just incase they broke my connection include a screenshot of my power levels. Hopefully someone can help

Sephiroth · 10-10-2013, 11:35

To complete the picture, a tracert to bbc.co.uk would be appreciated.

Rexz · 10-10-2013, 11:55

Here you go

C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC>tracert bbc.co.uk

Tracing route to bbc.co.uk [212.58.253.67]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms pfsense.firewall.home [192.168.2.254]
2 367 ms 7 ms 9 ms 10.14.88.1
3 928 ms 9 ms 9 ms basl-core-2b-ae10-722.network.virginmedia.net [81.96.2.233]
4 443 ms 163 ms 11 ms popl-bb-1b-ae11-0.network.virginmedia.net [81.96.0.149]
5 93 ms 13 ms 14 ms nrth-bb-1b-et-010-0.network.virginmedia.net [62.253.175.57]
6 588 ms 12 ms 276 ms tele-ic-4-ae0-0.network.virginmedia.net [62.253.174.18]
7 13 ms 14 ms 14 ms pos6-1.rt0.thdo.bbc.co.uk [212.58.239.237]
8 * * * Request timed out.
9 729 ms 425 ms 17 ms ae0.er01.cwwtf.bbc.co.uk [132.185.254.93]
10 606 ms 15 ms 480 ms 132.185.255.156
11 22 ms 22 ms 17 ms www-vip.cwwtf.bbc.co.uk [212.58.253.67]

Trace complete.

---------- Post added at 10:55 ---------- Previous post was at 10:44 ----------

Here is a screenshot of my pings too.

Sephiroth · 10-10-2013, 12:19

Quote:

Originally Posted by Rexz

So I lost connection about 5am this morning according to logs and had to power cycle the modem when I woke up to get a connection back. Upon checking the modem I now have a new upstream channel (now 2). Unfortunately with has brought pings from hell. I am now jumping from 21ms pinging bbc.co.uk to upwards of over 1000ms and back down. Even pfsense is reporting a RTT to the gateway of over 200ms which is usually only 8ms. ls. ......

First to say the power levels are fine.

It seems that there is delay at your gateway that propagates, albet not all the time on the tracert you've provided. A good diagnostic goin=g forward is the Thinkbroadband Quality Monitor which usually flushes out upstream issues.

I'm guessing here but based on other people's reports, they've either misconfigured the CMTS when they added the upstream channel, or despite the change there is upstream noise or congestion. One might even suggest a one day settling down period while they monitor things at the local head end (Basildon).

In the network log, are there any T3 events (other than when it was rebooting)?

Rexz · 10-10-2013, 12:37

I had a lot of t3 events at near 4am this morning and the internet was out when I woke up at 8. Infact just power cycled the modem and im left with no connection now. Hmm guess I wont be working today.

---------- Post added at 11:37 ---------- Previous post was at 11:30 ----------

Here is thinkbroadband.

and here is part of the SH2 log from this morning

Kushan · 10-10-2013, 13:02

Wow, hell of an issue there for sure. Might be worth giving them a call and hoping you get through to a somewhat competent agent (or at least one that's willing to speak to second line).

Sephiroth · 10-10-2013, 13:13

I was looking for T3s during normal running. Your log doesn't show any. So it's not a congested upstream, IMO. It seems to me that they've not left your upstream circuit in a state of grace. Packet loss indicates npoise on the upstream.

If you call VM, I recommend you put the phone down on an offshore agent! You can also publish this lot in the VM forums whgere a VM tech will check your circuit and let you know what's amiss.

Rexz · 10-10-2013, 13:24

Thanks for the replies guys. I have now posted all the images and an explanation on the VM forums. Hopefully someone will help. Also trying to phone to explain this will be a killer, especially if I get offshore....."have you tried rebooting the modem?"...... "&%#£!"

Sephiroth · 10-10-2013, 13:27

Put the phone down on offshore.

Kushan · 10-10-2013, 13:31

Quote:

Originally Posted by Sephiroth

I was looking for T3s during normal running. Your log doesn't show any. So it's not a congested upstream, IMO. It seems to me that they've not left your upstream circuit in a state of grace. Packet loss indicates npoise on the upstream.

If you call VM, I recommend you put the phone down on an offshore agent! You can also publish this lot in the VM forums whgere a VM tech will check your circuit and let you know what's amiss.

I don't really want to throw a spanner in the works here, but I was speaking to an old friend who still works for Virgin and they were of the opinion that Swansea these days were no better trained than offshore.

I've not had to deal with either so I can't really comment, but I would suggest maybe NOT just putting the phone down right away.

Posting on the VM forums may take a bit longer but would be the best way to ensure it gets to the right people.

Sephiroth · 10-10-2013, 13:42

Put the phone down oin the offshore agent and let us know how you get on with the UK agent!

jarrodstrachan · 10-10-2013, 19:10

I'm certain that the Swansea team is better than the offshore. Simply because they cant be any worse

Rexz · 10-10-2013, 23:57

Ok kind of serious now. Not sure if ringing tech support will help but I havnt had any connection for most of the day and my modem gets stuck in provisioning state in progress. Tomorrow will be another lost day of work.... anyone know what that means?

---------- Post added at 22:57 ---------- Previous post was at 21:15 ----------

Ok it seems power cycling the modem about 10 times got it working. Sometimes it wouldn't provision and then others it wouldn't supply an IP. Guess some channels are borked or something. Pings are slightly better but still spiking. Just waiting for the technicians to get back to me on the VM forums. Thanks for your help guys!

Sephiroth · 11-10-2013, 00:18

Rebooting the modem is a natural temptation and by all means continue - it doesn't nbring a borked network back.

It's what they're doing in the network that'll fix it. It's shocking though that VM's systems are unable to tell you what's going on.

Chrysalis · 11-10-2013, 06:19

Kushan I dont think its about the training its about the mindset. Offshore agents tend to rigidly follow scripts whilst onshore tend to use common sense and decide to skip the script against what they been trained to do but less frustration for them and customers. Plus there is issue where offshore can be hard to understand due to bad line quality and accent/language issues.

I agree with seph that I dont think its a typical congestion situation, looks like a borked network upgrade.