50M Think Broadband Ping Monitor Results (POST YOURS) - Page 156

roughbeast · 12-10-2012, 22:43

Quote:

Originally Posted by ferretuk

www.conspiracytheory.com is that way ------->

I object to that comment.

I'm not into conspiracy theories! Only facts. Apologise at your leisure.

Until yesterday, when the effects upon my service became clear I was defending VM to the hilt against habitual VM bashers and conspiracy theorists. My service has been virtually perfect for years.

I just got off the phone to CS. It has now been explained to me very clearly that there is a specific problem with servers (sic) in the country giving effects nationwide but not everywhere. They have been attempting to fix the problem for a few days and it is likely to continue for a week.

The reason, I am now told, that the effect is so pronounced in the evenings is because of normal increased traffic as the evening progresses and that VM engineers are deliberately reducing service, at the flick of a switch, nation-wide. This, and I am using their words, is to prevent 'network collapse'. A few areas are unaffected.

To compensate me my bill is being reduced by £20 this month. Tokenism?

We shall see shan't we? I will desist from ringing CS for a week, fully expecting an improvement after that time.

Isn't it a pity that VM didn't forewarn its customers about this, so we could ,where possible, make adjustments to our usage .

Coffeeguy · 12-10-2012, 23:20

Quote:

Originally Posted by roughbeast

<snip>I just got off the phone to CS</snip>

I wonder what 'your man in Bangalore' was smoking because I'd love some of it. My man in Mumbai told me VM were testing the Superhub 3.

roughbeast · 12-10-2012, 23:27

Quote:

Originally Posted by Coffeeguy

I wonder what 'your man in Bangalore' was smoking because I'd love some of it. My man in Mumbai told me VM were testing the Superhub 3.

Jeez. They tell us anything that comes into their heads.

What was he referring to? How can testing Superhub 3 degrade so many customer's service every evening?

Qtx · 12-10-2012, 23:29

Would be nice to know a fault reference number for this....

Coffeeguy · 12-10-2012, 23:47

Quote:

Originally Posted by roughbeast

Jeez. They tell us anything that comes into their heads.

What was he referring to? How can testing Superhub 3 degrade so many customer's service every evening?

He mentioned something about it having 'Super' powers but wouldn't elaborate

Qtx · 13-10-2012, 00:13

Quote:

Originally Posted by Coffeeguy

He mentioned something about it having 'Super' powers but wouldn't elaborate

Unfortunately the technicians were not thinking straight when they put the krytponite component in the super-hub...

qasdfdsaq · 13-10-2012, 00:45

Quote:

Originally Posted by roughbeast

Which server was that?

Quote:

Ping test not working. It is failing to find a ping server good enough it seems.
So here's a trace route to the ping test server selected as being the best!

tracert xilo.net

Tracing route to xilo.net [91.199.78.20]
over a maximum of 30 hops:

The Xilo pingtest server isn't at xilo.net

It's at 213.229.119.169

---------- Post added at 00:45 ---------- Previous post was at 00:44 ----------

Quote:

Originally Posted by craigj2k12

Thats not what happening here

Well actually... The amount of increase in latency is far higher than one would expect from any saturated link, and looks more like what you'd expect from an overloaded firewall or DPI module. That and the consistency of its effects across VM's entire network do make me wonder...

craigj2k12 · 13-10-2012, 04:06

Quote:

Originally Posted by qasdfdsaq

Well actually... The amount of increase in latency is far higher than one would expect from any saturated link, and looks more like what you'd expect from an overloaded firewall or DPI module. That and the consistency of its effects across VM's entire network do make me wonder...

Expand....................

kwikbreaks · 13-10-2012, 08:18

I'm guessing qas means....

Maybe they are testing their new shaping policies?

If too many are evading NNTP + P2P shaping by using VPNs perhaps they are trying some real DPI but haven't got the grunt to support it.

roughbeast · 13-10-2012, 09:16

Quote:

Originally Posted by qasdfdsaq

Which server was that?

The Xilo pingtest server isn't at xilo.net

It's at 213.229.119.169

---------- Post added at 00:45 ---------- Previous post was at 00:44 ----------

Well actually... The amount of increase in latency is far higher than one would expect from any saturated link, and looks more like what you'd expect from an overloaded firewall or DPI module. That and the consistency of its effects across VM's entire network do make me wonder...

Firstly, that speed test was NAMESCO London, the one that gives me the most consistent results; usually 104Mb. The bump is having a significant effect on my service.

Thanks regarding Xilo. I just went for the obvious.

Whatever the reason for the action VM is taking during the evening, it is clear that it is calculated. It isn't just the natural effect of congestion. It commences and ceases too abruptly for that. The CS explanation given me is that they are having temporary server problems and they are merely reducing the service at that time to avoid meltdown whilst they fix it. Pinch of salt taken before ringing them.

Quote:

Originally Posted by Qtx

Would be nice to know a fault reference number for this....

Yes it would. Since there is no fault with my local connection or network I was not given one.

Qtx · 13-10-2012, 10:59

Quote:

Originally Posted by roughbeast

Yes it would. Since there is no fault with my local connection or network I was not given one.

For this exact hump problem, they have given the fault ref F002195199 but it only is valid for those in the Worthing area, even though everyone else in the thread has told them its not limited to worthing....

japitts · 13-10-2012, 11:33

This is pretty typical for me in BS37 at the moment...

joglynne · 13-10-2012, 11:53

Maybe we should all go on the VM Forum and report these humps en masse. Be a tad difficult for VM to dismiss reports from all over the country and, with other members over there seeing what we report, it may set the ball moving for VM to have to have come clean about what appears to be an extremely widespread issue.

I was one of the earliest members on there when it was first set up and I have seen posts deleted in the past when they asked too many embarrassing questions but don't know whether that is still done?

ferretuk · 13-10-2012, 12:23

Quote:

Originally Posted by joglynne

Maybe we should all go on the VM Forum and report these humps en masse. Be a tad difficult for VM to dismiss reports from all over the country and, with other members over there seeing what we report, it may set the ball moving for VM to have to have come clean about what appears to be an extremely widespread issue.

I was one of the earliest members on there when it was first set up and I have seen posts deleted in the past when they asked too many embarrassing questions but don't know whether that is still done?

See here and others...

The way the VM Community forum splits speed issues between tiers doesn't help present a wider view of course,

Traduk · 13-10-2012, 12:46

I am getting the same bumps off Guildford (Surrey). They started on the 1st October and were preceded by a 2 hour outage on the 29th September which is highly probably when my area was upgraded to double upload.

In the past (months ago) there were two national outages following changes in down and up channels when we went from 4 to 5 to 8 down and up to 2 up.

As a casual observation it does look as though changes in user configuration causes major network problems and hopefully these bumps are a short term deliberate effect until something can be sorted out. There will be all hell to pay if they become a regular feature because mass complaints will be evident if we all suffer prime time useless service.

I did not really notice the reduction in what was an excellent service because over the past few weeks I have been using BT Infinity for most of my internet work.

[img]Download Failed (1)[/img]