[Merged] Manchester Issues - Page 3

Ignition · 04-12-2005, 15:29

Quote:

Originally Posted by Florence

I know they keep saying there is enough to manage 10mb but the faster you go the more the network needs to be spot on and unfortunately the bean counters at NTL seem to think selling staff is more important than engineers that keep things ticking over.

Wrong, the 'speed' getting chucked down the network is the same whether users are on 128k or 10Mbit, as you all share a part of a larger channel, and it's that channel and its' modulation that set how high a quality the network needs to be.

You get a slice of that larger channel however your modem is constantly receiving data from el uBR.

Sounds more like bandwidth congestion (clue being slowdowns at peak times), which isn't really an issue to do with local network maintenance more one for capacity planning. Some dodgy signal issues there potentially, with regard to connection stability, these can be checked out through the normal channels, and logs from modems will tell the story.

Quote:

Fri, 02 Dec 2005 20:49:42 GMT
1st 128K took 735 ms = 178329 Bytes/sec = approx 1484 kbits/sec
2nd 128K took 390 ms = 336082 Bytes/sec = approx 2796 kbits/sec
3rd 128K took 563 ms = 232810 Bytes/sec = approx 1937 kbits/sec
4th 128K took 547 ms = 239620 Bytes/sec = approx 1994 kbits/sec

RDHW's test is not supposed to be a reliable test, note your 2nd test's speeds are impossible, and averaging them all out gives you a quite acceptable speed across all 4 tests.

Use ADSLGuide or another speedtest but not RDHW's it was designed to be a basic guesstimator, its' sample sizes are too small for the higher speed services really, and can be thrown wildly out of kilter too easily.

craig5320 · 04-12-2005, 15:37

Ignition, would you happen to know if this graph is the result of bandwidth congestion?

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

I've just taken this shot about 5 minutes ago. This is what I get constantly 24/7 not just onpeak, i've tested right through the night until 5am, 9am through to 12pm, any time of day, this is what i get. This is downloading a 50mb file from my webspace on ntl. But happens what ever site i use, even my newshost, and it's been like this for nearly two weeks now.

Chrysalis · 04-12-2005, 16:57

The throughput drops to 0 whilst you downloading?

Rone · 04-12-2005, 19:40

I found hardly any webpages would even open last week, today its been almost perfect.
On friday a very helpful tech guy did every test he could, and concluded my modem, and everything my end were functioning ok.
That only leaves the network and\or congestion. Assuming it only got really bad due to the upgrade work [and could that be the cause?], i am hoping things might settle down.

Ignition · 04-12-2005, 19:57

Quote:

Originally Posted by craig5320

Ignition, would you happen to know if this graph is the result of bandwidth congestion?

I've just taken this shot about 5 minutes ago. This is what I get constantly 24/7 not just onpeak, i've tested right through the night until 5am, 9am through to 12pm, any time of day, this is what i get. This is downloading a 50mb file from my webspace on ntl. But happens what ever site i use, even my newshost, and it's been like this for nearly two weeks now.

That doesn't look like congestion no, congestion you'd see fluctuating speeds not all or nothing as you are getting.

altis · 04-12-2005, 20:29

eBay has virtually ground to a halt for me. I do hope something happens about this soon.

th'engineer · 04-12-2005, 20:52

just when you think that you would escape the recent problems

Sun, 4 Dec 2005 19:50:16 UTC
1st 128K took 532 ms = 246376 Bytes/sec = approx 2050 kbits/sec
2nd 128K took 531 ms = 246840 Bytes/sec = approx 2054 kbits/sec
3rd 128K took 562 ms = 233224 Bytes/sec = approx 1940 kbits/sec
4th 128K took 3172 ms = 41322 Bytes/sec = approx 344 kbits/sec

A load of bobbins again

Target Name: www.ntlworld.com
IP: 212.250.162.47
Date/Time: 04/12/2005 19:53:37
1 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms [192.168.123.254]
2 6 ms 15 ms 7 ms 8 ms 6 ms 6 ms 11 ms * 7 ms 16 ms [10.23.48.1]
3 28 ms 12 ms 16 ms 52 ms 12 ms 102 ms 9 ms * 16 ms 8 ms oldh-t2cam1-b-v115.inet.ntl.com [80.5.164.189]
4 12 ms 29 ms 7 ms 10 ms 8 ms 71 ms 13 ms 10 ms 12 ms 11 ms manc-t3core-1b-ge-111-0.inet.ntl.com [213.104.242.205]
5 24 ms 15 ms 9 ms 8 ms 65 ms 40 ms 12 ms 18 ms 9 ms 24 ms man-bb-b-so-400-0.inet.ntl.com [213.105.242.245]
6 13 ms 18 ms 16 ms 13 ms 44 ms 18 ms 28 ms 18 ms 17 ms 17 ms win-bb-a-so-300-0.inet.ntl.com [62.253.185.138]
7 21 ms 22 ms 14 ms 20 ms 17 ms 14 ms 114 ms 15 ms 15 ms 18 ms win-dc-a-v902.inet.ntl.com [62.253.187.222]
8 15 ms 24 ms 19 ms 18 ms 20 ms 13 ms 100 ms 15 ms 21 ms 19 ms www.ntlworld.com [212.250.162.47]

this might help

Florence · 04-12-2005, 21:04

Quote:

Originally Posted by th'engineer

just when you think that you would escape the recent problems

Sun, 4 Dec 2005 19:50:16 UTC
1st 128K took 532 ms = 246376 Bytes/sec = approx 2050 kbits/sec
2nd 128K took 531 ms = 246840 Bytes/sec = approx 2054 kbits/sec
3rd 128K took 562 ms = 233224 Bytes/sec = approx 1940 kbits/sec
4th 128K took 3172 ms = 41322 Bytes/sec = approx 344 kbits/sec

A load of bobbins again

Target Name: www.ntlworld.com
IP: 212.250.162.47
Date/Time: 04/12/2005 19:53:37
1 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms [192.168.123.254]
2 6 ms 15 ms 7 ms 8 ms 6 ms 6 ms 11 ms * 7 ms 16 ms [10.23.48.1]
3 28 ms 12 ms 16 ms 52 ms 12 ms 102 ms 9 ms * 16 ms 8 ms oldh-t2cam1-b-v115.inet.ntl.com [80.5.164.189]
4 12 ms 29 ms 7 ms 10 ms 8 ms 71 ms 13 ms 10 ms 12 ms 11 ms manc-t3core-1b-ge-111-0.inet.ntl.com [213.104.242.205]
5 24 ms 15 ms 9 ms 8 ms 65 ms 40 ms 12 ms 18 ms 9 ms 24 ms man-bb-b-so-400-0.inet.ntl.com [213.105.242.245]
6 13 ms 18 ms 16 ms 13 ms 44 ms 18 ms 28 ms 18 ms 17 ms 17 ms win-bb-a-so-300-0.inet.ntl.com [62.253.185.138]
7 21 ms 22 ms 14 ms 20 ms 17 ms 14 ms 114 ms 15 ms 15 ms 18 ms win-dc-a-v902.inet.ntl.com [62.253.187.222]
8 15 ms 24 ms 19 ms 18 ms 20 ms 13 ms 100 ms 15 ms 21 ms 19 ms www.ntlworld.com [212.250.162.47]

this might help

What do you expect when they cut back on engineers and only employ sales staff

suppose after December the only way is downhill fast for the northwest. Lets see if NTL can prove us wrong and employ some BB engineers as withjout them the network will be bad and the customers will leave

ant78 · 04-12-2005, 21:36

im getting crap speeds like 36kbs when im downloading demos, does this look alright, its the speed test results from the link at the top of the page,

Sun, 4 Dec 2005 20:36:18 UTC

1st 512K took 1750 ms = 299593 KB/sec, approx 2469 Kbps, 2.41 Mbps
2nd 512K took 1859 ms = 282027 KB/sec, approx 2324 Kbps, 2.27 Mbps
3rd 512K took 2391 ms = 219276 KB/sec, approx 1807 Kbps, 1.76 Mbps
4th 512K took 2484 ms = 211066 KB/sec, approx 1689 Kbps, 1.65 Mbps

Overall Average Speed = approx 2072 Kbps, 2.02 Mbps

im on 10meg and my old 2meg seems faster and more stable

Ignition · 04-12-2005, 23:18

Quote:

Originally Posted by Florence

What do you expect when they cut back on engineers and only employ sales staff

suppose after December the only way is downhill fast for the northwest. Lets see if NTL can prove us wrong and employ some BB engineers as withjout them the network will be bad and the customers will leave

Sorry to say this but Bill C, who I assume you are referring to, is not a network engineer, nor do broadband engineers maintain and service the local networks in the North West.

---------- Post added at 22:09 ---------- Previous post was at 22:04 ----------

Quote:

Originally Posted by th'engineer

just when you think that you would escape the recent problems

Sun, 4 Dec 2005 19:50:16 UTC
1st 128K took 532 ms = 246376 Bytes/sec = approx 2050 kbits/sec
2nd 128K took 531 ms = 246840 Bytes/sec = approx 2054 kbits/sec
3rd 128K took 562 ms = 233224 Bytes/sec = approx 1940 kbits/sec
4th 128K took 3172 ms = 41322 Bytes/sec = approx 344 kbits/sec

A load of bobbins again

Target Name: www.ntlworld.com
IP: 212.250.162.47
Date/Time: 04/12/2005 19:53:37
1 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms [192.168.123.254]
2 6 ms 15 ms 7 ms 8 ms 6 ms 6 ms 11 ms * 7 ms 16 ms [10.23.48.1]
3 28 ms 12 ms 16 ms 52 ms 12 ms 102 ms 9 ms * 16 ms 8 ms oldh-t2cam1-b-v115.inet.ntl.com [80.5.164.189]
4 12 ms 29 ms 7 ms 10 ms 8 ms 71 ms 13 ms 10 ms 12 ms 11 ms manc-t3core-1b-ge-111-0.inet.ntl.com [213.104.242.205]
5 24 ms 15 ms 9 ms 8 ms 65 ms 40 ms 12 ms 18 ms 9 ms 24 ms man-bb-b-so-400-0.inet.ntl.com [213.105.242.245]
6 13 ms 18 ms 16 ms 13 ms 44 ms 18 ms 28 ms 18 ms 17 ms 17 ms win-bb-a-so-300-0.inet.ntl.com [62.253.185.138]
7 21 ms 22 ms 14 ms 20 ms 17 ms 14 ms 114 ms 15 ms 15 ms 18 ms win-dc-a-v902.inet.ntl.com [62.253.187.222]
8 15 ms 24 ms 19 ms 18 ms 20 ms 13 ms 100 ms 15 ms 21 ms 19 ms www.ntlworld.com [212.250.162.47]

this might help

You missed a 102ms spike in hop 3, a 71ms and a 29ms variation in hop 4, a 40ms spike to hop 5, a 28ms spike to hop 6, a 114ms spike to hop 7 and coloured a 21ms response in hop 8 but missed a 24ms one.

If this was done via wireless to your USR router try it wired instead or check signal for variations, alternatively supply something a little more concrete as far as proof of this affecting service, I note no packet loss to end destination indicating no ongoing packet loss and just drops to routers here and there.

Some variation with regard to latency is sadly an inevitable part of speed upgrades, and aspects of the upgrades aren't complete anywhere in the country. When all is done this might well stabilise your service.

Alternatively follow standard troubleshooting procedures for yourself and start calling directors of ntl

---------- Post added at 22:18 ---------- Previous post was at 22:09 ----------

Code:

                                           Packets               Pings
Hostname                                %Loss  Rcv  Snt  Last Best  Avg  Worst
 4. tele-ic-1-ge-200-205.inet.ntl.com      0%   51   51     1    0    2     10
 5. nth-bb-b-so-600-0.inet.ntl.com         0%   51   51     6    3    6     26
 6. lee-bb-a-so-600-0.inet.ntl.com         0%   51   51     6    6   10     74
 7. lee-bb-b-ae0-0.inet.ntl.com            0%   51   51    10    6   11     61
 8. man-bb-a-so-700-0.inet.ntl.com         0%   50   50    10    7    9     32
 9. glfd-bam-1-atm403-1.inet.ntl.com       0%   50   50     8    7    9     24
10. oldh-t2cam1-a-ge-wan32.inet.ntl.com    0%   50   50     9    7   12     73
11. ubr01roch.inet.ntl.com                 2%   49   50    10    8   10     15

--- ubr01roch.inet.ntl.com ping statistics ---
100 packets transmitted, 100 packets received, 0% packet loss
round-trip min/avg/max/stddev = 8.090/10.302/15.299/1.504 ms

Looks like the baby is deprioritising rather than having a genuine issue.

Might be a local fault, however people are all in generally different networks with no real crossover points at the *access* level.

Core level appears at a very casual inspection to be ok.

04-12-2005, 23:41

Quote:

Originally Posted by thaiengineer

just when you think that you would escape the recent problems

Sun, 4 Dec 2005 19:50:16 UTC
1st 128K took 532 ms = 246376 Bytes/sec = approx 2050 kbits/sec
2nd 128K took 531 ms = 246840 Bytes/sec = approx 2054 kbits/sec
3rd 128K took 562 ms = 233224 Bytes/sec = approx 1940 kbits/sec
4th 128K took 3172 ms = 41322 Bytes/sec = approx 344 kbits/sec

A load of bobbins again

Target Name: www.ntlworld.com
IP: 212.250.162.47
Date/Time: 04/12/2005 19:53:37
1 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms 0 ms [192.168.123.254]
2 6 ms 15 ms 7 ms 8 ms 6 ms 6 ms 11 ms * 7 ms 16 ms [10.23.48.1]
3 28 ms 12 ms 16 ms 52 ms 12 ms 102 ms 9 ms * 16 ms 8 ms oldh-t2cam1-b-v115.inet.ntl.com [80.5.164.189]
4 12 ms 29 ms 7 ms 10 ms 8 ms 71 ms 13 ms 10 ms 12 ms 11 ms manc-t3core-1b-ge-111-0.inet.ntl.com [213.104.242.205]
5 24 ms 15 ms 9 ms 8 ms 65 ms 40 ms 12 ms 18 ms 9 ms 24 ms man-bb-b-so-400-0.inet.ntl.com [213.105.242.245]
6 13 ms 18 ms 16 ms 13 ms 44 ms 18 ms 28 ms 18 ms 17 ms 17 ms win-bb-a-so-300-0.inet.ntl.com [62.253.185.138]
7 21 ms 22 ms 14 ms 20 ms 17 ms 14 ms 114 ms 15 ms 15 ms 18 ms win-dc-a-v902.inet.ntl.com [62.253.187.222]
8 15 ms 24 ms 19 ms 18 ms 20 ms 13 ms 100 ms 15 ms 21 ms 19 ms www.ntlworld.com [212.250.162.47]

this might help

What you going to blame this on. You are not on a proxy ?. Those readings are not great but there not that bad . ?

Ignition · 04-12-2005, 23:51

Ooo Oldham exchange:

Local loop unbundling presence
Bulldog: Available
Easynet: Available

UKOnline and Bulldog both available, ADSL2+ has shown promising results both on shorter and longer lines.

craig5320 · 05-12-2005, 01:45

Quote:

Originally Posted by Chrysalis

The throughput drops to 0 whilst you downloading?

Yep, it's all or nothing. this is using either my pc or laptop, with or without router, both have been throughly checked for spyware, viruses, etc. but even if it was spyware using the connection the issue then would be with my programs and not shown in bandwidth monitor to like that.

I even tried a friends laptop via my router and got the same results.

NTL say everything is fine.

th'engineer · 05-12-2005, 07:27

Quote:

Originally Posted by Ignition

Sorry to say this but Bill C, who I assume you are referring to, is not a network engineer, nor do broadband engineers maintain and service the local networks in the North West..

Wonder if it will be network engineers or the broadband engineers contracted out next.
First they came for the CSrs and no one complained then they came fot the Service engineers/techs who next .

Contracting out does not work unless its controlled properley and thats from someone who works in an industry that has done in many times.
Contractors just take the money for the least work then sting you for extra work that you forgot to specify at large bucks rates.
Not contract staff are better value and will do more fot the £ if managed correctly

Quote:

Originally Posted by Ignition

You missed a 102ms spike in hop 3, a 71ms and a 29ms variation in hop 4, a 40ms spike to hop 5, a 28ms spike to hop 6, a 114ms spike to hop 7 and coloured a 21ms response in hop 8 but missed a 24ms one.

If this was done via wireless to your USR router try it wired instead or check signal for variations, alternatively supply something a little more concrete as far as proof of this affecting service, I note no packet loss to end destination indicating no ongoing packet loss and just drops to routers here and there.

Some variation with regard to latency is sadly an inevitable part of speed upgrades, and aspects of the upgrades aren't complete anywhere in the country. When all is done this might well stabilise your service.

Alternatively follow standard troubleshooting procedures for yourself and start calling directors of ntl

..

Thank you for the correction not perfect all the time

just wanted to check it was not a "local issue" before starting normal procedures

Quote:

Originally Posted by Ignition

Code:

                                           Packets               Pings
Hostname                                %Loss  Rcv  Snt  Last Best  Avg  Worst
 4. tele-ic-1-ge-200-205.inet.ntl.com      0%   51   51     1    0    2     10
 5. nth-bb-b-so-600-0.inet.ntl.com         0%   51   51     6    3    6     26
 6. lee-bb-a-so-600-0.inet.ntl.com         0%   51   51     6    6   10     74
 7. lee-bb-b-ae0-0.inet.ntl.com            0%   51   51    10    6   11     61
 8. man-bb-a-so-700-0.inet.ntl.com         0%   50   50    10    7    9     32
 9. glfd-bam-1-atm403-1.inet.ntl.com       0%   50   50     8    7    9     24
10. oldh-t2cam1-a-ge-wan32.inet.ntl.com    0%   50   50     9    7   12     73
11. ubr01roch.inet.ntl.com                 2%   49   50    10    8   10     15
 
--- ubr01roch.inet.ntl.com ping statistics ---
100 packets transmitted, 100 packets received, 0% packet loss
round-trip min/avg/max/stddev = 8.090/10.302/15.299/1.504 ms

Looks like the baby is deprioritising rather than having a genuine issue.

Might be a local fault, however people are all in generally different networks with no real crossover points at the *access* level.

Core level appears at a very casual inspection to be ok.

Any other checks suggested please specify

Quote:

Originally Posted by Bill C

Those readings are not great but there not that bad . ?
Those readings are not great but there not that bad . ?

Suggestions on improving them appreciated

garybuk · 05-12-2005, 10:03

Just a quick note on using Treacert's and Pings for seeing how well the network is performing:

I work with Cisco kit and have a few 25xx's at home and they dynamically rate the traffic so under times of heavy load or if the router is doing something then pings go up, this doesn't necessarily mean the networks going slow, just the router 'decides' to reply a bit later than normal, this won't happen in real data situations. It's called Prioritisation and IMCP comes last in most cases.