MWEB Uncapped ADSL Feedback - Part 2

Thanks for the feedback, let me know if the problem with packet loss returns

Soulcode said:

And now it's fine again:

C:\Users\Werner>pathping /q 250 /p 50 bbc.co.uk

Tracing route to bbc.co.uk [212.58.241.131]
over a maximum of 30 hops:
0 Werner-PC [192.168.1.107]
1 192.168.1.1
2 41-133-128-1.dsl.mweb.co.za [41.133.128.1]
3 tengig-0-0-0-110.vic-ipc-1.mweb.co.za [196.22.169.170]
4 vl-92.vic-hscore-1.mweb.co.za [196.22.189.2]
5 tengig0-0-0-2.11.vic-p-1.mweb.co.za [196.22.169.225]
6 pos0-3-2-0.lon-up-1.mweb.co.za [197.80.5.7]
7 176.67.177.131
8 xe-8-1-0.edge3.London1.Level3.net [195.50.124.133]
9 vl-3604-ve-228.csw2.London1.Level3.net [4.69.166.157]
10 ae-2-52.edge4.London1.Level3.net [4.69.139.106]
11 BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]
12 * * *
Computing statistics for 137 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 Werner-PC [192.168.1.107]
0/ 250 = 0% |
1 0ms 0/ 250 = 0% 0/ 250 = 0% 192.168.1.1
0/ 250 = 0% |
2 28ms 0/ 250 = 0% 0/ 250 = 0% 41-133-128-1.dsl.mweb.co.za [41.133.128.1]
0/ 250 = 0% |
3 28ms 0/ 250 = 0% 0/ 250 = 0% tengig-0-0-0-110.vic-ipc-1.mweb.co.za [196.22.169.170]
0/ 250 = 0% |
4 28ms 0/ 250 = 0% 0/ 250 = 0% vl-92.vic-hscore-1.mweb.co.za [196.22.189.2]
0/ 250 = 0% |
5 30ms 0/ 250 = 0% 0/ 250 = 0% tengig0-0-0-2.11.vic-p-1.mweb.co.za [196.22.169.225]
0/ 250 = 0% |
6 210ms 0/ 250 = 0% 0/ 250 = 0% pos0-3-2-0.lon-up-1.mweb.co.za [197.80.5.7]
0/ 250 = 0% |
7 212ms 0/ 250 = 0% 0/ 250 = 0% 176.67.177.131
0/ 250 = 0% |
8 211ms 0/ 250 = 0% 0/ 250 = 0% xe-8-1-0.edge3.London1.Level3.net [195.50.124.133]
0/ 250 = 0% |
9 215ms 0/ 250 = 0% 0/ 250 = 0% vl-3604-ve-228.csw2.London1.Level3.net [4.69.166.157]
0/ 250 = 0% |
10 243ms 0/ 250 = 0% 0/ 250 = 0% ae-2-52.edge4.London1.Level3.net [4.69.139.106]
0/ 250 = 0% |
11 242ms 0/ 250 = 0% 0/ 250 = 0% BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]

Trace complete.


The ping in WoW is 30ms higher than always, but at least there seems to be no packet loss.
The packet loss has come and gone a couple of times though, so I'm not gonna breathe a sigh of relief just yet.

Click to expand...

Same thing as last night and at a similar time too, getting high ping to my local server (Cape Town: http://www.pingtest.net/result/70800503.png - I usually get around 20ms I think) and unplayable numbers to EUW (Frankfurt: http://www.pingtest.net/result/70800555.png - I usually get ~170ms). It went back to normal at about 2am last night, am I being unusually shaped or throttled? Wouldn't really make sense to me considering I'm not a heavy user.

MWEB Guy said:

Good Evening AcidRazor, there was a problem in JHB earlier today. Please let me know if you have rebooted your router as yet?

Click to expand...

Did so just now, 230ms solid ping. First time I had to reboot a router for it to help when the ISP had issues. I miss my Netgear. This piece of **** Dlink sucks

And now it sucks again. Even the latencies are high now:
C:\Users\Werner>pathping /q 100 /p 50 bbc.co.uk

Tracing route to bbc.co.uk [212.58.241.131]
over a maximum of 30 hops:
0 Werner-PC [192.168.1.107]
1 192.168.1.1
2 41-133-128-1.dsl.mweb.co.za [41.133.128.1]
3 tengig-0-0-0-110.vic-ipc-1.mweb.co.za [196.22.169.170]
4 vl-92.vic-hscore-1.mweb.co.za [196.22.189.2]
5 tengig0-0-0-2.11.vic-p-1.mweb.co.za [196.22.169.225]
6 pos0-3-2-0.lon-up-1.mweb.co.za [197.80.5.7]
7 176.67.177.131
8 xe-8-1-0.edge3.London1.Level3.net [195.50.124.133]
9 vl-3604-ve-228.csw2.London1.Level3.net [4.69.166.157]
10 * ae-2-52.edge4.London1.Level3.net [4.69.139.106]
11 BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]
12 * * *
Computing statistics for 55 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 Werner-PC [192.168.1.107]
0/ 100 = 0% |
1 0ms 1/ 100 = 1% 1/ 100 = 1% 192.168.1.1
0/ 100 = 0% |
2 342ms 0/ 100 = 0% 0/ 100 = 0% 41-133-128-1.dsl.mweb.co.za [41.133.128.1]
0/ 100 = 0% |
3 341ms 0/ 100 = 0% 0/ 100 = 0% tengig-0-0-0-110.vic-ipc-1.mweb.co.za [196.22.169.170]
0/ 100 = 0% |
4 322ms 2/ 100 = 2% 2/ 100 = 2% vl-92.vic-hscore-1.mweb.co.za [196.22.189.2]
0/ 100 = 0% |
5 335ms 1/ 100 = 1% 1/ 100 = 1% tengig0-0-0-2.11.vic-p-1.mweb.co.za [196.22.169.225]
0/ 100 = 0% |
6 514ms 1/ 100 = 1% 1/ 100 = 1% pos0-3-2-0.lon-up-1.mweb.co.za [197.80.5.7]
0/ 100 = 0% |
7 528ms 1/ 100 = 1% 1/ 100 = 1% 176.67.177.131
0/ 100 = 0% |
8 523ms 0/ 100 = 0% 0/ 100 = 0% xe-8-1-0.edge3.London1.Level3.net [195.50.124.133]
0/ 100 = 0% |
9 518ms 1/ 100 = 1% 1/ 100 = 1% vl-3604-ve-228.csw2.London1.Level3.net [4.69.166.157]
0/ 100 = 0% |
10 544ms 1/ 100 = 1% 1/ 100 = 1% ae-2-52.edge4.London1.Level3.net [4.69.139.106]
0/ 100 = 0% |
11 554ms 0/ 100 = 0% 0/ 100 = 0% BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]

Trace complete.


I don't get it

@mwebguy sent you a pm thanks.

Ever since Telkom has upgraded the 384kbps users to 1Mbps our latencies (on the WSTB [Stellenbosch] exchange) has increased drastically!

Here is my proof (Zabbix graphs showing my WAN usage on the left axis and Latency to Google on the right axis - all for the time between 12:00 - 18:00):

02 Oct - Tuesday - all good (15ms avg):
*

03 Oct - Wednesday - problems started (34ms avg):
*

05 Oct - Friday - bad (47ms avg):
*

07 Oct - Sunday - bad (61ms avg):
*

08 Oct - Monday - pretty bad (66ms avg):
*

10 Oct - Wednesday - bad (53ms avg):
*

12 Oct - Friday - bad (62ms avg):
*

14 Oct - Sunday - all good (15ms avg):
*

* Click thumbnails for full size images.

For trace routes, see what I've posted on the 9th of Oct in the List of congested exchanges? thread. My results were the same with my WebAfrica account, where the 2nd-hop would have a latency of 70ms+, instead of just like 10ms.

As you can probably see from the graphs, the issue is mostly during the week.
Unfortunately I can't really post latency graphs for after hours, because then I'm mostly downloading stuff, which makes the latency go up even more than due to the congested exchange.

I also just posted a select few images, because posts are limited to 8 images. I only started monitoring my connection from the 28th of Sept, and then added left axis data on the 6th of Oct.
My latency axis should be pretty accurate, because it pings the Google host 3x during a 5s period.

@MWEB Guy:
Should I log this as a fault with MWEB, seeing that I'm renting my line & uncapped account with MWEB? Or should I also log a fault with Telkom and get all my friends who are experiencing the same to do the same?

Torrents moving on at 0.6kb/s everything else is fine . Interesting when you stop/start torrent is going on 170kb/s then drops.

Must thank MWEB today, epic local latencies noticed.. Durban IPC doing good.
Downloads are also performing well for me (I only use NNTP)

Code-Red said:

Same thing as last night and at a similar time too, getting high ping to my local server (Cape Town: http://www.pingtest.net/result/70800503.png - I usually get around 20ms I think) and unplayable numbers to EUW (Frankfurt: http://www.pingtest.net/result/70800555.png - I usually get ~170ms). It went back to normal at about 2am last night, am I being unusually shaped or throttled? Wouldn't really make sense to me considering I'm not a heavy user.

Click to expand...

Hi Code-Red

Are you experiencing these ping test while gaming mainly? How is your normal browsing?

Pada said:

Ever since Telkom has upgraded the 384kbps users to 1Mbps our latencies (on the WSTB [Stellenbosch] exchange) has increased drastically!

Here is my proof (Zabbix graphs showing my WAN usage on the left axis and Latency to Google on the right axis - all for the time between 12:00 - 18:00):

02 Oct - Tuesday - all good (15ms avg):
*

03 Oct - Wednesday - problems started (34ms avg):
*

05 Oct - Friday - bad (47ms avg):
*

07 Oct - Sunday - bad (61ms avg):
*

08 Oct - Monday - pretty bad (66ms avg):
*

10 Oct - Wednesday - bad (53ms avg):
*

12 Oct - Friday - bad (62ms avg):
*

14 Oct - Sunday - all good (15ms avg):
*

* Click thumbnails for full size images.

For trace routes, see what I've posted on the 9th of Oct in the List of congested exchanges? thread. My results were the same with my WebAfrica account, where the 2nd-hop would have a latency of 70ms+, instead of just like 10ms.

As you can probably see from the graphs, the issue is mostly during the week.
Unfortunately I can't really post latency graphs for after hours, because then I'm mostly downloading stuff, which makes the latency go up even more than due to the congested exchange.

I also just posted a select few images, because posts are limited to 8 images. I only started monitoring my connection from the 28th of Sept, and then added left axis data on the 6th of Oct.
My latency axis should be pretty accurate, because it pings the Google host 3x during a 5s period.

@MWEB Guy:
Should I log this as a fault with MWEB, seeing that I'm renting my line & uncapped account with MWEB? Or should I also log a fault with Telkom and get all my friends who are experiencing the same to do the same?

Click to expand...

Hi Pada

Thank you for the information.

This is happening with your Web Africa account as well then it could be exchange or line related. You mentioned your line is with us, please PM me the MWEB email address or customer number so that I can log it with Telkom.

Also, if your friends are experiencing the same problem, I would suggest they log it with Telkom as well.

arni1954 said:

Torrents moving on at 0.6kb/s everything else is fine . Interesting when you stop/start torrent is going on 170kb/s then drops.

Click to expand...

Hi arni1954
Is this the current speed you getting? Does your speed increase during the night at all?

Still terrible on my MWEB account this morning. Here is a pathping with a Web Africa account:

C:\Users\Werner>pathping /q 100 /p 50 bbc.co.uk

Tracing route to bbc.co.uk [212.58.241.131]
over a maximum of 30 hops:
0 Werner-PC [192.168.1.107]
1 192.168.1.1
2 obfn-ip-bb-1.ipnet.wa.co.za [41.185.96.1]
3 196.220.59.222
4 wnls-cr1-vl-801.wa.co.za [41.185.0.67]
5 41-66-150-41-29.WC-RO-DCE-1-to-WEB001-CPE.africainx.net [41.66.150.41]
6 CORE_GB-LD-GS-MEC-1_TO_GB-LD-GS-MSC-1_AINX_P2P__1G_gb [41.84.12.145]
7 195.50.122.217
8 BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]
9 * * *
Computing statistics for 40 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 Werner-PC [192.168.1.107]
0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.1.1
0/ 100 = 0% |
2 51ms 0/ 100 = 0% 0/ 100 = 0% obfn-ip-bb-1.ipnet.wa.co.za [41.185.96.1]
0/ 100 = 0% |
3 52ms 0/ 100 = 0% 0/ 100 = 0% 196.220.59.222
0/ 100 = 0% |
4 51ms 0/ 100 = 0% 0/ 100 = 0% wnls-cr1-vl-801.wa.co.za [41.185.0.67]
0/ 100 = 0% |
5 51ms 0/ 100 = 0% 0/ 100 = 0% 41-66-150-41-29.WC-RO-DCE-1-to-WEB001-CPE.africainx.net [41.66.150.41]
0/ 100 = 0% |
6 --- 100/ 100 =100% 100/ 100 =100% CORE_GB-LD-GS-MEC-1_TO_GB-LD-GS-MSC-1_AINX_P2P__1G_gb [41.84.12.145]
0/ 100 = 0% |
7 201ms 0/ 100 = 0% 0/ 100 = 0% 195.50.122.217
0/ 100 = 0% |
8 200ms 0/ 100 = 0% 0/ 100 = 0% BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]

Trace complete.


Here is the MWEB one:

C:\Users\Werner>pathping /q 100 /p 50 bbc.co.uk

Tracing route to bbc.co.uk [212.58.241.131]
over a maximum of 30 hops:
0 Werner-PC [192.168.1.107]
1 192.168.1.1
2 41-133-128-1.dsl.mweb.co.za [41.133.128.1]
3 tengig-0-0-0-107.vic-ipc-1.mweb.co.za [196.22.169.158]
4 vl-92.vic-hscore-1.mweb.co.za [196.22.189.2]
5 tengig0-0-0-2.11.vic-p-1.mweb.co.za [196.22.169.225]
6 pos0-3-2-0.lon-up-1.mweb.co.za [197.80.5.7]
7 176.67.177.131
8 xe-8-1-0.edge3.London1.Level3.net [195.50.124.133]
9 vl-3604-ve-228.csw2.London1.Level3.net [4.69.166.157]
10 * ae-2-52.edge4.London1.Level3.net [4.69.139.106]
11 BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]
12 * * *
Computing statistics for 55 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 Werner-PC [192.168.1.107]
0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.1.1
0/ 100 = 0% |
2 30ms 0/ 100 = 0% 0/ 100 = 0% 41-133-128-1.dsl.mweb.co.za [41.133.128.1]
0/ 100 = 0% |
3 31ms 3/ 100 = 3% 3/ 100 = 3% tengig-0-0-0-107.vic-ipc-1.mweb.co.za [196.22.169.158]
0/ 100 = 0% |
4 31ms 4/ 100 = 4% 4/ 100 = 4% vl-92.vic-hscore-1.mweb.co.za [196.22.189.2]
0/ 100 = 0% |
5 34ms 4/ 100 = 4% 4/ 100 = 4% tengig0-0-0-2.11.vic-p-1.mweb.co.za [196.22.169.225]
0/ 100 = 0% |
6 211ms 1/ 100 = 1% 1/ 100 = 1% pos0-3-2-0.lon-up-1.mweb.co.za [197.80.5.7]
0/ 100 = 0% |
7 213ms 1/ 100 = 1% 1/ 100 = 1% 176.67.177.131
0/ 100 = 0% |
8 214ms 1/ 100 = 1% 1/ 100 = 1% xe-8-1-0.edge3.London1.Level3.net [195.50.124.133]
0/ 100 = 0% |
9 217ms 3/ 100 = 3% 3/ 100 = 3% vl-3604-ve-228.csw2.London1.Level3.net [4.69.166.157]
0/ 100 = 0% |
10 212ms 0/ 100 = 0% 0/ 100 = 0% ae-2-52.edge4.London1.Level3.net [4.69.139.106]
2/ 100 = 2% |
11 212ms 2/ 100 = 2% 0/ 100 = 0% BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]

Trace complete.


This is getting really frustrating

wikk3d88 said:

Im on 1mb uncapped, paying an additional rental fee because telkom upgraded the line to 2mb, and im getting what i guess a 512kb line feels like. Not quite as shyt as 384, but thats not saying much now is it.
Seriously considering changing ISPs if this service continues

Click to expand...

I'm on the old 1meg paying R488pm on tha AIC Package, Telkom only now recently upgraded my line to 1meg.
While it was syncing at 1meg on the router and at mweb, but had a 512 Telkom Connection.
Thought Mweb was suppose to sort this mess out with Telkom since i"m paying them for the All Inclusive Pakage.
Speeds are also Horrible for a 1meg product and was considering upgrading to 2meg, but considering this unstable service of Mweb Lately i don't think so...
ADSL Stands for Always on and Available Internet and Uncapped @ the Speed i am paying for..
Not Depending on Conditions and when things change Day2Day... What's the use of ADSL then ?
Might as well go 3G and have internet everywhere...

Hi Soulcode

Please can I ask you to send me a traceroute to the same site with your MWEB account as well as the Web Africa account. You can PM me the results.

MWEB Guy said:

Hi Soulcode

Please can I ask you to send me a traceroute to the same site with your MWEB account as well as the Web Africa account. You can PM me the results.

Click to expand...

Sent

Hi

Also been noticing some bad performance for the past 2 days. Was struggling to download Easyfile from SARS earlier, and knowing its usually a problem their side, I downloaded the file to our server at Hetzner (Germany) and then tried to download it from there. Same thing, download varies between 4 and 30 KB/s and then eventually times out. Have so far had 4 failed download attempts, and busy with #5 now, crawling at 30KB/s. Our transfers from the Germany server usually run at full speed (1mbps uncapped mweb). I can deal with the slower speed, but its the timing out thats bothering us right now...

Ok, while writing all the above, I tried doing an FTP download instead, and its coming down at between 80 & 90 KB/s.... and completed! So could it be your HTTP cache servers causing the http download speeds/timeouts?

Yeah getting really erratic speeds on everything atm. On my Afrihost it is fine

racer said:

Hi

Also been noticing some bad performance for the past 2 days. Was struggling to download Easyfile from SARS earlier, and knowing its usually a problem their side, I downloaded the file to our server at Hetzner (Germany) and then tried to download it from there. Same thing, download varies between 4 and 30 KB/s and then eventually times out. Have so far had 4 failed download attempts, and busy with #5 now, crawling at 30KB/s. Our transfers from the Germany server usually run at full speed (1mbps uncapped mweb). I can deal with the slower speed, but its the timing out thats bothering us right now...

Ok, while writing all the above, I tried doing an FTP download instead, and its coming down at between 80 & 90 KB/s.... and completed! So could it be your HTTP cache servers causing the http download speeds/timeouts?

Click to expand...

Good Evening racer, thanks for the on how your http performs through the course of this evening.

dasco144 said:

Yeah getting really erratic speeds on everything atm. On my Afrihost it is fine

Click to expand...

Evening Dasco, what type of speeds are you obtaining?

Hi Guys

I've noticed particularly erratic speeds ever since upgrading to a 2mb line with Telkom and then later upgrading with MWEB(my ISP).

I just wanted to find out if there is any network issues at the moment from MWEB's side?

With my 1meg account i usually had full speeds at all times.

I'm on a 2mb uncapped account with MWEB , my line is with Telkom.

Any info or advice will be greatly appreciated If you need more info just let me know.

My speeds are down as well.. I also should be downloading over 400kb by now, at the moment I am getting 275kb..

Being like this for a week now... are there network problems????