Telkom Internet Capped/Uncapped User Feedback (Pt2)

Things are still pear shaped - Hulu still battling.
 
Poor night on the DL front last night.

Almost time to give up...

How much longer, ranger?

Strange, my downloads seem to be relatively OK during Night Surfer time - usenet, cannot comment on torrents.

Daytime streaming and general browsing is beyond painful and severely frustrating.
100MB Android SW update took about 3.5 hours yesterday...

Telkom Internet was certainly not contributing to the spirit of 'reconciliation' on Reconciliation Day...
 
Something just got better.

International pretty much back to normal standard for me.
Normal ping and data transfer.

Hallelujah, 3 days of torture over.
 
Something just got better.

International pretty much back to normal standard for me.
Normal ping and data transfer.

Hallelujah, 3 days of torture over.
Looks like something was done :)

Code:
[mick@mick1 ~]$ mtr -rc 20 bbc.co.uk
Start: Wed Dec 17 11:59:41 2014
HOST: mick1.mickza                Loss%   Snt   Last   Avg  Best  Wrst StDev
  1.|-- newcop                     0.0%    20    0.3   0.2   0.1   0.3   0.0
  2.|-- ???                       100.0    20    0.0   0.0   0.0   0.0   0.0
  3.|-- wblv-ip-hsll-2-gi-3-0-0-v  0.0%    20  118.3  51.6  15.8 180.2  55.5
  4.|-- ipc-3-up.south.dsl.telkom  0.0%    20   16.7  16.1  15.6  17.0   0.0
  5.|-- ipc-aggr-1.south.dsl.telk  0.0%    20   25.9  16.7  15.6  25.9   2.2
  6.|-- wbs-ip-ll-1-wan.telkom-ip  0.0%    20   16.9  16.6  15.8  19.4   0.6
  7.|-- lon-ip-hsll-1-gig-0-1-0.t  0.0%    20  160.7 160.3 159.7 160.8   0.0
  8.|-- xe-0-1-0-19.r00.londen10.  0.0%    20  161.6 160.9 160.3 161.7   0.0
  9.|-- ae-7.r23.londen03.uk.bb.g  0.0%    20  160.8 160.9 160.0 164.9   1.1
 10.|-- bbc-linx.pr01.rbsov.bbc.c  0.0%    20  161.5 164.9 160.1 184.5   5.8
 11.|-- ???                       100.0    20    0.0   0.0   0.0   0.0   0.0
 12.|-- ???                       100.0    20    0.0   0.0   0.0   0.0   0.0
 13.|-- ae0.er01.telhc.bbc.co.uk   0.0%    20  163.5 162.3 161.1 163.7   0.5
 14.|-- 132.185.255.149            0.0%    20  171.3 171.1 170.6 171.8   0.0
 15.|-- fmt-vip71.telhc.bbc.co.uk  0.0%    20  170.7 170.8 170.2 174.0   0.7
[mick@mick1 ~]$

(ignore hop 3 I've got iPlayer and downloads running on a couple of systems).
 
My international is buggered again... usual story, local fine, pings fine, tracert fine etc, just slow, and to be honest, getting old now. Been a few months of this ****, works fine one day, buggered the next, wash, rinse, repeat... I've never had problems with telkom, but for the last 6mnths or so, its been awful. I understand **** happens, but this has gone on long enough

http://speedtest.ookla.com/result/2784552384.png
http://speedtest.ookla.com/result/2784553604.png

http://testmy.net/QSH6esR.Ort1cXQ.png

Traceroute has started…

traceroute: Warning: bbc.co.uk has multiple addresses; using 212.58.244.18
traceroute to bbc.co.uk (212.58.244.18), 64 hops max, 72 byte packets
1 10.0.0.2 (10.0.0.2) 3.598 ms 0.895 ms 0.708 ms
2 rrba-ip-bng-1.north.dsl.telkomsa.net (105.224.208.1) 10.460 ms 9.835 ms 10.870 ms
3 ipc-1-up.north.dsl.telkomsa.net (105.224.0.6) 13.141 ms 14.221 ms 11.877 ms
4 ipc-aggr-1.north.telkomsa.net (105.224.0.13) 13.300 ms 12.829 ms 13.237 ms
5 165.165.216.221 (165.165.216.221) 12.959 ms 12.919 ms 13.405 ms
6 lon-ip-hsll-1-gig-7-1-0.telkom-ipnet.co.za (196.43.9.58) 189.911 ms 188.853 ms 189.761 ms
7 xe-0-1-0-19.r00.londen10.uk.bb.gin.ntt.net (83.231.146.245) 189.935 ms 187.676 ms 188.213 ms
8 ae-7.r23.londen03.uk.bb.gin.ntt.net (129.250.6.54) 195.517 ms 194.641 ms 187.807 ms
9 bbc-linx.pr01.rbsov.bbc.co.uk (195.66.236.103) 189.491 ms 189.575 ms 189.493 ms
10 * * *
11 * * *
12 ae0.er01.telhc.bbc.co.uk (132.185.254.109) 193.222 ms 191.398 ms 190.542 ms
13 132.185.255.149 (132.185.255.149) 197.681 ms 197.577 ms 199.209 ms
14 fmt-vip72.telhc.bbc.co.uk (212.58.244.18) 188.769 ms 188.447 ms 188.313 ms
 
Last edited:
My international is buggered again... usual story, local fine, pings fine, tracert fine etc, just slow, and to be honest, getting old now.

This is still the IPC issue. Shaping effects are magnified by latency, so intl is affected more.

Cape Town's IPC upgrade is ready to go, but had to wait for a change we just completed in Gauteng.

So, this should hopefully be fixed tomorrow (today?).

I checked the date we ordered this additional transmission to cater to more IPC demand: 25 June .... (but we ordered upgrades using the last transmission capacity in Sept.).
 
I checked the date we ordered this additional transmission to cater to more IPC demand: 25 June .... (but we ordered upgrades using the last transmission capacity in Sept.).

Thanks ranger for the update.
That said, when looking back on this thread, it is clear that this IPC issue seems to be recurring every other month...
Is there no way of better anticipating the demand and planning accordingly? Hope that the answer will not turn the question into the usual rhetorical one...

I find it real difficult to accept that my circa R800 per month over the last year could only buy me about 6 month's worth of fair Internet access...
 
This is still the IPC issue. Shaping effects are magnified by latency, so intl is affected more.

Cape Town's IPC upgrade is ready to go, but had to wait for a change we just completed in Gauteng.

So, this should hopefully be fixed tomorrow (today?).

I checked the date we ordered this additional transmission to cater to more IPC demand: 25 June .... (but we ordered upgrades using the last transmission capacity in Sept.).

Thanks for the update, much appreciated
 
This is still the IPC issue. Shaping effects are magnified by latency, so intl is affected more.

Cape Town's IPC upgrade is ready to go, but had to wait for a change we just completed in Gauteng.

So, this should hopefully be fixed tomorrow (today?).

I checked the date we ordered this additional transmission to cater to more IPC demand: 25 June .... (but we ordered upgrades using the last transmission capacity in Sept.).

It's time that the TI execs get some balls and explain to your group CEO how Wholesale are destroying TI's value.

It also exposes the CTO as he made it very clear in September that they had no outstanding IPC orders on hand.
 
It's time that the TI execs get some balls and explain to your group CEO how Wholesale are destroying TI's value.

It also exposes the CTO as he made it very clear in September that they had no outstanding IPC orders on hand.

Someone also said there's no congestion but we know there is dslam backhaul issues.

Also does the problem lie solely with TW or is it also an issue with transmission not having capacity? Does TI monitor utilisation and do forecasting and forward this on to TW & Transmission in a timeous manner in order for planning to be done and capacity to be provisioned?

These TI issues take months to resolve which is unacceptable, they (TI, TW, Transmission etc) need to get their house in order.
 
It's time that the TI execs get some balls and explain to your group CEO how Wholesale are destroying TI's value.

It also exposes the CTO as he made it very clear in September that they had no outstanding IPC orders on hand.

Well it's already destroyed for me (TI's value, that is). Cancellation effective 10 January.

It's been very interesting swapping between my TI accounts (capped and uncapped) and Afrihost capped. Afrihost just works. Page loads, lookups, downloads, nntp - hell, even torrents in the middle of the day @ 1 MBps.

Perhaps it's because (I'm guessing) I break out straight on to MTN Business fibre at my exchange?

Anyway, what does a person do with an unused TI 4 megabit uncapped account for 3 weeks?
 
Thanks ranger for the update.
That said, when looking back on this thread, it is clear that this IPC issue seems to be recurring every other month...
Is there no way of better anticipating the demand and planning accordingly? Hope that the answer will not turn the question into the usual rhetorical one...

As I have explained before, on previous occasions we were able to get more transmission from request for proposal to IPC circuit working in 6 weeks.

We requested upgrades to the Cape Town IPC to use > 80% of the transmission in June, and as soon as it was implemented (on 25 June) requested proposals for more transmission capacity. We introduced about 2 business days of delay getting signatures, and this transmission was completed on 21 Nov., but the creation of the IPC circuits (which was on the dame order) was completed last Thursday. Only after the creation of the IPC (logical) circuits themselves could we request upgrades. In the meantime, we used the last transmission for IPC upgrades at the end of August.

So, based on previous experience we had given enough lead time, but on this occasion it took 4 months more.

The IPC upgrades are now complete, and we are currently using about 60% of the transmission available in the Cape Town site.

We have also been finalising other capacity planning and forecasting models to give a longer planning horizon. So, thos shouldn't happen again.
 
Top
Sign up to the MyBroadband newsletter
X