Afrihost Business Uncapped Feedback - 2

[AfriGenie]

so if I sign up with SAIX and buy 1Mbps capacity, Telkom are going to go and deploy millions of Rands worth of routers across the country for me?

I believe that would be included in the IPC costs. At the moment that major consumers of IPC would be the IS and MTN infrastructures. Since most ISPs use one of those two, network traffic would be routed through the appropriate DSLAM, BRAS and then IPC infrastructure.

 

[kimbo]

It looks like you are connected to a completely different Exchange. I've reset your ADSL line from my side, can you check if there is any change in your connection.

Still the same. I can barely brows, barely do anything. Nothing is working

Tracing route to www.afrihost.com [197.242.144.102]
over a maximum of 30 hops:

1 1 ms <1 ms <1 ms 192.168.1.1
2 38 ms 38 ms 39 ms 105-236-8-1-esr-lo.mtnbusiness.co.za [105.236.8.
1]
3 47 ms 47 ms 47 ms ipc-recieve-jh-2a.za.mtnbusiness.net [41.181.178
.17]
4 47 ms 47 ms 49 ms qux-jh-dca-2.za-b.za.mtnbusiness.net [41.181.165
.115]
5 48 ms 47 ms 53 ms jh-dca-2.za--qux-b.za.mtnbusiness.net [41.181.16
5.114]
6 50 ms 49 ms 73 ms jh-cr-2.za--jh-dca-2.za-a.mtnns.net [196.44.0.22
4]
7 48 ms 48 ms 48 ms 196.44.31.99
8 49 ms 48 ms 48 ms 196.31.220.8
9 48 ms 48 ms 48 ms 196.31.220.27
10 47 ms 48 ms 47 ms 196.31.63.202
11 47 ms 48 ms 58 ms vl9-ah-ha-2.za.mtnbusiness.net [196.30.42.133]
12 48 ms 48 ms 48 ms cms-gm.afrihost.com [197.242.144.102]

Trace complete.

 

[AfriGenie]

Whooo Hoooo Yeah


Tracing route to www.afrihost.co.za [197.242.144.102]
over a maximum of 30 hops:

1 3 ms 1 ms <1 ms 192.168.1.1
2 8 ms 8 ms 8 ms 105-236-1-1-esr-lo.mtnbusiness.co.za [105.236.1.
1]
3 17 ms 17 ms 17 ms 41.181.221.214
4 17 ms 18 ms 17 ms rb-dca-2.za--qux-b.za.mtnbusiness.net [41.181.16
5.118]
5 18 ms 18 ms 18 ms rb-cr-1.za--rb-dca-2.mtnns.net [196.44.31.122]
6 18 ms 17 ms 18 ms 196.44.31.97
7 19 ms 18 ms 18 ms 196.31.220.10
8 20 ms 18 ms 18 ms 196.31.220.21
9 20 ms 17 ms 17 ms 196.31.63.202
10 18 ms 19 ms 18 ms vl9-ah-ha-2.za.mtnbusiness.net [196.30.42.133]
11 18 ms 18 ms 18 ms cms-gm.afrihost.com [197.242.144.102]

Trace complete.

Yay!

Let's hope you stay at those levels!

 

[cortexenslaver]

With which ISPs did you test specifically? I want to see if I can correlate this from my side as well.

Afrihost (1GB Free, Capped, Bus. Uncapped)
Axxess (Free)
Web Africa (Capped, Home Uncapped)
Vox Fat Pipe
MWEB (Capped, Uncapped)
FNB (Capped test account)


The above are all over the place, and have been tested from 3 locations (Milnerton, Sunningdale, Montague Gardens)

Telkom (Capped, Uncapped) - Both operate as normal, but there was an instance today at 12:30 exactly that the line in Montague Gardens dropped and then struggled to authenticate for about 10 minutes... then it was back to normal.

 

[Rickster]

No, imagine your exchange is a highway.

If there is a huge amount of traffic, will one lane (protocol) be faster than another?

No.

Wait, I might be wrong here.

When my exchange was congested I would get good download speed but the packet loss an ping were horrible.

If I remember correctly.

 

[ghostR]

I believe that would be included in the IPC costs. At the moment that major consumers of IPC would be the IS and MTN infrastructures. Since most ISPs use one of those two, network traffic would be routed through the appropriate DSLAM, BRAS and then IPC infrastructure.

So no matter whether my line is migrated to you or not, exchange congestion can and is now dependent on the ISP, because apparently MTN now have their own network infrastructure at every exchange. So if a customer, no matter whether their line is with you or not, has signs of congestion, they must report it to Afrihost for Afrihost to fix. Is this correct?

 

[AfriGuy]

So no matter whether my line is migrated to you or not, exchange congestion can and is now dependent on the ISP, because apparently MTN now have their own network infrastructure at every exchange. So if a customer, no matter whether their line is with you or not, has signs of congestion, they must report it to Afrihost for Afrihost to fix. Is this correct?

Congestion takes place before the BRAS. After the BRAS traffic breaks out to the relevant IPC links, anything before that is owned, run and maintained by Telkom This is why everyone is fighting so hard for LLU.
Not sure what reporting metrics Telkom use to monitor congestion.

 

[Quad!]

My Internet was working perfectly.. Then all of a sudden dead.. Tested everything and looks like authentication problem.. Problem sins 13:10 2015-04-13

Who is the monkey that played with the wires at the exchange?

Still no connection!!!

 

[ghostR]

Congestion takes place before the BRAS. After the BRAS traffic breaks out to the relevant IPC links, anything before that is owned, run and maintained by Telkom This is why everyone is fighting so hard for LLU.
Not sure what reporting metrics Telkom use to monitor congestion.

So you're very quick to accuse Telkom of running congested exchanges based on a customer's tracert, so how are you identifying this based on a tracert, and ruling out your equipment at the exchanges before doing so? You don't have access to this router nor its reports (if this dedicated router exists), yet we have now ascertained that you can diagnose this fault and get it fixed if need be. So you have an obligation to your customers now to rule this out every time someone posts high latency on the second hop, as that latency may actually be caused by your routers and not congestion at all. And since you claim that the congestion happens before the BRAS, and we know your router doesn't respond to ICMP requests, it seems pretty damn irresponsible to claim congestion based on high second hop latency.
Or, you have your wires crossed and this is actually not the case at all

 

[electrofreeze]

So you're very quick to accuse Telkom of running congested exchanges based on a customer's tracert, so how are you identifying this based on a tracert, and ruling out your equipment at the exchanges before doing so? You don't have access to this router nor its reports (if this dedicated router exists), yet we have now ascertained that you can diagnose this fault and get it fixed if need be. So you have an obligation to your customers now to rule this out every time someone posts high latency on the second hop, as that latency may actually be caused by your routers and not congestion at all. And since you claim that the congestion happens before the BRAS, and we know your router doesn't respond to ICMP requests, it seems pretty damn irresponsible to claim congestion based on high second hop latency.
Or, you have your wires crossed and this is actually not the case at all

+100000

come on ah stop blaming other people and pull your finger out you !!!!!!!!!!

 

[Rickster]

So you're very quick to accuse Telkom of running congested exchanges based on a customer's tracert, so how are you identifying this based on a tracert, and ruling out your equipment at the exchanges before doing so? You don't have access to this router nor its reports (if this dedicated router exists), yet we have now ascertained that you can diagnose this fault and get it fixed if need be. So you have an obligation to your customers now to rule this out every time someone posts high latency on the second hop, as that latency may actually be caused by your routers and not congestion at all. And since you claim that the congestion happens before the BRAS, and we know your router doesn't respond to ICMP requests, it seems pretty damn irresponsible to claim congestion based on high second hop latency.
Or, you have your wires crossed and this is actually not the case at all

Actually he said

Congestion takes place before the BRAS, anything before that is owned, run and maintained by Telkom

So every piece of hardware in the exchange is owned and accessed only by telkom.

But you claiming AH have some access to the exchanges which is not true.

 

[ghostR]

Actually he said



So every piece of hardware in the exchange is owned and accessed only by telkom.

But you claiming AH have some access to the exchanges which is not true.

And this?

Each backbone provider - IS / MTN - would have their own separate router or network infrastructure at the Exchange. That's why if you were to use an Axxess account you would probably see similar results.

 

[Rickster]

And this?

Yes they do have their own routers, but they are maintained and owned by Telkom.

 

[AfriGuy]

And this?

This all takes place after the BRAS, as I've said in my initial comment about this.
Short answer, is no, we don't have access to the exchange directly. Traffic breaks out to IPC links after the BRAS, this would display as the 3rd hop, generally.

So the exchange, being the second hop, is run by Telkom. We have zero access to this.

 

[ghostR]

Yes they do have their own routers, but they are maintained and owned by Telkom.

Might be the case, but it's dedicated hardware for MTN, in which case MTN have a responsibility to ensure that the hardware Telkom provision for them is working, not the client of MTN who must contact Telkom to talk to them about hardware for MTN

 

[ghostR]

This all takes place after the BRAS, as I've said in my initial comment about this.
Short answer, is no, we don't have access to the exchange directly. Traffic breaks out to IPC links after the BRAS, this would display as the 3rd hop, generally.

So the exchange, being the second hop, is run by Telkom. We have zero access to this.

Afripeople, on what hop are you now stating this exchange issue can manifest? We are talking about the one where your equipment operated by Telkom, but dedicated to MTN only, was the problem, as you claim earlier in the thread it was. I'd just like to know which hop you are stating it will show up as

Also, can you confirm that you have no capacity constraints still? Because it seems as if what we call capacity constraints, you like to call "increased demand" and I think we're all talking about the same thing. You are just polishing the turd a little more

 

[ghostR]

This all takes place after the BRAS, as I've said in my initial comment about this.
Short answer, is no, we don't have access to the exchange directly. Traffic breaks out to IPC links after the BRAS, this would display as the 3rd hop, generally.

So the exchange, being the second hop, is run by Telkom. We have zero access to this.

It also seems as if you're now trying to take increased demand/Afrihost capacity problems and start to blame Telkom for them by stating that there are faulty routers that may show up on your IPC hop. Either it shows up on hop 2, in which case you need to investigate every congestion symptom and rule out the MTN hardware at that exchange first, or it's hop 3, in which case you just invented a brand new way to pass the buck and claim that your capacity issues are actually just faulty Telkom hardware at exchanges around SA. Either way you guys are approaching Openweb levels of trust now

 

[AfriGuy]

Might be the case, but it's dedicated hardware for MTN, in which case MTN have a responsibility to ensure that the hardware Telkom provision for them is working, not the client of MTN who must contact Telkom to talk to them about hardware for MTN

As nice as that sounds, I'm afraid it's a bit more complex than that.

 

[AfriGuy]

Afripeople, on what hop are you now stating this exchange issue can manifest? We are talking about the one where your equipment operated by Telkom, but dedicated to MTN only, was the problem, as you claim earlier in the thread it was. I'd just like to know which hop you are stating it will show up as

Also, can you confirm that you have no capacity constraints still? Because it seems as if what we call capacity constraints, you like to call "increased demand" and I think we're all talking about the same thing. You are just polishing the turd a little more

I think exchange congestion has been so deeply discussed across MyBroadband. Regardless, congestion happens at the exchange.
So let's assume your exchange is congested, and you have a router (nothing special) and your Afrihost account.

Here's how the traceroute works:

Hop 1 is your router, sitting in your house. This device is managed by you.
Hop 2 is your exchange, owned, run and maintained by Telkom.
Hop 3 is breakout traffic to MTN's IPC links. From here on out, traffic is routed, run, owned and maintained by MTN.

Hope that clears things up.
So again, exchange congestion happens at the exchange, which is not run, owned, controlled or maintained by anyone other than Telkom.

 

[whatwhat]

So what about AfriGenie statement that said this?

"Each backbone provider - IS / MTN - would have their own separate router or network infrastructure at the Exchange."

Would that show under Hop 2 then?