Afrihost Capped ADSL Feedback (MTN) - New beginning

Status
Not open for further replies.
1254547921.png

Did you bring out the 33.6 modem?
 
Morning everyone. Things seemed generally better on the network last night, however we did see demand spike a little around 10pm in the South and latency rose for about 20-30mins while we got that traffic under control. All networks seems mostly normal otherwise. We are still seeing high demand during those peak periods but we seem to be seeing much better latency for gaming and realtime services.

I see some people in the south did report higher 3rd hop latency between 10 and 10:30 which seems to match my reports, but that should only be in the South. Other regions should have seen consistent latency.
 
So it's official, we can now start comparing Afrihost to Eskom, as they appear to be running the exact same business model, that being force clients to use less in order to maintain their fragile infrastructure :p

No need to reply to this post, just my once off Friday rant :)
 
Morning everyone. Things seemed generally better on the network last night, however we did see demand spike a little around 10pm in the South and latency rose for about 20-30mins while we got that traffic under control. All networks seems mostly normal otherwise. We are still seeing high demand during those peak periods but we seem to be seeing much better latency for gaming and realtime services.

I see some people in the south did report higher 3rd hop latency between 10 and 10:30 which seems to match my reports, but that should only be in the South. Other regions should have seen consistent latency.

I'm not in the south and have been reporting high 3rd hop latency for 3 days now.

How was the network better last night? Did you guys not read the posts here?
 
Last edited:
I'm not in the south and have been reporting high 3rd hop latency for 3 days now.

How was the network better last night? Did you guys not read the posts here?

I have to report based on the overall pciture. I definitely read the posts, which is why I wanted to clarify that what was reported in the South is consistent, but the North overall latency was normal according to our own tests, monitoring and reports from our channels.
 
Morning everyone. Things seemed generally better on the network last night, however we did see demand spike a little around 10pm in the South and latency rose for about 20-30mins while we got that traffic under control. All networks seems mostly normal otherwise. We are still seeing high demand during those peak periods but we seem to be seeing much better latency for gaming and realtime services.

I see some people in the south did report higher 3rd hop latency between 10 and 10:30 which seems to match my reports, but that should only be in the South. Other regions should have seen consistent latency.
Must have missed my speedtests after midnight then...
 
So it's official, we can now start comparing Afrihost to Eskom, as they appear to be running the exact same business model, that being force clients to use less in order to maintain their fragile infrastructure :p

No need to reply to this post, just my once off Friday rant :)

I'd like reply, hope that's OK :)

I do understand why you would think that, but I think it's a different situation. Not to criticise Eskom, but they have a very different set of circumstances and I think there's differing information in the press and that's a huge part of their difficulties.

The difficulty we face is challenging, but we're seeing progress and I believe we'll continue to see improvement. I definitely don't think we're going backwards, but forwards, and we have a good idea of what the problems are and how we can tackle them. Ideally, we'd love to do it faster, but unfortunately a lot of the solutions we implement then have to be monitored. Because the network is live, there is always the danger of implementing too many concurrent changes which might be a mess to untangle if any one change creates a bigger problem. Again, not to criticise Eskom, but we definitely have solutions on the horizon and the experience will improve. We don't have a monopoly and we offer month to month service, so we have to get this right.
 
Last edited:
So when are we getting a straight answer regarding us folk in the South? Its DAMN frustrating to have a cap you can barely use only for it to expire come the 1st...?
 
So when are we getting a straight answer regarding us folk in the South? Its DAMN frustrating to have a cap you can barely use only for it to expire come the 1st...?

I've done my best to answer, the short answer is that there is high demand on the network at peak times and we're seeing degraded performance.

The long answer is that we believe this demand is inflated due to rogue P2P traffic that should be shaped (for Uncapped users) and we are working on closing those loopholes in our shaping policies. We have identified some new P2P traffic signatures and improved identification of existing signatures to address this. We are still working on this continuously until we've improved the impact of demand on the overall network performance.
 
I've done my best to answer, the short answer is that there is high demand on the network at peak times and we're seeing degraded performance.

The long answer is that we believe this demand is inflated due to rogue P2P traffic that should be shaped (for Uncapped users) and we are working on closing those loopholes in our shaping policies. We have identified some new P2P traffic signatures and improved identification of existing signatures to address this. We are still working on this continuously until we've improved the impact of demand on the overall network performance.

Clearly those shaping policy's that are being implemented are having an effect on the capped users accounts also.
 
Very intermittent, strange that the 3rd hop times out as well on the second trace. With network latency I usually expect it to be consistently bad, but this seems a lot more intermittent.

Have you tested on another ISP network?
I get the same on Axxess. They use MTN as well.

My Speedtest is normal at present, but is often higher than any other ISP. Gaming also is a major problem.
 
Clearly those shaping policy's that are being implemented are having an effect on the capped users accounts also.

The shaping policy's are not the cause, the issue is the demand. The shaping policies are there to help and address the demand and we're making incremental progress but it's not a turnkey solution. We have to take the wins as they come but I do believe at the end of the process we'll be back to a great internet experience for our clients.
 
Honestly, if you are seeing users using more than what they are supposed to use, block their accounts by willingly bypassing the FUP/AUP. That has been what MWEB has been doing for years.

However, that you aren't and are letting them degrade the network for everyone is really puzzling.
 
The shaping policy's are not the cause, the issue is the demand. The shaping policies are there to help and address the demand and we're making incremental progress but it's not a turnkey solution. We have to take the wins as they come but I do believe at the end of the process we'll be back to a great internet experience for our clients.

When is this process due to end? Why did this situation come about all of a sudden?
 
So I tried gaming on my AH account for the first time in about 2 weeks and it was awful. Literally nothing has changed in 2 weeks. 30% loss in mumble and utterly horrid performance in GTA Online. Not even worth playing.

Switch to telkom and my loss doesn't even reach 0.05%. Glad to see nothing has changed...
 
Status
Not open for further replies.
Top
Sign up to the MyBroadband newsletter
X