Afrihost Business Uncapped Feedback - 2

Status
Not open for further replies.
HP_Exarkun said:
And another evening that there is no connection to any Usenet providers on Business.
Normal capped works, other ISP's Works but Afrihost Business does not.

And the reply from support? Please send us more traceroutes and speed test
Click to expand...

I've tested some NNTP stuff earlier and all seems good from my end.
What servers are you using?
 
Dianysis said:
Yeah there are line issues - since Feb. But seeing as though nothing is being done (no ETR from Telkom), I am looking at the situation from all angles. And yes, maybe grasping at straws, but better than doing nothing.
Thanks for looking at the traceroute.
Click to expand...

We'll push Telkom where we can.
 
C:\Users\Joshua>tracert www.afrihost.com

Tracing route to www.afrihost.com [41.181.185.18]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 140 ms 77 ms 61 ms 41.181.51.181
4 62 ms 60 ms 59 ms 41.181.51.182
5 83 ms 82 ms 82 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.198.188]
6 79 ms 80 ms 79 ms unc-cpt-1.mtnns.net [196.44.18.8]
7 103 ms 101 ms 101 ms pe-cr-2.za--el-mse-1.za.mtnns.net [196.44.31.66]
8 120 ms 119 ms 115 ms jh-dca-1.za--jh-cr-1.za-a.mtnns.net [196.44.0.217]
9 96 ms 106 ms 91 ms 18.185.181.41.in-addr.aserv.co.za [41.181.185.18]

Trace complete.

latency climbing again.
 
8-BALL said:
C:\Users\Joshua>tracert www.afrihost.com

Tracing route to www.afrihost.com [41.181.185.18]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 140 ms 77 ms 61 ms 41.181.51.181
4 62 ms 60 ms 59 ms 41.181.51.182
5 83 ms 82 ms 82 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.198.188]
6 79 ms 80 ms 79 ms unc-cpt-1.mtnns.net [196.44.18.8]
7 103 ms 101 ms 101 ms pe-cr-2.za--el-mse-1.za.mtnns.net [196.44.31.66]
8 120 ms 119 ms 115 ms jh-dca-1.za--jh-cr-1.za-a.mtnns.net [196.44.0.217]
9 96 ms 106 ms 91 ms 18.185.181.41.in-addr.aserv.co.za [41.181.185.18]

Trace complete.

latency climbing again.
Click to expand...

Same here....
 
8-BALL said:
C:\Users\Joshua>tracert www.afrihost.com

Tracing route to www.afrihost.com [41.181.185.18]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 140 ms 77 ms 61 ms 41.181.51.181
4 62 ms 60 ms 59 ms 41.181.51.182
5 83 ms 82 ms 82 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.198.188]
6 79 ms 80 ms 79 ms unc-cpt-1.mtnns.net [196.44.18.8]
7 103 ms 101 ms 101 ms pe-cr-2.za--el-mse-1.za.mtnns.net [196.44.31.66]
8 120 ms 119 ms 115 ms jh-dca-1.za--jh-cr-1.za-a.mtnns.net [196.44.0.217]
9 96 ms 106 ms 91 ms 18.185.181.41.in-addr.aserv.co.za [41.181.185.18]

Trace complete.

latency climbing again.
Click to expand...

:(
The latency reflects on our end too... working on having it normalised.
 
Latency creeping up again. P2P standing still.

Tracing route to www.mweb.co.za [196.2.63.110]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms wall.exa [10.0.0.254]
2 6 ms 6 ms 7 ms 105-236-13-129-esr-lo.mtnbusiness.co.za [105.236.13.129]
3 74 ms 72 ms 113 ms ipc-recieve-tb-1a.mtnbusiness.net [41.181.53.150]
4 67 ms 62 ms 79 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.198.188]
5 145 ms 144 ms 146 ms unc-cpt-1.mtnns.net [196.44.18.8]
6 165 ms 164 ms 169 ms pe-cr-2.za--el-mse-1.za.mtnns.net [196.44.31.66]
7 149 ms 154 ms 153 ms 196.44.0.72
8 158 ms 154 ms 151 ms 1122.te0-0-2-0.vic-pr-3.optinet.net [196.22.161.133]
9 123 ms 121 ms 113 ms te0-0-0-3.vic-p-2.optinet.net [197.80.7.34]
10 117 ms 116 ms 139 ms te0-0-0-0.cpt-p-1.optinet.net [197.84.4.34]
11 195 ms 192 ms 191 ms vl11.cpt-hscore-1.optinet.net [197.84.5.238]
12 166 ms 159 ms 167 ms 196.28.178.66
13 195 ms 183 ms 192 ms cte-core-sw2.vwol.net [196.41.144.35]
14 151 ms 159 ms 167 ms www.mweb.co.za [196.2.63.110]

Trace complete.
 
AfriGuy said:
Very strange indeed.
Maybe try change some ports around?

Are other downloads also slow for you?
Click to expand...

Nope, all the other services(youtube etc.) runs full speed.

The only problem is Usenet during the evening. After midnight up to early afternoon it runs almost full speed. Early evening it dies and then I can only reconnect about 11ish.

It seems to be throttled
 
Is this going to become a thing now with p2p shaped to death in the evenings?
 
standerr said:
Latency creeping up again. P2P standing still.

Tracing route to www.mweb.co.za [196.2.63.110]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms wall.exa [10.0.0.254]
2 6 ms 6 ms 7 ms 105-236-13-129-esr-lo.mtnbusiness.co.za [105.236.13.129]
3 74 ms 72 ms 113 ms ipc-recieve-tb-1a.mtnbusiness.net [41.181.53.150]
4 67 ms 62 ms 79 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.198.188]
5 145 ms 144 ms 146 ms unc-cpt-1.mtnns.net [196.44.18.8]
6 165 ms 164 ms 169 ms pe-cr-2.za--el-mse-1.za.mtnns.net [196.44.31.66]
7 149 ms 154 ms 153 ms 196.44.0.72
8 158 ms 154 ms 151 ms 1122.te0-0-2-0.vic-pr-3.optinet.net [196.22.161.133]
9 123 ms 121 ms 113 ms te0-0-0-3.vic-p-2.optinet.net [197.80.7.34]
10 117 ms 116 ms 139 ms te0-0-0-0.cpt-p-1.optinet.net [197.84.4.34]
11 195 ms 192 ms 191 ms vl11.cpt-hscore-1.optinet.net [197.84.5.238]
12 166 ms 159 ms 167 ms 196.28.178.66
13 195 ms 183 ms 192 ms cte-core-sw2.vwol.net [196.41.144.35]
14 151 ms 159 ms 167 ms www.mweb.co.za [196.2.63.110]

Trace complete.
Click to expand...

I can definitely see your latency is horrible, which is consistent with what I'm seeing from reports of overall latency in the south last night. We're still looking for any opportunity to bring latency down, but demand is extremely high during that peak window. If there is a way to get a quick win on this, I'm sure our guys will find it :(
 
HP_Exarkun said:
Nope, all the other services(youtube etc.) runs full speed.

The only problem is Usenet during the evening. After midnight up to early afternoon it runs almost full speed. Early evening it dies and then I can only reconnect about 11ish.

It seems to be throttled
Click to expand...

It sounds like maybe the increases in latency might affect your speeds or connection. I would definitely say that NNTP and P2P seem to be affected when demand is excessive and latency increases. Possibly due to the nature of the traffic and not receiving confirmation of packets received :(
 
dj2381 said:
Is this going to become a thing now with p2p shaped to death in the evenings?
Click to expand...

It's definitely not shaping, or we could give you clear guidelines on how or what is happening. This is pretty much based on demand the it's pretty unpredictable. We don't want any services to perform poorly, especially on our unshaped products, so we'll keep looking at any viable ways of keeping latency down :(
 
So you are telling me demand has increased so much since last week that p2p is now dead in the evenings?
 
dj2381 said:
So you are telling me demand has increased so much since last week that p2p is now dead in the evenings?
Click to expand...

Demand has been an issue in the evenings for a few weeks now, but P2P behaviour is very hard to predict. I do see from posts here that a few people report poor P2P or very little throughput when demand is at peak. That could be due to latency or just overall contention. We're definitely not blocking P2P on our side.
 
Oh dear this sounds more and more like the Eskom PR department spin that we get fed every day. "Load shedding has been implemented due to excessive demand."

AH consistently deny that the latency and other ongoing performance issues are caused by a capacity problem but then in the same sentence will blame it on "excessive demand". Surely these are one and the same thing and depending on your point of view there are only two possible solutions. If you choose to call it excessive demand then it is necessary to reduce the demand and that can only be done by culling customers. There is of course also shaping but that is so vehemently denied that it would not be an option. If you choose to call it "inadequate capacity" then the solution is simply to increase capacity.

As all the "fine tuning", "tweaking" and "optimizing" that has been going on round the clock for the last few months has failed to solve the problem perhaps this is not the correct path to be following.

From the tone of the posts on this thread it is becoming apparent that this issue is nearing breaking point and the lame excuses and explanations that are dished out are now beginning to make matters worse rather than better.
 
NarrowBand said:
Oh dear this sounds more and more like the Eskom PR department spin that we get fed every day. "Load shedding has been implemented due to excessive demand."

AH consistently deny that the latency and other ongoing performance issues are caused by a capacity problem but then in the same sentence will blame it on "excessive demand". Surely these are one and the same thing and depending on your point of view there are only two possible solutions. If you choose to call it excessive demand then it is necessary to reduce the demand and that can only be done by culling customers. There is of course also shaping but that is so vehemently denied that it would not be an option. If you choose to call it "inadequate capacity" then the solution is simply to increase capacity.

As all the "fine tuning", "tweaking" and "optimizing" that has been going on round the clock for the last few months has failed to solve the problem perhaps this is not the correct path to be following.

From the tone of the posts on this thread it is becoming apparent that this issue is nearing breaking point and the lame excuses and explanations that are dished out are now beginning to make matters worse rather than better.
Click to expand...

At the moment we are looking at this problem from multiple different perspectives, and temporarily increasing capacity whilst we determine what exactly is going wrong with the network management is definitely an option! And one that we are certainly considering!

The major issue with "excessive demand" has primarily been with P2P services on Uncapped accounts that should be shaped at times, have not always been shaped as thoroughly as they should be. This is generally more of a network management issue than a lack of capacity issue.
 
From hero to zero, what happened to being the no1 Isp in the country? Seems like someone upstairs has decided to cut costs to make maximum profit instead of focusing on the customers, while all these issues aren't entirely afrihosts fault, how are other providers getting it right? You guys are now bottom of.the log.
 
Status
Not open for further replies.
Back
Top
Sign up to the MyBroadband newsletter
X