Afrihost Business Uncapped Feedback - 2

Status
Not open for further replies.
I've been having these strange extreme slowdowns lasting for about 1-2 minutes. During that time all traffic just stops. I can barely do anything. Then it just comes back again.
Don't know if anybody has been having this.

Location: Cape Town
 
:mad: I am absolutely clueless right now to what is going on with telkom or afrihost, after waiting 4 days to get my connection back I am still sitting with a disconnect problem, always happens in the late evening or early hours, then several more times during the day, this problem appeared overnight a few weeks back, what in the hell is causing this??? that's all I want to know, but I get no answers from either, I have already opened a support ticket for the 30th time maybe, I suspect a general response of we will escalate and get a technician on it, and I will be posting the same complaint next week as it has been every week...
 
I've been having these strange extreme slowdowns lasting for about 1-2 minutes. During that time all traffic just stops. I can barely do anything. Then it just comes back again.
Don't know if anybody has been having this.

Location: Cape Town

It could be spikes in latency causing your traffic or requests to timeout. You would see this on a traceroute or speedtest if you ran it. It could also be an intermittent line, which is dropping and reconnecting continuously, which may seem like you don't lose connection but there are those short periods in between when you don't actually have a connection.

If it's latency,we'd need to see whether that is latency on the network or something internal to your network, like if you have a large download running or someone is streaming HD on the network and the player occasionally saturates the line to buffer ahead. We'd have to do some troubleshooting to see what it could be, can't really say without some data :(
 
:mad: I am absolutely clueless right now to what is going on with telkom or afrihost, after waiting 4 days to get my connection back I am still sitting with a disconnect problem, always happens in the late evening or early hours, then several more times during the day, this problem appeared overnight a few weeks back, what in the hell is causing this??? that's all I want to know, but I get no answers from either, I have already opened a support ticket for the 30th time maybe, I suspect a general response of we will escalate and get a technician on it, and I will be posting the same complaint next week as it has been every week...

There shouldn't be any need to complain, so we'll need to pick this up on our side. It sounds like you may have an intermittent line, in which case a tech has to come out and investigate what's happening.

Can you PM me so I can chase this up with our guys and Telkom?
 
I have to reset my modem every time, it doesn't reconnect by itself, it happens during downloads and during periods when there is no one on the pc, I could be watching tv and come back an hour later and it has dropped, no steam, no p2p, no updates at all and it just drops. I will do some traceroutes and post them shortly.
 
Here we go:

C:\>tracert www.ign.com

Tracing route to a1005.g.akamai.net [41.181.159.82]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.0.1
2 * * * Request timed out.
3 11 ms 10 ms 10 ms 41.181.221.253
4 10 ms 10 ms 41 ms 41.181.221.254
5 11 ms 10 ms 10 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.19
8.188]
6 12 ms 10 ms 12 ms compj-cpt-1.mtnns.net [196.44.18.2]
7 10 ms 10 ms 44 ms fosc-cpt-1.mtnns.net [196.44.18.7]
8 11 ms 10 ms 10 ms 41.181.159.82

Trace complete.

C:\>tracert www.news24.com

Tracing route to www.news24.com [41.86.110.200]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.0.1
2 * * * Request timed out.
3 103 ms 122 ms 102 ms 41.181.54.85
4 11 ms 9 ms 10 ms ipc-recieve-tb-2a.mtnbusiness.net [41.181.54.86]

5 40 ms 10 ms 43 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.19
8.188]
6 11 ms 10 ms 10 ms compj-cpt-1.mtnns.net [196.44.18.2]
7 11 ms 10 ms 11 ms ct-cr-2.za--tb-cr-1.za.mtnns.net [196.44.31.134]

8 11 ms 10 ms 11 ms 196.44.0.74
9 20 ms 21 ms 23 ms 1122.te0-0-0.cpt-pr-1.optinet.net [196.28.178.14
1]
10 22 ms 22 ms 24 ms tengig-0-0-0-1-cpt-p-1.mweb.co.za [197.84.7.1]
11 19 ms 18 ms 19 ms vl-12-cpt-hscore-2.mweb.co.za [197.84.5.254]
12 19 ms * 19 ms vp-02-14.bb.ctn.c27.za.net [196.28.178.70]
13 19 ms 19 ms 22 ms gig5-2-cpt-opt-65-1.optinet.net [196.41.133.166]

14 20 ms 19 ms 19 ms OPTI-CPT-N1-Hosting.optinet.net [197.84.196.6]
15 21 ms 19 ms 19 ms 197-84-208-146.cpt.mweb.co.za [197.84.208.146]
16 * 19 ms 19 ms 41-86-110-200.mweb.co.za [41.86.110.200]

Trace complete.

C:\>tracert www.cnn.com

Tracing route to turner.map.fastly.net [185.31.17.73]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.0.1
2 * * * Request timed out.
3 87 ms 116 ms 115 ms 41.181.53.213
4 10 ms 9 ms 10 ms ipc-send-tb-3a.mtnbusiness.net [41.181.53.214]
5 11 ms 10 ms 10 ms tb-dca-2.za--qux-i.za.mtnbusiness.net [66.8.11.1
86]
6 11 ms 11 ms 11 ms compj-cpt-1.mtnns.net [196.44.18.2]
7 154 ms 154 ms 154 ms ls-cr-2--tb-cr-1.uk-b.mtnns.net [196.44.31.113]

8 154 ms 154 ms 154 ms ls-pr-2.uk--ls-cr-2.uk-a.mtn.net [209.212.111.18
7]
9 218 ms 169 ms 196 ms 195.50.122.181
10 * * * Request timed out.
11 * * * Request timed out.
12 * * * Request timed out.
13 173 ms 174 ms 174 ms 185.31.17.73

Trace complete.

C:\>tracert www.mybroadband.co.za

Tracing route to www.mybroadband.co.za [197.242.89.170]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.0.1
2 * * * Request timed out.
3 48 ms 115 ms 11 ms 41.181.54.85
4 11 ms 10 ms 10 ms ipc-recieve-tb-2a.mtnbusiness.net [41.181.54.86]

5 11 ms 9 ms 10 ms tb-dca-2.za--qux-c.za.mtnbusiness.net [41.181.19
8.188]
6 11 ms 11 ms 10 ms unc-cpt-1.mtnns.net [196.44.18.8]
7 11 ms 12 ms 11 ms 196.44.31.106
8 11 ms 11 ms 12 ms nl-ha-2.za--rb-cr-1.za-a.mtnns.net [196.44.0.121
]
9 18 ms 17 ms 18 ms africainx.cinx.net.za [196.223.22.47]
10 35 ms 34 ms 34 ms CORE.GP-CN-HET-MEE-1.TO.GP-HV-ICT-MEE-1.DFA.P2P.
10G.za [41.84.13.66]
11 35 ms 34 ms 34 ms 41-66-132-246-f6.HET001-CPE-1-to-GP-CN-HET-MEE-1
.africainx.net [41.66.132.246]
12 * 35 ms 35 ms core-access-switch1.jnb1.host-h.net [197.189.193
.1]
13 * 40 ms 35 ms row-access-switch1-row3-4.jnb1.host-h.net [197.1
89.193.36]
14 * 35 ms 35 ms 197.242.89.170

Trace complete.

C:\>
 
It could be spikes in latency causing your traffic or requests to timeout. You would see this on a traceroute or speedtest if you ran it. It could also be an intermittent line, which is dropping and reconnecting continuously, which may seem like you don't lose connection but there are those short periods in between when you don't actually have a connection.

If it's latency,we'd need to see whether that is latency on the network or something internal to your network, like if you have a large download running or someone is streaming HD on the network and the player occasionally saturates the line to buffer ahead. We'd have to do some troubleshooting to see what it could be, can't really say without some data :(

I'll monitor it and see if it happens again. Was worried it has something to do with the changes you guys are rolling out.
The connection is definitely up during these dips. I'm monitoring the throughput on my firewall and everything just comes to a complete standstill. All traffic throughput drops to almost nothing, including running downloads. Opening a browser tab waits for about 10 seconds before the actual page starts loading.

Unfortunately I have not been able to do a tracert while this is happening. Only happened twice this morning. Last one was about 7h45.
 
A technician has already been here, the problem is at the exchange, he said he would either do some software changes at the exchange or move me to a new port, I suppose I'm still on the same port with the same problems.

Edit: I would jump on that VDSL bandwagon if there was an option for a 10MB/s line, according to the tech, I would have a much more stable line with more realistic speeds, unfortunately the 20Mb/s option is out of reach for me.
 
Last edited:
At Arfrihost: I think your routing and QoS is generally messed up. I am not going to do diagnostics for you, as it is quite simple for you to run a "traceroute 199.189.118.227" from your own network and then compare it to anyone of your competitors and you will see that your routing in general is messed up. This is not even an IPC issue. Hence the erratic and poor performance of your network when it comes to gaming.

FWIW - connected my PS4 to the IS network and had absolutely no issue. Switched back to Afrihost and lag.
 
A technician has already been here, the problem is at the exchange, he said he would either do some software changes at the exchange or move me to a new port, I suppose I'm still on the same port with the same problems.

Edit: I would jump on that VDSL bandwagon if there was an option for a 10MB/s line, according to the tech, I would have a much more stable line with more realistic speeds, unfortunately the 20Mb/s option is out of reach for me.

It does look like it could be congestion or an line/exchange issue, so make sense from the tech. I wish that we could get the 2nd hop ping again, seems like it's gone away since many exchanges were upgraded.

I doubt they'll do 10Mbps over VDSL, but it depends. If they completely phase out ADSL (i.e 4Mbps and 2Mbps) then they may consider it, so everything would be more fibre based, which would be great. It must be an ultimate plan somewhere. I would guess though that once fibre is more wide-spread, it will be 20Mbps and up (or maybe even starting at 40Mbps) - at least on paper.
 
At Arfrihost: I think your routing and QoS is generally messed up. I am not going to do diagnostics for you, as it is quite simple for you to run a "traceroute 199.189.118.227" from your own network and then compare it to anyone of your competitors and you will see that your routing in general is messed up. This is not even an IPC issue. Hence the erratic and poor performance of your network when it comes to gaming.

FWIW - connected my PS4 to the IS network and had absolutely no issue. Switched back to Afrihost and lag.

It is important for us to have a sense of the client experience. We do analyse our routing and generally an additional hop or two here will not add additional significant latency to the destination.

We're still pushing out major updates and changes so running those kind of tests now would be moot, it's best to monitor as changes are implemented and then test end-to-end when the new changes are in place and pass all our system checks.
 
I'll monitor it and see if it happens again. Was worried it has something to do with the changes you guys are rolling out.
The connection is definitely up during these dips. I'm monitoring the throughput on my firewall and everything just comes to a complete standstill. All traffic throughput drops to almost nothing, including running downloads. Opening a browser tab waits for about 10 seconds before the actual page starts loading.

Unfortunately I have not been able to do a tracert while this is happening. Only happened twice this morning. Last one was about 7h45.

We're definitely not making changes in the mornings, so that is not likely to be the cause. Drop me a PM and let's see how we can troubleshoot this :(
 
I would take a 20Mb's VDSL line with a 10Mb/s business uncapped option, but that's no longer a bundle which means the prices would go up :cry:

Edit: Actually would that be possible? I just checked and its not too much more than I am paying right now.
 
Last edited:
I would take a 20Mb's VDSL line with a 10Mb/s business uncapped option, but that's no longer a bundle which means the prices would go up :cry:

Edit: Actually would that be possible? I just checked and its not too much more than I am paying right now.

You don't have to match your line and data in a bundle. You should be able to test out possible combinations on the order form.
 
I've been having issues with hanging web sites lately. pinging google results in a lot of dropped packets. here is a tracert

Tracing route to www.google.co.za [216.58.223.35]
over a maximum of 30 hops:

1 1 ms 1 ms 1 ms 192.168.1.254
2 33 ms 31 ms 31 ms 105-236-15-1-esr-lo.mtnbusiness.co.za [105.236.15.1]
3 * 33 ms * ipc-recieve-jh-1a.za.mtnbusiness.net [41.181.178.5]
4 36 ms 35 ms 34 ms qux-jh-dca-2.za-b.za.mtnbusiness.net [41.181.165.115]
5 34 ms 34 ms 34 ms jh-dca-2.za--qux-q.za.mtnbusiness.net [196.31.180.4]
6 47 ms 43 ms 50 ms 41.181.180.10
7 36 ms 37 ms 37 ms rb-cr-2.za--jh-cr-1.za.mtnns.net [196.44.0.43]
8 54 ms 48 ms 47 ms 41.181.139.99
9 * 38 ms 34 ms 72.14.194.74
10 * 38 ms 34 ms 72.14.239.129
11 38 ms 38 ms 37 ms jnb01s08-in-f3.1e100.net [216.58.223.35]
 
I've been having issues with hanging web sites lately. pinging google results in a lot of dropped packets. here is a tracert

Tracing route to www.google.co.za [216.58.223.35]
over a maximum of 30 hops:

1 1 ms 1 ms 1 ms 192.168.1.254
2 33 ms 31 ms 31 ms 105-236-15-1-esr-lo.mtnbusiness.co.za [105.236.15.1]
3 * 33 ms * ipc-recieve-jh-1a.za.mtnbusiness.net [41.181.178.5]
4 36 ms 35 ms 34 ms qux-jh-dca-2.za-b.za.mtnbusiness.net [41.181.165.115]
5 34 ms 34 ms 34 ms jh-dca-2.za--qux-q.za.mtnbusiness.net [196.31.180.4]
6 47 ms 43 ms 50 ms 41.181.180.10
7 36 ms 37 ms 37 ms rb-cr-2.za--jh-cr-1.za.mtnns.net [196.44.0.43]
8 54 ms 48 ms 47 ms 41.181.139.99
9 * 38 ms 34 ms 72.14.194.74
10 * 38 ms 34 ms 72.14.239.129
11 38 ms 38 ms 37 ms jnb01s08-in-f3.1e100.net [216.58.223.35]

Would you be able to check what the current DNS settings on your ADSL modem are?

It should probably be on Automatic, would you mind switching that to manual then entering the following DNS Settings:
Primary DNS Server: 196.7.7.7
Secondary DNS Server: 8.8.8.8

Can you then please test your connection to Google services and see if there is any improvement.
 
Would you be able to check what the current DNS settings on your ADSL modem are?

It should probably be on Automatic, would you mind switching that to manual then entering the following DNS Settings:
Primary DNS Server: 196.7.7.7
Secondary DNS Server: 8.8.8.8

Can you then please test your connection to Google services and see if there is any improvement.

DNS is not going to cause this:
3 * 33 ms * ipc-recieve-jh-1a.za.mtnbusiness.net [41.181.178.5]

That's 2/3 packets dropped at your IPC.
 
DNS is not going to cause this:
3 * 33 ms * ipc-recieve-jh-1a.za.mtnbusiness.net [41.181.178.5]

That's 2/3 packets dropped at your IPC.

Traceroutes don't really give a detailed break down of packet loss, to do that you will need to use something like Visual Route.

If the DNS isn't responding quickly enough it can also result in time outs on traceroutes.
 
Just use -d like so

tracert -d mybroadband.co.za

to stop DNS lookups.

Tracing route to mybroadband.co.za [197.242.89.170]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.0.1
2 6 ms 6 ms 6 ms 105.236.14.1
3 16 ms 16 ms 16 ms 41.181.221.218
4 16 ms 55 ms 16 ms 41.181.165.118
5 18 ms 17 ms 18 ms 196.44.0.155
6 17 ms 17 ms 24 ms 196.44.0.42
7 17 ms 17 ms 17 ms 196.44.0.72
8 * * * Request timed out.
9 39 ms 40 ms 41 ms 41.84.13.66
10 42 ms 39 ms 42 ms 41.66.132.246
11 44 ms 42 ms 44 ms 197.189.193.1
12 43 ms 42 ms 42 ms 197.189.193.36
13 40 ms 37 ms 36 ms 197.242.89.170
 
Traceroutes don't really give a detailed break down of packet loss, to do that you will need to use something like Visual Route.

If the DNS isn't responding quickly enough it can also result in time outs on traceroutes.

Dns timeouts will not cause stars on traces. They will cause numeric addresses to appear.

A star in a traceroute means a timeout receiving the icmp ttl exceeded packet expected.
It is a direct irrefutable sign of packet loss (assuming that hop does function correctly in traceroutes, which it does in this case ) .
 
Status
Not open for further replies.
Top
Sign up to the MyBroadband newsletter
X