Afrihost Capped DSL Feedback

Latency to Cape Town has spiked to 180ms. The moment this happens performance goes through the floor. Unfortunately there is no explanation other than congestion in CT. It's the same thing every time/evening. And the same hop every time.

No problem to BRAS:

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| router.asus.com - 0 | 111 | 111 | 0 | 0 | 1 | 0 |
| 155.239.255.250 - 0 | 111 | 111 | 16 | 16 | 18 | 17 |

But there's my old friend causing the problem - cpt-rx1.ip.adsl.co.za. Obviously a busy boy in the evenings.

| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| router.asus.com - 0 | 57 | 57 | 0 | 0 | 1 | 0 |
| No response from host - 100 | 11 | 0 | 0 | 0 | 0 | 0 |
| cpt-tx2.ip.adsl.co.za - 7 | 45 | 42 | 40 | 41 | 55 | 41 |
| cpt-rx1.ip.adsl.co.za - 2 | 53 | 52 | 168 | 192 | 206 | 193 |
| cpt-net1.ip.adsl.co.za - 0 | 57 | 57 | 40 | 42 | 82 | 82 |
| optinet.ixp.capetown - 0 | 57 | 57 | 41 | 42 | 50 | 42 |
| bu-ether1.cpt-p-1.optinet.net - 14 | 36 | 31 | 165 | 214 | 231 | 215 |
| te0-7-0-3.cpt-p-2.optinet.net - 0 | 56 | 56 | 178 | 213 | 229 | 207 |
| tengig-0-7-0-1-vic-p-1.mweb.co.za - 0 | 57 | 57 | 59 | 72 | 116 | 60 |
| 197-84-4-101.cpt.mweb.co.za - 0 | 57 | 57 | 59 | 61 | 80 | 61 |
| 201.bu-ether5.vic-pe-2.optinet.net - 3 | 48 | 47 | 168 | 202 | 217 | 210 |
| 197-81-226-81.jhb.mweb.co.za - 2 | 53 | 52 | 146 | 188 | 210 | 199 |
| 197-81-226-61.jhb.mweb.co.za - 0 | 57 | 57 | 58 | 62 | 85 | 59 |
| 197-81-229-4.jhb.mweb.co.za - 2 | 53 | 52 | 58 | 59 | 66 | 59 |
| 197-81-229-15.jhb.mweb.co.za - 2 | 53 | 52 | 177 | 209 | 223 | 211 |
| 41-86-112-68.mweb.co.za - 0 | 57 | 57 | 58 | 58 | 67 | 58 |


During the day I can get that latency to London, UK!!

:( There is definitely something off there. I've asked our Network Team to follow-up on these results.
 
And the local loss in durban is back................................

Edit: it's not on *.adsl.co.za

It's neotel: 41.162.84.192
 
And the local loss in durban is back................................

Edit: it's not on *.adsl.co.za

It's neotel: 41.162.84.192

Could you maybe post the entire traceroute so that I can check what's going on.
 
Were you able to test over the weekend? How did the latency hold-up?


Trace complete.
PS C:\Users\Oliver> tracert bbc.co.uk
Unable to resolve target system name bbc.co.uk.
PS C:\Users\Oliver> tracert bbc.co.uk

Tracing route to bbc.co.uk [212.58.244.23]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms csp1.zte.com.cn [192.168.1.1]
2 * * * Request timed out.
3 19 ms 18 ms 19 ms jhb-up3.ip.adsl.co.za [169.1.5.29]
4 100 ms 98 ms 99 ms jhb-in2.ip.adsl.co.za [169.1.5.30]
5 100 ms 100 ms 100 ms jhb-net1.ip.adsl.co.za [169.1.5.52]
6 94 ms 94 ms 92 ms 169-1-5-74.ip.afrihost.co.za [169.1.5.74]
7 87 ms 84 ms 84 ms 41.169.57.216
8 20 ms 19 ms 19 ms 13.179.206.41.in-addr.arpa [41.206.179.13]
9 279 ms 265 ms 259 ms 85.55.6.216.in-addr.arpa [216.6.55.85]
10 202 ms 194 ms 202 ms 2.178.206.41.in-addr.arpa [41.206.178.2]
11 249 ms 250 ms 257 ms 109.159.231.80.in-addr.arpa [80.231.159.109]
12 189 ms 189 ms 189 ms 5.158.231.80.in-addr.arpa [80.231.158.5]
13 182 ms 181 ms 181 ms 30.158.231.80.in-addr.arpa [80.231.158.30]
14 186 ms 186 ms 186 ms 1.139.231.80.in-addr.arpa [80.231.139.1]
15 * * 186 ms 42.139.231.80.in-addr.arpa [80.231.139.42]
16 187 ms 202 ms 187 ms 130.130.231.80.in-addr.arpa [80.231.130.130]
17 3372 ms 1581 ms 435 ms 195.219.83.102
18 * * * Request timed out.
19 * * * Request timed out.
20 196 ms 204 ms 199 ms NIU-SOLUTIO.ear1.London15.Level3.net [212.187.173.210]
21 * * * Request timed out.
22 265 ms 248 ms 235 ms ae0.er01.telhc.bbc.co.uk [132.185.254.109]
23 197 ms 197 ms 196 ms 132.185.255.148
24 196 ms 195 ms 195 ms 212.58.244.23


So I guess not great?.... This is right now.
 
Trace complete.
PS C:\Users\Oliver> tracert bbc.co.uk
Unable to resolve target system name bbc.co.uk.
PS C:\Users\Oliver> tracert bbc.co.uk

Tracing route to bbc.co.uk [212.58.244.23]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms csp1.zte.com.cn [192.168.1.1]
2 * * * Request timed out.
3 19 ms 18 ms 19 ms jhb-up3.ip.adsl.co.za [169.1.5.29]
4 100 ms 98 ms 99 ms jhb-in2.ip.adsl.co.za [169.1.5.30]
5 100 ms 100 ms 100 ms jhb-net1.ip.adsl.co.za [169.1.5.52]
6 94 ms 94 ms 92 ms 169-1-5-74.ip.afrihost.co.za [169.1.5.74]
7 87 ms 84 ms 84 ms 41.169.57.216
8 20 ms 19 ms 19 ms 13.179.206.41.in-addr.arpa [41.206.179.13]
9 279 ms 265 ms 259 ms 85.55.6.216.in-addr.arpa [216.6.55.85]
10 202 ms 194 ms 202 ms 2.178.206.41.in-addr.arpa [41.206.178.2]
11 249 ms 250 ms 257 ms 109.159.231.80.in-addr.arpa [80.231.159.109]
12 189 ms 189 ms 189 ms 5.158.231.80.in-addr.arpa [80.231.158.5]
13 182 ms 181 ms 181 ms 30.158.231.80.in-addr.arpa [80.231.158.30]
14 186 ms 186 ms 186 ms 1.139.231.80.in-addr.arpa [80.231.139.1]
15 * * 186 ms 42.139.231.80.in-addr.arpa [80.231.139.42]
16 187 ms 202 ms 187 ms 130.130.231.80.in-addr.arpa [80.231.130.130]
17 3372 ms 1581 ms 435 ms 195.219.83.102
18 * * * Request timed out.
19 * * * Request timed out.
20 196 ms 204 ms 199 ms NIU-SOLUTIO.ear1.London15.Level3.net [212.187.173.210]
21 * * * Request timed out.
22 265 ms 248 ms 235 ms ae0.er01.telhc.bbc.co.uk [132.185.254.109]
23 197 ms 197 ms 196 ms 132.185.255.148
24 196 ms 195 ms 195 ms 212.58.244.23


So I guess not great?.... This is right now.

Seeing a spike on the 17th hop, but it doesn't seem to affect latency at the destination. Are you experiencing slow access?
 
Seeing a spike on the 17th hop, but it doesn't seem to affect latency at the destination. Are you experiencing slow access?

Access is extremely slow and ping is sitting around 250-280 on League of Legends, compared to my normal 200. My worries is the 4th* hop... It usually sits at around 20ms.
 
I don't think this clearly points to a capacity issue. We're not seeing high latency or packet loss on our network at the moment. Please send me a traceroute test and ping to bras.afrihost.com and we'll be able to see the origin of your latency more clearly.

Every single night the latency rises and packet loss occurs. You can see that is graphed and that is clear. I will try get a traceroute but latency to bras.afrihost.com is fine as you can see on the same site I linked.
 
So this might be related, but isn't there issues with the Seacom cable at this time?
Wouldn't that affect International traffic?

And since WACS is running from Cape Town isn't that maybe why I'm getting better results on WA that utilizes IS?
 
Afri Gents, why does the sms message say i have used 140.491 GB of 220.461, but CZ says i have about 20 gigs left. Im confused a bit here :confused:
 
Afri Gents, why does the sms message say i have used 140.491 GB of 220.461, but CZ says i have about 20 gigs left. Im confused a bit here :confused:

Uh... I have no idea... Could you PM me the username and I'll check for you quickly.
 
Access is extremely slow and ping is sitting around 250-280 on League of Legends, compared to my normal 200. My worries is the 4th* hop... It usually sits at around 20ms.

If it happens again, could you run a traceroute to Afrihost.com or the LoL server please. Our Network Team is investigating any reports of latency but we'll need the traceroute results to figure out what's going on.
 
So this might be related, but isn't there issues with the Seacom cable at this time?
Wouldn't that affect International traffic?

And since WACS is running from Cape Town isn't that maybe why I'm getting better results on WA that utilizes IS?

We don't use Seacom though. :( All our international bandwidth runs over WACS.
 
Top
Sign up to the MyBroadband newsletter
X