Afrihost - Pure Fibre Feedback Thread

Status
Not open for further replies.
waylander said:
Had multiple tests running over the last 12 hours:

1. CS:GO cannot connect to servers hosted at Xneelo.
2. DNS requests have some sort of delay/queue/buffer going on that delays requests to even the closest ones hosted at NAP/JINX
3. My iPhone and iPad cannot connect to many services, but Facebook/Instagram/Whatsapp/Wikipedia work fine.
4. Intermittent 10-15 second outages every now and then

Something is up ya'll fr
Click to expand...
Hi waylander, this is definitely not good, 12 hours is enough time to get an accurate reading. Which service do you have with us? Can you send me the email address linked to it so that I can also take a look?
 
[
Seeyou said:
@ Any AH rep - is there any further confirmation of a fibre cancellation/migration to another ISP beyond the client zone status saying the cancellation is effective 2020-05-31? Can I trust the line will be released on 1 June or do I need to confirm with someone?
Click to expand...

Hi Seeyou, you don't any further confirmation than that. The line will definitely be released on the first if the cancellation is effective on the 31st of May as per Clientzone.
 
Icemanbrfc said:
Mitchells Fibre/Vuma Reach.. Free standing
Click to expand...
There's sometimes a glitch on the Vumatel portal where addresses need to be created, even though there's fibre coverage. This is referred to as a premise creation. This could be the case for your buddy, you are welcome to PM me his email address so that I can take a look into his profile.
 
AfriNatic said:
There's sometimes a glitch on the Vumatel portal where addresses need to be created, even though there's fibre coverage. This is referred to as a premise creation. This could be the case for your buddy, you are welcome to PM me his email address so that I can take a look into his profile.
Click to expand...
Pm sent
 
CF buggering around again, I see i'm being routed to DBN instead of CPT

Code: 
traceroute to www.iol.co.za (104.22.29.113), 20 hops max, 52 byte packets
 1  10.23.23.1 (10.23.23.1)  1.123 ms  0.451 ms  0.391 ms
 2  * * *
 3  165-73-14-113.ip.afrihost.co.za (165.73.14.113)  6.530 ms  5.040 ms  5.516 ms
 4  169-1-21-2.ip.afrihost.co.za (169.1.21.2)  6.249 ms  2.345 ms  3.442 ms
 5  169-1-21-3.ip.afrihost.co.za (169.1.21.3)  2.933 ms
    169-1-21-2.ip.afrihost.co.za (169.1.21.2)  3.199 ms  3.120 ms
 6  169-1-21-3.ip.afrihost.co.za (169.1.21.3)  3.542 ms  5.129 ms
    102.67.176.74 (102.67.176.74)  19.722 ms
 7  vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  33.065 ms
    vlan1724.terdbn-igw1.net.echosp.link (102.67.176.29)  30.307 ms
    vlan1624.terdbn-igw1.net.echosp.link (102.67.176.25)  29.021 ms
 8  vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  32.163 ms
    vlan1231.terdbn-igw1.net.echosp.link (102.67.176.13)  30.164 ms
    vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  30.897 ms
 9  196-10-141-198.ixp.durban (196.10.141.198)  30.837 ms  30.587 ms  30.990 ms
10  104.22.29.113 (104.22.29.113)  29.587 ms  30.075 ms  29.395 ms
 
SpoonSpatula said:
pings were erratic on Thursday.

And performance on video conferencing apps during the day have been horrendous (stuttering, latency, defaulting to lowest video quality, etc).
Click to expand...

You are lucky you get 9ms ping

My ping sits nothing below 38ms

Today however, had a zoom class for my kid and on 100mb fibre it stuttered like a mofo

I put everything off in the house to try make it manageable.
 
blunt said:
CF buggering around again, I see i'm being routed to DBN instead of CPT

Code: 
traceroute to www.iol.co.za (104.22.29.113), 20 hops max, 52 byte packets
1  10.23.23.1 (10.23.23.1)  1.123 ms  0.451 ms  0.391 ms
2  * * *
3  165-73-14-113.ip.afrihost.co.za (165.73.14.113)  6.530 ms  5.040 ms  5.516 ms
4  169-1-21-2.ip.afrihost.co.za (169.1.21.2)  6.249 ms  2.345 ms  3.442 ms
5  169-1-21-3.ip.afrihost.co.za (169.1.21.3)  2.933 ms
    169-1-21-2.ip.afrihost.co.za (169.1.21.2)  3.199 ms  3.120 ms
6  169-1-21-3.ip.afrihost.co.za (169.1.21.3)  3.542 ms  5.129 ms
    102.67.176.74 (102.67.176.74)  19.722 ms
7  vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  33.065 ms
    vlan1724.terdbn-igw1.net.echosp.link (102.67.176.29)  30.307 ms
    vlan1624.terdbn-igw1.net.echosp.link (102.67.176.25)  29.021 ms
8  vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  32.163 ms
    vlan1231.terdbn-igw1.net.echosp.link (102.67.176.13)  30.164 ms
    vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  30.897 ms
9  196-10-141-198.ixp.durban (196.10.141.198)  30.837 ms  30.587 ms  30.990 ms
10  104.22.29.113 (104.22.29.113)  29.587 ms  30.075 ms  29.395 ms
Click to expand...
Were you connected to the router or ONT directly during the test? The latency on the first hop is a little concerning.
 
scienide said:
You are lucky you get 9ms ping

My ping sits nothing below 38ms

Today however, had a zoom class for my kid and on 100mb fibre it stuttered like a mofo

I put everything off in the house to try make it manageable.
Click to expand...
Hi scienide, Do you mind if I have a look into this? Please PM the email address linked to your profile.
 
blunt said:
CF buggering around again, I see i'm being routed to DBN instead of CPT

Code: 
traceroute to www.iol.co.za (104.22.29.113), 20 hops max, 52 byte packets
1  10.23.23.1 (10.23.23.1)  1.123 ms  0.451 ms  0.391 ms
2  * * *
3  165-73-14-113.ip.afrihost.co.za (165.73.14.113)  6.530 ms  5.040 ms  5.516 ms
4  169-1-21-2.ip.afrihost.co.za (169.1.21.2)  6.249 ms  2.345 ms  3.442 ms
5  169-1-21-3.ip.afrihost.co.za (169.1.21.3)  2.933 ms
    169-1-21-2.ip.afrihost.co.za (169.1.21.2)  3.199 ms  3.120 ms
6  169-1-21-3.ip.afrihost.co.za (169.1.21.3)  3.542 ms  5.129 ms
    102.67.176.74 (102.67.176.74)  19.722 ms
7  vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  33.065 ms
    vlan1724.terdbn-igw1.net.echosp.link (102.67.176.29)  30.307 ms
    vlan1624.terdbn-igw1.net.echosp.link (102.67.176.25)  29.021 ms
8  vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  32.163 ms
    vlan1231.terdbn-igw1.net.echosp.link (102.67.176.13)  30.164 ms
    vlan1131.terdbn-igw1.net.echosp.link (102.67.176.5)  30.897 ms
9  196-10-141-198.ixp.durban (196.10.141.198)  30.837 ms  30.587 ms  30.990 ms
10  104.22.29.113 (104.22.29.113)  29.587 ms  30.075 ms  29.395 ms
Click to expand...

Yeah Cloudflare is messing around again.

Cloudflare Status

Welcome to Cloudflare's home for real-time and historical data on system performance.
www.cloudflarestatus.com www.cloudflarestatus.com

You are going via JHB to DUR btw. 102.67.176.74 is in JHB.

Mybroadband.co.za is going to London too.
 
SpoonSpatula said:
I've also been getting very poor international speeds this week (issue started around Tuesday/Wednesday?). Accessing certain sites during the day have been hit and miss (at times it wouldn't let me load MyBB, Google, etc). Strangely I have not experienced any packet loss. Steam, Battle.Net, etc also dropped completely on Wednesday night and pings were erratic on Thursday.

And performance on video conferencing apps during the day have been horrendous (stuttering, latency, defaulting to lowest video quality, etc).

I've attached a few speedtest results. Using Openserve, located in PTA East. 200 down/100 up. I have had no issues with local speeds thus far.

Local:

View attachment 840777

View attachment 840779

London:

View attachment 840781

View attachment 840783

US:

View attachment 840785
Click to expand...

Please take note that Speedtesting to Vodafone, Vox, Cool Ideas servers in London, Germany, Amsterdam is not a clear indication of issues on Afrihost's network. The data will be served by the respective ISP via peering. It won't give an accurate reading of international issues because if those ISPs are suffering some transit issues it will seem as if Afrihost is slow when it's not.

Below I have done trace route to the Vox Telecom speedtest server in London. As you can see traffic is handed over to Vox in JHB and from there it's all Vox network. Vox uses EASSy for EU traffic from JHB. This means that even though Afrihost uses WACS the traffic will go over Vox network via EASSy giving the impression of high latency to EU.

vox_london.png


A quick trace route to CoreIX London server gives me the below result. Both servers are at the same IX in LINX but the latency is speed is different for both. The Vox one above goes via Vox network and the trace route below goes via Afrihosts own transit. Also take not Core IX London is a very busy server with a lot of people especially in SA doing tests to it due to it being a neutral server so the server might be struggling bandwidth wise too.

coreix_lon.png
 
Last edited:
waylander said:
I'll be running tests via the ONT to rule out anything in my (exotic) network environment at home - I just can't figure out why this all happened all of a sudden
Click to expand...

I had a similar issue the other day

openserve_centurion_fault6.png
 
waylander said:
Did you manage to find out what it was? I've wrecked my brain swapping out cables, moving my APs around, firmware updates, resetting configs... last resort is plugging a desktop into the ONT and seeing if its anything related to the line.
Click to expand...

I tried everything but did not get a solution. I though it was DNS related. The drops would make my Tenda AC11 restart so I though the DNS replied with something that my Tenda did not like or had a firmware bug and then trigger the restart. Switched over to the TP-Link and it did the same minus the restarts. Now have the Huawei H112-370 hooked up and the issue is no more. Not sure what it was or if it's still there but the Huawei seems fine.
 
Last edited:
LabAnimal said:
Whats up Afrihost, slow slow slow!. :(
Click to expand...
waylander said:
I'll be running tests via the ONT to rule out anything in my (exotic) network environment at home - I just can't figure out why this all happened all of a sudden
Click to expand...
It aint you - Afrihost is the problem - been bad on and off for the past few days. I'm on a 20meg line, and i'm watching pics on popular pages take its time to buffer and load... takes about 20seconds for 1 averagely small pic....
 
LabAnimal said:
It aint you - Afrihost is the problem - been bad on and off for the past few days. I'm on a 20meg line, and i'm watching pics on popular pages take its time to buffer and load... takes about 20seconds for 1 averagely small pic....
Click to expand...

Cloudflare DUR and CPT is rerouted and most traffic pulled from JHB or London. Might be the reason why you are getting sluggish performance. What does speedtests show?
 
cavedog said:
Cloudflare DUR and CPT is rerouted and most traffic pulled from JHB or London. Might be the reason why you are getting sluggish performance. What does speedtests show?
Click to expand...
1.8meg down, 3.7 up, 89ms ping. it says its local server is 123net. veeeeeeeeeeeery slow. even the speed test took long to load.
 
LabAnimal said:
1.8meg down, 3.7 up, 89ms ping. it says its local server is 123net. veeeeeeeeeeeery slow. even the speed test took long to load.
Click to expand...

Yes that is slow what does trace route look like? Openserve?
 
Tracing route to openserve.co.za [165.143.151.55]
over a maximum of 30 hops:

1 <1 ms 2 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 110 ms 99 ms 100 ms 165-73-14-65.ip.afrihost.co.za [165.73.14.65]
4 61 ms 71 ms 99 ms 165-73-14-90.ip.afrihost.co.za [165.73.14.90]
5 102 ms 99 ms 109 ms dbn-in1.ip.adsl.co.za [169.1.5.204]
6 * 101 ms 101 ms rndf-ip-ll-1-wan.telkom-ipnet.co.za [196.25.134.29]
7 72 ms 81 ms 102 ms 196.25.146.8
8 111 ms 106 ms 110 ms 196.25.146.253
9 * 107 ms 107 ms 196.43.51.78
10 104 ms 112 ms 111 ms telkom-sa-tienet-gw.telkom-ipnet.co.za [196.25.54.62]
11 112 ms 111 ms 112 ms 165.143.151.55

Trace complete.
 
Status
Not open for further replies.
Back
Top
Sign up to the MyBroadband newsletter
X