Just to mention a few points and maybe I will get in trouble for it but that is what is happening currently.
Inbound traffic is a mix of Seacom and Angola Cables.
Outbound Traffic is Openserve/Telkom.
Currently Afrihost is picking up inbound traffic via Cape Town. This means the Following.
Gauteng people would have outbound go via Telkom JHB to KZN > EASSY > EU and inbound traffic Seacom > Durban > Cape Town > JHB. This causes latency.
Durban will in some cases go outbound straight over EASSY to EU but inbound EASSY > Durban > Cape Town > Durban. This is causing latency.
Traffic needs to be transit to respective pickup points there might be congestion on there. Currently there isn't enough international transit available. Peak time like now will have an issue. It's not shaping or throttling it's just there isn't any available in the short term without any very lengthy expensive contracts. That is just what it is.
Local remains unaffected. All 3 peering points are operational and if not traffic will go to the nearest one from there but NAP-DUR, NAP-CPT and NAP-JHB peering is working fine.
IPC will obviously not congest as other ISPs did mention before IPC bursts over allocations and won't ever congest on Afrihost and with the new Openserve models that would be a thing of the past.
Quickly doing a trace or ping to bras.afrihost.com will point out congestion. If you experience high latency that is a problem on Openserve side as bras.afrihost.com does not hit the Afrihost network
ping bras.afrihost.com
bgp.he.net
speedtest.mybroadband.co.za