South Africa’s biggest forum. Discuss, discover, and connect with thousands of members.
Nope
MTN don't have anything to do with our network management.
I have a question. If the accounts aren't being shaped, is Afrihost forcing Protocol Prioritisation for anyone who currently has it off? Would having your customers turning it on help with easing the load?
BWAHAHAHA! And how many issues have you claimed has been escalated to MTN?
It's ironic how "congestion" only affects P2P protocols, but none of the others... It's not often that I get to say this, but it looks like your support staff has a better clue as to what is actually going on. You just haven't had the time to tell them what they can / can't admit yet. Fortunately for us, the truth came out before you could hush them....
As mentioned, we're well aware that the accounts aren't performing as they should - and we're addressing this as a top priority.
We've been extremely transparent about this, from the get go.
This is due to an increase in demand.
Again, this is by no means a permanent way forward for the accounts. They will return to what they used to be.
You still haven't answered my question. How is it possible that P2P is slower than Speedtest.net? Surely if there was no shaping/prioritisation/optimisation then they would run at similar speeds ?
You still haven't answered my question. How is it possible that P2P is slower than Speedtest.net? Surely if there was no shaping/prioritisation/optimisation then they would run at similar speeds ?
Can we have a date?
Because all speedtest sites are whitelisted, no ISP will ever confirm this though....
AH this is when as a "respectable" ISP suppose to come up with compensation for your clients.
I'm sure this is something being looked into my Management, it's not something that I can authorise at this stage though![]()
Humour me please... as a customer who spends R7000+ with AH per month ... How is it possible that P2P is slower than let's say a Speedtest.net measurement?
You need to remember that speedtests are run via tiny packets of data whereas torrents are multiple threads, larger packets across the network.
Simply put, if these packets become too much a contention level is reached where they then degrade.
I'm sure this is something being looked into my Management, it's not something that I can authorise at this stage though![]()
Because all the naughty iOS update users are blocking the system and there is an unshaped bottleneck that has nothing to do with the IPC but is being optimised and monitored and they are sorry but it will be back to speed soon soon.
Kapeeeesh!
I don't mean to sound obtuse but if you could point me in the direction of which expert said that torrents and http run at different speeds because of variable packet size? But for the sake of debate then let's assume that somehow this is true... Then why if I run a web server (plain ol' HTTP) listening on tcp port 80, tcp port 40000 and tcp port 10000 with a single file download with single thread do I get different speeds for downloading from each?