Official Web Africa thread

Vamp1r3 said:
Literally unusable. Barely loads web pages, can't even stream 144p youtube. Their own website doesn't even load.

The drops in speed and massive increase in latency are consistent every single night between 20h00 - 23h00. What is happening? Why is this suddenly an issue?

What the hell has happened to this ISP
Click to expand...

I am having the same issues as you Vamp1r3 and convex , whats happening webafrica ?? been great and now its unusable
 
Update from the network team
We have isolated this and have confirmed that there is an issue on a segment of our network affecting certain users and are we working to resolve this with our providers as fast as we can.
 
Vamp1r3 said:
Literally unusable. Barely loads web pages, can't even stream 144p youtube. Their own website doesn't even load.

The drops in speed and massive increase in latency are consistent every single night between 20h00 - 23h00. What is happening? Why is this suddenly an issue?

What the hell has happened to this ISP
Click to expand...
Thought it was just me, ffs.

No notice on Web Africa's website.
 
WebAfrica Helper said:
Update from the network team
We have isolated this and have confirmed that there is an issue on a segment of our network affecting certain users and are we working to resolve this with our providers as fast as we can.
Click to expand...

Heard this one before. ISP crowdsourcing network repair?
 
infection said:
So,

Today I finally had someone from WebAfrica call me about my ticket I opened in February (rather late than never I guess).

Long story short, they've given me a test account with another ISP (I will keep the name out for now) and requested some trace routes.

So, I've decided run the exact same tests on both accounts. Made sure all the devices on the network except for the computer I am running the test from was physically disconnected. WiFi turned off as well.

WinMTR
Ran 20 iterations with the destination being www.webafrica.co.za. To ensure that there is no skewed results with packets being flooded before a response is received I've set it to a 5 second interval rate between each iteration and a 32 bytes size as well as disabling resolving host names.
WebAfrica - 3rd hop @ 70% packet loss. Latency was better than normal during the test.
Other ISP - 0% packet loss on the whole route. Saw some higher latency at certain hops on the route, but everything still under 100ms.

The funny thing on this test was, I had better latency on the last hop (www.webafrica.co.za) from the Other ISP at an average of 64ms compared to that of WebAfrica itself at an average of 790ms.

Ping
Just a standard ping, running 500 iterations with the destination being www.webafrica.co.za
WebAfrica - Average latency was actually good during this test and the packet loss wasn't high at all. Ended up with 11ms minimum, 199ms maximum and 15ms average latency with a 2% packet loss.
Other ISP - Average latency was good during this test and there was zero (that's right, nada) packet loss. Ended up with 10ms minimum, 367ms maximum and 14ms average latency.

tracert
Good old trace route that everyone asks for, is in-accurate at determining packet loss and latency (if faults are not consistent) and takes ages to run. So, with this test I ran 50 tracert's to www.webafrica.co.za. For this test I set a rule, if the third hop (which is on the ISP, or their backbones', network and the hop my problem has been consistent on) is either above 100ms or times out, I count it as a bad trace route.
WebAfrica - Out of the 50 trace routes ran, 26 (52%) of them was flagged as a bad trace route with either timing out or high latency, with some reaching 2000+ms.
Other ISP - Out of the 50 trace routes ran 2 (4%) of them was flagged as a bad trace route with one timing out and the other only have the first RTT indicating a time out while the others returned a response time, but I still counted it as bad.

Pathping
As a last test, I ran a pathping to www.webafrica.co.za and left all settings at default. This means it would run the path 100 times.
WebAfrica - Again here hop 3 was consistent on it's bad "performance" with an RTT of 927ms and 63% packet loss.
Other ISP - Right of the bat this looks like an acceptable pathping report. No high latency, with the 3rd hop sitting at a solid 13ms and nothing popping up shouting "this is the problem". There was some packet loss later in the path, but again what I would put under an acceptable value of 3%.

I've sent through my findings on the above tests to WebAfrica and will update once I get a response.
Click to expand...

Are you in the Cape ?

So basically another round of conclusive evidence we have been sending them for the last 2 months using other ISP accounts and only now do they start taking this seriously.

Proof their network is ****ed in the evenings at least in the WC and EC.

I cannot stress enough how piss poor their problem solving efforts have been.
 
infection said:
Yes I am.

In any case, got a response today from Ricardo at WebAfrica.

Even after he gave me an Afrihost acount to test it on and there is no packet loss on the Afrihost account and the latency is fine, he still comes spitting out bull**** that it is the ADSL line.

Even after WebAfrica Helper here admitted to an issue on a segment on their network Ricardo still said there is no network fault.

This company has become a joke, their support are sub par and their employee's are either too lazy to work or too incompetent to read basic trace routes, pings and ping paths.
Click to expand...

Fine on Afrihost broken on Webafrica “but it’s your line” is the best Ricardo can come up with

Ok buddy. Keep drinking that koolaid.

Same story here. Got VOX and Afrihost flying in the evenings.

Sorry WA April fools has come and gone.
 
If you want to cry get pingplotter (free trial) and run the tests in the evenings to either news24.com or webafrica.co.za themselves.

You should see latency averages of around 1.5 to 2 SECONDS on that 3rd or 4th hop.

BUT but "it's your line" ! :D
 
I can't believe this is still going on. What a crazy fall from grace for WA, having been possibly the best ISP in the country, to having no idea how to be an ISP anymore. Its so stupid.

I've been a client for years now, but last night's experience at 19:30 (! we didnt even get to 8pm) was so shocking I feel I need to cancel. How do I go about doing that? Can I do it online? The thought of calling these clowns one more time kills me.
 
Markd said:
I can't believe this is still going on. What a crazy fall from grace for WA, having been possibly the best ISP in the country, to having no idea how to be an ISP anymore. Its so stupid.

I've been a client for years now, but last night's experience at 19:30 (! we didnt even get to 8pm) was so shocking I feel I need to cancel. How do I go about doing that? Can I do it online? The thought of calling these clowns one more time kills me.
Click to expand...

out of interest, how is it looking tonight?
 
I have no opened another ticket with WebAfrica, since I'm still experiencing high latencies in CS:GO in the evening - where as on VOX I have no packet loss, jitter or high latency issue!

Traceroute on WebAfrica at like 21:00:
Code: 
Tracing route to 155.133.238.162 over a maximum of 30 hops

  1    <1 ms    <1 ms    <1 ms  router.chris.local [192.168.11.1]
  2     *        *        *     Request timed out.
  3   106 ms   107 ms   106 ms  196.38.75.121
  4    98 ms   106 ms   115 ms  196.38.75.122
  5    89 ms    86 ms    80 ms  196.38.77.88
  6    70 ms    71 ms    74 ms  mi-za-cpt-p7-te0-7-0-0.ip.isnet.net [168.209.2.12]
  7    72 ms    70 ms    73 ms  168.209.100.101
  8    72 ms    72 ms    74 ms  pr2-pkl-xe-2-2-0.ip.isnet.net [168.209.1.179]
  9    82 ms    78 ms    84 ms  196-60-9-22.ixp.joburg [196.60.9.22]
 10    89 ms    85 ms    86 ms  192.168.210.55
 11    62 ms    63 ms    60 ms  155.133.238.162

Trace complete.

vs. Traceroute on VOX at like 21:02:
Code: 
Tracing route to 155.133.238.162 over a maximum of 30 hops

  1    <1 ms    <1 ms    <1 ms  router.chris.local [192.168.11.1]
  2     *        *        *     Request timed out.
  3    23 ms    23 ms    23 ms  telkom-upload-vlan-barrack-ipc.vox.co.za [41.193.121.9]
  4    23 ms    23 ms    22 ms  vox-upload-teraco-cpt-barrack-ipc.vox.co.za [41.193.121.10]
  5    25 ms    23 ms    24 ms  vox-b2b-pts-barrack-ipc-int.vox.co.za [209.203.1.41]
  6    25 ms    24 ms    24 ms  41.193.120.45
  7    37 ms    36 ms    35 ms  41.193.106.105
  8    35 ms    35 ms    36 ms  41.193.106.97
  9    43 ms    42 ms    42 ms  41.193.106.101
 10    41 ms    41 ms    41 ms  41.193.119.86
 11    41 ms    42 ms    41 ms  196-60-9-22.ixp.joburg [196.60.9.22]
 12    42 ms    41 ms    41 ms  192.168.210.55
 13    41 ms    41 ms    41 ms  155.133.238.162

Trace complete.

When I ran the traceroute on my Mikrotik router, I also had 6% packet loss on the 3rd hop - like a few others.

I will most definitely cancel my account if this continues for another 2 weeks.
 
Looks pretty decent with Vox, it appears to be where most are heading, packages are quite similar.
Just remember, if you cancel now they still gonna keep you in for the next calendar month. So rather put the cancellation notice in now, and you can always have that reversed should you change your mind.
For me the opportunity to resolve has passed ages ago. Logging tickets and poor service will just make me mad, far easier to just move.
 
There clearly is some sort of routing issue. It took WA about 3 weeks to respond to my initial ticket. Then i had to convince them several times that it wasnt my adsl line. They handed my ticket over to netops like a week ago, heard nothing since...
 
TheVovo said:
There clearly is some sort of routing issue. It took WA about 3 weeks to respond to my initial ticket. Then i had to convince them several times that it wasnt my adsl line. They handed my ticket over to netops like a week ago, heard nothing since...
Click to expand...

The only issue is that for the last 2 months in the Cape their network has been ****ed in the evenings, perhaps lack of capacity resulting in crippling latency or more nefarious causes.

So ****ed in fact the 90’s called and they want their 56k dial-up back.

Sad times.
 
Soooo I've noticed for the past 3 days that the evening latency problem they've been experiencing is getting a lot better.

I'm getting an average ping of 15-30 ms to telkom's saix.

Is the Phoenix finally rising from the ashes or am I the only lucky one ?
 
thewusman said:
The only issue is that for the last 2 months in the Cape their network has been ****ed in the evenings, perhaps lack of capacity resulting in crippling latency or more nefarious causes.

So ****ed in fact the 90’s called and they want their 56k dial-up back.

Sad times.
Click to expand...

Not saying you're not having a problem but my downloads and streaming are working fine on a 10 Mbps ADSL line. I'm in Melkbostrand.
 
TheVovo said:
There clearly is some sort of routing issue. It took WA about 3 weeks to respond to my initial ticket. Then i had to convince them several times that it wasnt my adsl line. They handed my ticket over to netops like a week ago, heard nothing since...
Click to expand...

am interested to know if you having an issue right now/tonight?
 
thewusman said:
The only issue is that for the last 2 months in the Cape their network has been ****ed in the evenings, perhaps lack of capacity resulting in crippling latency or more nefarious causes.

So ****ed in fact the 90’s called and they want their 56k dial-up back.

Sad times.
Click to expand...

could you check for me now if you still having that additional latency.
 
treebsd said:
could you check for me now if you still having that additional latency.
Click to expand...

yup still crap

C:\Users\Mark>ping utas.fut.ea.com

Pinging utas.fut.ea.com [134.213.37.197] with 32 bytes of data:
Reply from 134.213.37.197: bytes=32 time=169ms TTL=244
Reply from 134.213.37.197: bytes=32 time=176ms TTL=244
Reply from 134.213.37.197: bytes=32 time=168ms TTL=244
Reply from 134.213.37.197: bytes=32 time=169ms TTL=244

Ping statistics for 134.213.37.197:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 168ms, Maximum = 176ms, Average = 170ms

C:\Users\Mark>tracert utas.fut.ea.com

Tracing route to utas.fut.ea.com [134.213.37.197]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 31 ms 28 ms 25 ms 196.38.75.122
4 26 ms 25 ms 26 ms 196.38.77.88
5 295 ms 367 ms * mi-za-cpt-p7-te0-7-0-0.ip.isnet.net [168.209.2.12]
6 341 ms 348 ms 454 ms core1b-dock-ten-0-4-0-0sub8.ip.isnet.net [168.209.246.65]
7 177 ms 174 ms 171 ms 195.66.224.116
8 * * * Request timed out.
9 530 ms 454 ms 399 ms 134.213.97.213
10 536 ms 507 ms 522 ms coreb-core6.lon3.rackspace.net [164.177.137.25]
11 171 ms 173 ms 170 ms core6-eaaggr1b.lon3.rackspace.net [31.222.133.219]
12 169 ms 169 ms 169 ms 134.213.37.197

Trace complete.

this **** is just never going to end. I've given up hope on WA.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X