MWEB + THE WEEKEND = FAIL

I have been doing some calculations. At R539 per month, it works out to R18 per day. In the last two months I have lost about 7 days of connectivity, so that is R126.

So I would like to put a challenge to MWeb:
If you truly feel 'sorry' for the outages, are you going to refund your customers for the downtime?

Your competition, Telkom, did it, if you have reference numbers, which I have:
CAS-209297-7C7KFC and CAS-209298-4XMVCQ


That wouldn't work 2 well as half of us can't evan contact the call centre to get a refrence number lol
 
As a Tier 1 ISP, MWeb should have a backup system in place. I'm not paying Neotel or Seacom for Internet access.

+1, though what I think you're implying as a "backup system" (namely a 1:1 failover) would never happen, purely for cost reasons.
 
You are probably a 384 customer, I am quad 4Mb uncapped subscriber. I was phoned by them this afternoon.

No, I'm a 4MB uncapped customer. And why should it make a difference. A customer is a customer. Surely it is not to difficult to send an email out to all customers - they manage to do it with their bills and when they want to tell us "what's new"
 
Ya, and its also costing me less then three rand a gig...

I dont constantly rape the system for 200+ GB per month, I actually use this service for work but I need uncapped as I frequently upload and download big sets of data. And yes I was planing on working on a Friday night, some of us have too...
 
+1, though what I think you're implying as a "backup system" (namely a 1:1 failover) would never happen, purely for cost reasons.

The problem at the moment seems to be that MWeb is routing all traffic via SAT-3, but there is zero throughput (at least for users on this forum). I'm wondering if there isn't a deeper problem, since last time MWeb managed to route some traffic via SAT-3 last time there was a 'planned' outage on SEA-ME-WE-4. Maybe their 'shaping' only works on the SEACOM circuit, and can't handle the load?
 
Mervyn Goliath I am in direct and ongoing contact with the most senior executive at NEOTEL and the latest feedback is as Rudi has indicated ealier ... they have located the physical break in the fiber and have dispatched a repair team ... the prognosis is 3 to 4 hours to full service restoration ... BUT they are also in the process o...f rerouting traffic over an unaffected circuit ... i'm hoping that we will have the service restored at least partially before the actual break is restored.

From facebook
 
How does a fibre optic cable just break at ten past six on a Friday evening?

And where, oh where *weep* is the damned *weep* redundancy? Not even a *slow* connection internationally at the moment.
 
The problem at the moment seems to be that MWeb is routing all traffic via SAT-3, but there is zero throughput (at least for users on this forum). I'm wondering if there isn't a deeper problem, since last time MWeb managed to route some traffic via SAT-3 last time there was a 'planned' outage on SEA-ME-WE-4. Maybe their 'shaping' only works on the SEACOM circuit, and can't handle the load?

Does MWeb automatically and always have immediate access to SAT-3, or is it only by prior arrangement (as with the SEACOM outage last month)?
 
The problem at the moment seems to be that MWeb is routing all traffic via SAT-3, but there is zero throughput (at least for users on this forum). I'm wondering if there isn't a deeper problem, since last time MWeb managed to route some traffic via SAT-3 last time there was a 'planned' outage on SEA-ME-WE-4. Maybe their 'shaping' only works on the SEACOM circuit, and can't handle the load?

Hm, their shaping *should* be connection agnostic (though would likely need tweaking dependent on the performance capabilities of the backup route), who knows, I suppose anything is possible. :)
 
How does a fibre optic cable just break at ten past six on a Friday evening?

And where, oh where *weep* is the damned *weep* redundancy? Not even a *slow* connection internationally at the moment.

<tinfoilhat>
MWeb sees they can't pay their bills at the end of the month. They decide to introduce severe shaping and feign instability to reduce overall bandwidth usage, effectively not using more than their purchased capacity. This in turns saves their business while screwing their customers in the process.​
</tinfoilhat>
 
Last edited:
I also would love to know how a fibre cable, which is meant to be enclosed in a highly protective casing just "breaks". If there was an earthquake, it would make more sense..
For some reason Facebook still works.. interesting.
 
Does MWeb automatically and always have immediate access to SAT-3, or is it only by prior arrangement (as with the SEACOM outage last month)?

AFAIK MWeb are always routing part of their traffic via SAT-3 (this is what tracert has been saying for several weeks).

When their SEACOM routing fails they switch all the traffic to SAT-3, but this seems to have failed badly tonight, for some reason (which is why I suspect a more complex problem).

Edit: I'm still puzzled that pings can get to London, but no further...

Hm, their shaping *should* be connection agnostic (though would likely need tweaking dependent on the performance capabilities of the backup route), who knows, I suppose anything is possible. :)

+1. The shaping might only work when high-priority traffic demand is less than 30% of capacity. If there is too much demand, everything just seems to collapse for low-priority (consumer) users.
 
Last edited:
I also would love to know how a fibre cable, which is meant to be enclosed in a highly protective casing just "breaks". If there was an earthquake, it would make more sense..
For some reason Facebook still works.. interesting.

I can read but I cant log into facebook...
 
Top
Sign up to the MyBroadband newsletter
X