Vodacom denies sabotaged to blame for network outage

gibberish

MTN might have some splainin to do.
Can they satisfactorily account for their whereabouts at the time of the outage? Hah! I thought not. It's like nine-eleven all over again! Maybe it was the same guy dug up their cable as tossed that bolt into the Koeberg nuclear reactor? Hmmm I wonder what the DA could have against Vodacom? Or is he just a versatile freelancer? The plot thickens...clear reactor? Hmmm I wonder what the DA could have against Vodacom? Or is he just a versatile freelancer? The plot thickens...
 
Last edited:
/goes of too read the article first ;)

:edit
So it's not sabotage or rica, but failure at two points:

What exactly happened at those two points will be made clear later then ?
Then obviously can this be avoided in the future by something like a fall-over ?
 
Last edited:
Pr⊕phet said:
/goes of too read the article first ;)

:edit
So it's not sabotage or rica, but failure at two points:

What exactly happened at those two points will be made clear later then ?
Then obviously can this be avoided in the future by something like a fall-over ?
Click to expand...

Maybe, depending on the type and location of the failures. As I posted somewhere else, all networks experience failures on a daily basis, yet these events where large a number of subs are affected are very infrequent. You can probably count them on one hand over the last decade.

The normal failures are handled by the redundancy inherent in the network but what happens if both the primary and backup system fail?

I'm busy building a new 12TB NAS (O/T question: FreeNAS or UnRaid?), so let me give a relevant example.

We build RAID systems to ensure redundancy with RAID5 most popular as it allows you to drop any one disk and not compromise the data. I've had a few disk failures over the years but every time I could replace the failed disk before the next one fails.

What will happen if you have 2 disks that fail at the same time or within a very time period of each other? Total, catastrophic loss of data.

Even if you scale it up to the size of the network, you always have some level of exposure, irrespective of how much redundancy you build in.
 
jannievanzyl said:
What will happen if you have 2 disks that fail at the same time or within a very time period of each other? Total, catastrophic loss of data.
Even if you scale it up to the size of the network, you always have some level of exposure, irrespective of how much redundancy you build in.
Click to expand...

An analogy is like a bucket of water, you can only carry it that far. For 6 makulus you can have hot standby disks, mirrored raids, online backup, offline backup and geographic replication.
 
r00igev@@r said:
An analogy is like a bucket of water, you can only carry it that far. For 6 makulus you can have hot standby disks, mirrored raids, online backup, offline backup and geographic replication.
Click to expand...

Yes, let's just have 6 of everything. Then you can complain how everything is so expensive because they need to fund and maintain it all. Heck why dont they just build another Vodaworld to have on standby incase the other one burns down

Seriously, get over it. No matter how much redundancy you put in, how many precautions you put in place, there is always an small chance that that can still fail. And it did.
 
Last edited:
jannievanzyl said:
Maybe, depending on the type and location of the failures. As I posted somewhere else, all networks experience failures on a daily basis, yet these events where large a number of subs are affected are very infrequent. You can probably count them on one hand over the last decade.

The normal failures are handled by the redundancy inherent in the network but what happens if both the primary and backup system fail?

I'm busy building a new 12TB NAS (O/T question: FreeNAS or UnRaid?), so let me give a relevant example.

We build RAID systems to ensure redundancy with RAID5 most popular as it allows you to drop any one disk and not compromise the data. I've had a few disk failures over the years but every time I could replace the failed disk before the next one fails.

What will happen if you have 2 disks that fail at the same time or within a very time period of each other? Total, catastrophic loss of data.

Even if you scale it up to the size of the network, you always have some level of exposure, irrespective of how much redundancy you build in.
Click to expand...

I know it's still very expensive, but are you using SSDs for your NAS, or SATA/SCSI drives?

My feeling is that SSDs will last much longer, cos there's no moving parts inside + the much improved read/write speeds.
 
kingrob said:
I know it's still very expensive, but are you using SSDs for your NAS, or SATA/SCSI drives?

My feeling is that SSDs will last much longer, cos there's no moving parts inside + the much improved read/write speeds.
Click to expand...

2TB SATA disks. I would shudder to think what a 2TB SSD would cost. Do you even get them in that size?

The system is predominantly used to store media, so it's write once, read occasionally. Thus I'm not too worried about outright performance, especially on the write cycle. I discovered a system called UnRaid that seems better suited for this type of occasional use, massive media store than a normal RAID5 system. It's even more attractive if you've got a bunch of different size disks laying around and you want to make a NAS with redundancy from them. This is something you can't do with normal RAID systems where disks must be the same size.

In terms of pricing, I was surprised how the costs came down. A 12TB system (10TB - RAID) came to about R8K and that includes a fancy chassis. Compared to my first 1GB drive (R20K - circa 1990) the price/capacity ratio increased by a factor 30 000!
 
Confused

"Vodacom claims sabotage to its network

By: I-Net Bridge
2011-07-01 15:41
Johannesburg - Vodacom Group [JSE:VOD] CEO Pieter Uys on Friday claimed that the mobile operator had been a victim of sabotage on its fibre networks.

In an interview with regional radio station Jacaranda 94.2, Uys said that at least one of the recent outages to hit the group was caused by sabotage.":confused:
 
r00igev@@r said:
SSDs have a lower lifetime in terms of access count.
Click to expand...

But they're improving the lifetime.

As to RAID - we also do use RAID5/RAID6, but also do backup to tape. So if the worst do happen, then we will still have our data.

It's horrible... the massive amounts of data one need to back up... :sick:
 
The_Librarian said:
But they're improving the lifetime.

As to RAID - we also do use RAID5/RAID6, but also do backup to tape. So if the worst do happen, then we will still have our data.

It's horrible... the massive amounts of data one need to back up... :sick:
Click to expand...

Yeah, and often the tapes can write the data faster than what it can be read off the hard disk (especially if there is an application causing latency in between)
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X