South Africa’s biggest forum. Discuss, discover, and connect with thousands of members.
garp said:Never mind though - remember the last nationwide Vodacom outage in March? Nobody could make calls for most of the day, yet by the following day Vodacom had issued so much spin that the mainstream papers were reporting that Vodacom had experienced a "minor" problem that only affected 5% of its subscribers.
Heh, well if on average only 5% of subscribers use the network over a several hour period, then only those 5% would have been "affected", haha! (Spin spin spin ..)garp said:Vodacom had experienced a "minor" problem that only affected 5% of its subscribers.
Gambit said:Vodacom should implement some fault tolerance.
But last time they didn't have anyone to blame it on, except themselvesrpm said:The previous outage was shrouded in mystery and although they had a good idea of what the problem was they did not want to divulge any information.
Server redundancy is very good and is a must for mission critical systems but when all hardware is working 100% and software screws up then most of the time it does not help. From their execuse it sounds to me like they need error handling in their software.tibby.dude said:Actually they do ... they use HP Tandem NonStop servers for their call control.
These specialized computers are designed for fault tolerance and 24x7 operation and is also used by most of the local banks and the S.A Reserve Bank to ensure that your money goes uninterrupted to where it is supposed to every millisecond of the day.
Typically uptimes are measured in decades.
http://h20223.www2.hp.com/nonstopcomputing/cache/76385-0-0-0-121.aspx