FNB online banking woes continue

Hi Steve can I move some of my services back to you?
 
and one users even posted a screenshot of the online banking outage.
Click to expand...

Whaaaaaaaaaaaaaaat, that **** is off the chain ! A screenshot nogal.

This user even has a funky toolbar in his IE browser ftw !
 
11 500 volts - They can expect a hugh electricity bill from JHB Muncipality....Lol
 
Lesson: Keep backup Data Centre isolated from primary,preferably by physical distance :)
 
FNB seem to have a lot of internet banking troubles... I remember a few years back they had these issues.
 
rorz0r said:
11 500 volts? From what?
Click to expand...

11 500 volts, I'm guessing that their DC has their own medium voltage power supply and step down transformer. Probably something tripped to prevent damage from the spike
 
cyberbob1979 said:
11500 Volts blew up some stuff? Eskom might have a little bill waiting for them...

Or someone stole the Ground Neutral Copper bar again :whistle:
Click to expand...

Michael Jordaan:
My 1st smile of the morning “@justjohan: See, even electrons want to switch to FNB.”
 
PsyWulf said:
Lesson: Keep backup Data Centre isolated from primary,preferably by physical distance :)
Click to expand...

I think they do. As far as I understand (not sure what FNB's setup is I can just speak from what I have heard from other financial services guys) is that they run a main system,with a another system in perfect sync right next to it. This is for high availability and not disaster recovery. Then they also run a separate system some distance away which is the back-up and disaster recovery site. The DR site is not in perfect sync and can be out of date by a couple of seconds because the main system does not wait for the DR site for completion as it does for the high-availability system.

So the power spike brought down the high-availability setup - not much you can do there, as this was obviously a fringe event. However, bringing a DR site online is not a small thing. It is built to recover operations within days of natural disaster and not to ensure high availability over a period of an hour or two.

Besides, you have a million little checks to run in cases like these. For example - with a power spike that couldn't be filtered you cannot be sure of the integrity of your persistent storage (I assume they run fiber channel into storage that use battery-backed RAM as buffering). They have checks and balances and backups and transaction logs to sort out the mess even in cases such as this BUT it affects availability.

Now I speak only from general knowledge. I am not sure if anybody on the forum have more information about FNB's specific setup?
 
Oh I understand having an HA replicated onsite,they said prod and backup taken out by the voltage spike though,thus my comment ;-) fortunate they had an off-site,too bad switching over has so many niggles but thats technology for you
 
PsyWulf said:
Oh I understand having an HA replicated onsite,they said prod and backup taken out by the voltage spike though,thus my comment ;-) fortunate they had an off-site,too bad switching over has so many niggles but thats technology for you
Click to expand...

Yup backup is NOT DR.
 
Johand said:
I think they do. As far as I understand (not sure what FNB's setup is I can just speak from what I have heard from other financial services guys) is that they run a main system,with a another system in perfect sync right next to it. This is for high availability and not disaster recovery. Then they also run a separate system some distance away which is the back-up and disaster recovery site. The DR site is not in perfect sync and can be out of date by a couple of seconds because the main system does not wait for the DR site for completion as it does for the high-availability system.

So the power spike brought down the high-availability setup - not much you can do there, as this was obviously a fringe event. However, bringing a DR site online is not a small thing. It is built to recover operations within days of natural disaster and not to ensure high availability over a period of an hour or two.

Besides, you have a million little checks to run in cases like these. For example - with a power spike that couldn't be filtered you cannot be sure of the integrity of your persistent storage (I assume they run fiber channel into storage that use battery-backed RAM as buffering). They have checks and balances and backups and transaction logs to sort out the mess even in cases such as this BUT it affects availability.

Now I speak only from general knowledge. I am not sure if anybody on the forum have more information about FNB's specific setup?
Click to expand...

They have 2 separate DC's as far as I'm aware, HA being available at both sites locally and then DR in place to switch between the 2 sites. I'm unsure what the difference in sync is between the 2 sites but I would imagine being a bank it's very little to nothing.
 
The site is up (sort of). I tried to create a payment recipient and make a payment, but I get an OTP error. The payment page does not offer a place to put in the OTP, nor was one sent. HUGE security hole. Shame on you FNB!!!!
 
erb said:
The site is up (sort of). I tried to create a payment recipient and make a payment, but I get an OTP error. The payment page does not offer a place to put in the OTP, nor was one sent. HUGE security hole. Shame on you FNB!!!!
Click to expand...

Security hole? You can't make the payment. What hole?
Systems are being moved to DR at the moment.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X