Techies, what's with the RBS computer?

kittiej

Do any of you lovely techie people have any idea what's up with RBS/NW computers?

If there is such a problem then couldn't they have rolled back their system to before the glitch?

Seems quite odd to me, is it going on names/ acc nos since some can access funds and others not. Why not a blanket stop on all accounts?

I just wondered.

Thanks

Dreamnine

My guess is that they've been hacked...

Of course they're not going to say,"Sorry, but we've been hacked - we hope to be back up soon."

My RBS debit card wouldn't work in Morrisons today, most annoying.

bazster

Without being directly involved it's impossible to say, but most likely they "fixed" or "upgraded" something and didn't test it properly. The law of unintended consequences is a merciless enemy in the world of software.

With systems of the size (and probable antiquity) of theirs, restoring to an earlier point would most likely be a very long-winded job, leading the geeks instead to try to patch it back together as it stands. And by the time they've been hacking it about for a few hours, they will most likely have rendered it impossible to go back, leaving them no option but to keep laboriously bashing and bending it back into shape however long it takes.

fwor

kittiej wrote: »

Do any of you lovely techie people have any idea what's up with RBS/NW computers?

It would be nice to think that some of the techies here had the skill to be able to diagnose a problem like that from outside the system, but the fact is that the computer systems of any mainstream bank are going to be ~incredibly~ complex. Multiple systems, interconnected in complicated ways.

If I had to guess I'd go for the least sinister and most mundane of likely causes: two ancient, legacy system components that were never originally designed to be interconnected had stopped talking to one another as expected.

You're not likely to get anything other than guesswork on any public forum - anyone with direct intimate knowledge of how their systems work would face immediate dismissal for revealing any of that knowledge here...

[Deleted User]

Right, i deal with hospital systems. They're also big interconnected things with ancient and modern software/hardware intertwined. This will be considered a P1 issue in our world - a complete failure of the system (from what i've read, it's effected ATMs, online access, telephone and branch access).

Now, if they've not fixed it within a few hours, you can consider this is likely to be hardware/data/security issue. I know integration, and this really isn't an integration issue (you wouldn't get calamitous issues like this unless there was some badly tested large software upgrade going on). Software also doesn't normally break this amazingly.

So - security? Maybe some of their core databases were compromised; either physically or electronically. It's not inconceivable to have a data center fail and cause this. However, you'd expect them to have high-availability & DR procedures.

My guess is they've had some kind of physical disaster, tried to use their DR procedures and they've not worked properly. Someone needs to look into their non-functional requirements a bit more. There's also a possibly their internal network has been compromised and they've had to kill the system to stop a virus propagating (fairly unlikely since it'll be a closed and tightly regulated network)

But... you never know. I once had to spend 2 hours trying to fix a problem that absolutely killed a multi-million point NHS system (thankfully not live at the time) just because the wrong windows registry setting had been entered.

Corncrake2

It was certainly a big un !
NatWest (dunno about the rest of them) went down at 2am Wed morning, contrary to the news reporting problems "today".
BBC news reported this (Frid) morning that NatWest/RBS said tech problem solved, so ,,
Went to my local branch just now, guess what, queue out the door and manager very sorry but the sudden demand crashed them again ! Might be Mon. before it is all sorted (they hope )

ChiefGrasscutter

Badly tested major software upgrade sounds exactly right.
The modern pressures to get the software improvements out on time, under budget leads to sub-standard coding and testing that is likely to be 'abbreviated'.

agrinnall

ChiefGrasscutter wrote: »

Badly tested major software upgrade sounds exactly right.
The modern pressures to get the software improvements out on time, under budget leads to sub-standard coding and testing that is likely to be 'abbreviated'.

Sadly this is almost certainly correct, it's yet another thing that makes me glad I'm no longer there. I don't get paid as much working in academia but it's very unlikely that my university job could cause national financial chaos!

Nilrem

ringo_24601 wrote: »

My guess is they've had some kind of physical disaster, tried to use their DR procedures and they've not worked properly. Someone needs to look into their non-functional requirements a bit more. There's also a possibly their internal network has been compromised and they've had to kill the system to stop a virus propagating (fairly unlikely since it'll be a closed and tightly regulated network)
.

That sounds quite likely, especially if they've got a contract with some of the big "computer"companies for the DR plan.
IIIRC there is a bit of a stink at the moment about how IBM in the US has been handling it's contracts for similar things, basically not having anything like the staff or resources to actually cope with the contracts they've signed (IE not having the actual personal anywhere near where they would be needed - i think it's been suggested they would need to fly in staff from outside the US in some cases as they'd outsourced so much).

From what I can remember a lot of the big banks were reportedly still using equipment/software that was 10-15 years old in places not to long ago because getting a replacement system up and running for those bits, would break compatibility with other systems* (so replacing system 1 would require replacing system 2, and updating 3,4 and 5).

Whatever happened was obviously fairly major, and unexpected, as IIRC the banks normally do any planned works over the weekend specifically because it gives them time to fix unexpected bugs before the working week, and outages normally don't affect most services as seems to be the case here (normally it'll be cash machines, or online banking etc).



*In much the same way as some manufacturing companies have been quietly buying up legacy hardware because they might have a bespoke bit of software that won't run on modern PC's, controlling a machine that costs millions to replace (IIRC some looms for carpets etc still require floppies to load patterns on them, and specific software for the patterns). I think NASA was stockpiling 8086/8088 chips and the like just before Intel stopped making them, as various bits of their hardware were built around them.

Notmyrealname

It might not even be the computers at all. Badly configured routers have brought down large networks before.

Dreamnine

It says here that a software update is responsible: http://www.scotsman.com/news/uk/rbs-unable-to-say-when-banking-problem-will-be-fixed-1-2370937