[LINK] Bug leaves NetBank customers in the dark

[Scott Howard]

On Fri, Jul 3, 2009 at 11:15 PM, Bernard Robertson-Dunn
<brd at iimetro.com.au>wrote:

> My view is that the bank's IT development and support is fragmented.
> It's being done by people with little or no knowledge of what the bank
> is all about. A lot of camels will get developed and I suspect that
> NetBank is one.


It's been a few years since I've had anything to do with Commbank/EDS, but
for the 3+ years before that I worked very closely with many of the CommBank
systems and both CommBank and EDS people, and while what you're saying may
be true of some areas it's certainly not the case across the board.

Specifically for NetBank, every party involved with the systems - be they
CommBank, EDS or external vendors - had an excellent and indepth
understanding of the entire infrastructure, as well as their use.  Despite
only being a vendor, I could tell you the approximate volumes of
transactions across any day, which days would be busier than others (October
15th will be the busiest day for NetBank in the next few months, for
example), which areas within the bank would be impacted under certain
failure conditions, etc.

>From a development perspective, their CommSee project was an amazing success
- especially given the cross-company involvement required between CommBank,
EDS and others - and IMHO the main reason it worked so well due to how well
CommBank and EDS worked together.  A few years ago they made the decision to
have more cross-company project teams, even to the point of many CBA people
moving to EDS locations and vica versa - and within about a month of that
starting the improvements were obvious.  After seeing how well this worked
on CommSee I believe they have now rolled it out across other projects as
well.

NetBank itself was always a strange beast.  Although I believe things have
changed a bit in the past few years, the problem at the time was that
although it had issues and problems, NetBank actually ran amazingly well.
So well in fact that most people weren't willing to change the
infrastructure involved (even though it had a few fairly major flaws) for
risk of causing further problems, especially given that the system was well
enough designed that these flaws didn't normally result in customer-facing
issues.


> In the case of the bank's current troubles, the problem is probably an
> interface issue between the mid-range NetBank applications and those
> running on their mainframes.


Unless their design has gone backwards, that's not going to be the cause.
Without going into too much detail, the front-end systems are suitably
isolated from the backend systems to avoid any such problems.


http://business.smh.com.au/business/bug-leaves-netbank-customers-in-the-dark-20090703-d7qz.html
>
> A Sydney IT veteran with experience working at the bank and IT services
> company EDS said: "I would be very concerned that [the bank has] an
> issue like this that lasts so long.
>
> "Either [the bank's] IT department wasn't prepared for the issue it's
> facing, or their disaster recovery plan hasn't worked," the IT expert,
> who asked not to be named, said.
>
> "They should have highly detailed disaster recovery plans and that
> should spell out how long it would take to recover from anything," he said.


I have no details on the current problems beyond what I've read in the
press, but it's fairly clearly not a "disaster" in the sense that one would
revert to a "disaster recovery plan". I'm not surprised the "IT expert"
asked not to be named when he's making comments with as little value as
those...


Scott.