POL00029605
POL00029605
Message
From: Steve Allchorn
on behalf of Steve Alichorn <-
Sent:
To:
ce:
Subject:
Lesley
I'11 hopefully get most of the loose ends on this sorted out by COP Monday.
Quite correct. As James Brett has described - the test area in the basement in 2008 was a physical build
of servers that represented the live Horizon system at the time as closely as possible but were
completely separate to that live production system. I'l] further confirm with JB and FS. In terms of what
Mr Rudkin allegedly saw and heard on his visit, i've asked Second Sight to establish Mr Rudkin's level of
understanding and knowledge of the test environment prior to him allegedly visiting it. The view I'm
getting is that he knew very little about it. It is therefore not inconceivable that he made certain
assumptions regarding his experiences.
I'll also button down what was in situ on the 2nd floor.
I'll also speak with Martin Rolfe (Test Manager) on Monday. He is probably the last opportunity in
attempting to correlate the test transaction visit set up by the Bureau Pre-Order Automation project on
the 19/8/2008 and Mr Rudkin's alleged visit_on that same day, as Martin was the test manager on the
project and appears to have been in Bracknell on that date.
steve
-----original Message-----
From: Lesley J Sewel
Sent: 14 June 2013 17:38
To: Steve Allchorn
Cc: Alwen Lyons; Susan Crichton; Gina Gould
Subject: Re: In Confidence - Second Sight Line of Enquiry
Thx steve
I've had a call with FJ this afternoon.
Can you please track these actions with FJ.
I have asked them what was on the second floor. Initial view is that it was the project team for HNGx -
no access to live Horizon. Fujitsu are to confirm this.
But more interesting - out of this email I've tested Fujitsu on the infrastructure. This looks as if the
test system infrastructure wasn't even in the data centre - it was physically separate (in the basement)
from the production system with no means of accessing the live data centre at all! This is material,
Fujitsu confirmed on the call that the test infrastructure was NOT in the datacentre, however I have
asked James to confirm this and to validate that even with the right log-on and access rights it would
not be possible to access the production system. If this is the case then SR 5 has no foundation.
Steve: can you also test the hypothesis with James B.
I've also asked James to contact Simon about comms - The two comms teams need to be connected as we build
our comms strategy and do the pre-mortem.
Have a good weekend.
L
Gina: please print for my Ss file.
sent from my iPad
On 14 Jun 2013, at 11:35, "Steve Allchorn" <j
Both
James's initial summary response to my line of enquiry.
Lesley (in particular from a technical perspective) if you have any further questions arising from
>
>
>
>
>
James's email could you let me know asap and I'll establish answers.
POL-0026087
POL00029605
POL00029605
steve
From: James Brett
Sent: 13 june 2013 08:17
To: Steve Allchorn
Subject: RE: In Confidence - Second Sight Line of Enquiry
steve,
Vv VV VY VV VV¥
In 2008, we were in test preparation mode for HNG, and in parallel supporting testing of the existing
Horizon estate. In the basement at that time there would have been 4 separate test environments
>
>* BTCl & BTC3 : These two test environments were used for functional testing of changes being
made to Horizon at that time, eg introduction of MoneyGram
>
>* V&I: The Horizon Volume test environment.
>
>* REL : The Horizon release test environment (where deployment of software packages were
tested)
>
> Along with these environments, preparation activities were underway in the basement to build a volume
and release environment for HNG, but these environment would not have been in a working state at time of
the statement.
>
> POL had access to the functional test environment, and I've asked people in my team around at the time
and no-one can recall any external visits, or Mr Rudkin specifically
>
> Fujitsu controlled the Volume and Release environments. But it would be unlikely that POL visitors
would be shown access to these as their focus wasn't really counter functionality; they were designed to
provide performance capability and software deployment to the data centre platforms rather than counter.
>
> The key point here is the word test environment. In Horizon days, the live Horizon Data Centre was
dual-located in Wigan and Bootle, access to this site was strictly controlled, and you would not have
been able to hook up a PC and interfere with transaction databases. To create a test environment in
Horizon days, we had to physically build a set of servers that represented the live configuration in
Wigan/Bootle. These servers were hosted in the Basement in Bracknell, along with test counters to connect
to them. Access to the test environments is (and was then) controlled via secure rooms, and user logon
authentication. However, as it is a test environment, there would have been terminals where interrogation
of databases would have been possible. But, this would have been interrogation of the test databases, as
there was complete physical separation between test and live.
>
> HNG is little different in that the live data centre is hosted in two sites in Belfast (IRE11 and
IRE19). Production uses IRE11 and the various test environments use IRE19. If we need to do a disaster
recovery, the test environments are quickly backed up to disk, and production is switched to IRE19. There
is network and disc partitioning to ensure live and test cannot interact. This is a moot point though, as
in 2008, the HNG environments in the basement were a long way from being operational.
>
> Tester's rarely get to see daylight, and can be creative at times. The basement may have been given
affectionate names by the testers over time, but I have never heard of the basement referred to as
"covert operations’.
>
> It is fairly common practice to invite external stakeholders to review new system operation, and I do
recall work at the time to validate the new HNG UI. In 2008, HNG development had just commenced and the
approach to HNG UI testing was two use a small group of superusers (i.e. brought in UI specialists) to
inform at the early build stages. Later, in 2009, when HNG functionality had matured enough, a formal UI
test phase was held, attended by true end users. Unfortunately there are no records as to who attended
this UI test, but I can confirm it was held in 2009, and it took place on the first floor rather than
basement.
>
> For perspective, there is live access available at Bracknell, and there would have been in Horizon days
too. However, this access is available only to Fujitsu's SSC (System Support Centre) team, who provide
expert support to helpdesk staff. They are based on the 6th floor, which is the most secure floor in
Bracknell. Visitors are by appointment only and are not allowed to be unattended. The ssc team follow
strict protocols relating to access and interrogation of live data, and their access is logged and
auditable. There is no access to SSC systems from the basement.
>
> AS I understand it, Horizon audit archives are held for a period of 7 years. Wouldn't it be possible to
retrieve the live Horizon audit for the date in question and look for the transaction Mr Rudkin saw? Even
if it were subsequently removed, the insertion and removal would have been logged for audit purposes.
Would probably need more information to search the live audit log though....
Regards
James
James Brett I Senior Test Manager
vvvvvyvy
POL-0026087
POL00029605
POL00029605
BRAOL, Bracknell RG12 8SN
-bret
{Footer 5]
<image001. png>
<image002. jpg>
POL-0026087