FUJ00226307
FUJ00226307
From: Jenkins Gareth Gl
Sent: Fri 02/10/2009 9:35:48 AM (UTC)
To: Chambers Anne 0, ___._GRO_
Subject: RE: Draft paper on Horizon Data Integrity
Anne,
I gather you've being trained this morning. Please can you give me a ring when you get a chance about this.
Regards
Gareth
Gareth Jenkins
Distinguished Engineer
Applications Architect
Royal Mail Group Account
SS Please consider the environment - do you really need to print this email?
This e-mail is only for the use of its intended recipient. Its contents are subject to a duty of confidence and may be privileged. Fujitsu Services
does not guarantee that this e-mail has not been intercepted and amended or that it is virus-free.
From: Chambers Anne O
Sent: 01 October 2009 18:19
To: Jenkins Gareth GI
Subject: RE: Draft paper on Horizon Data Integrity
Section 2
You mention incremental sequence numbers in the audit section at the end, but could mention it earlier too, to
make clear that each message has a unique identifier which stays with it when it is replicated.
Each individual message has a checksum.
Not quite sure what you should be saying about CRC read failures. We aren't currently checking old event logs
for these when doing audit retrievals, and I don't want us landed with even more checks to make. If there are
CRC errors, SMC normally raise a call and we trash the messagestore and let it rebuild - but probably don't want
to say that! But if we don't say, with they ask?
Event logs - more than 18 months but not sure if it is 7 years.
3.1.1, 3.1.2 The user will get an AP message warning them that the last session ended in error, but it only tells
them to check AP transactions, not others.
3.1.3.1 If banking recovery is not completed immediately after the counter is replaced, this is reported on the
banking reconciliation reports and followed up.
FUJ00226307
FUJ00226307
From: Jenkins Gareth GI
Sent: 01 October 2009 12:39
To: Hodgkinson Allan AJ; Worrell Jeremy; Chambers Anne O; Sweeting Jim; Bailey Chris
Cc: Smith Latoya; Burton John
Subject: Draft paper on Horizon Data Integrity
All,
Jeremy has asked me to produce a paper on Horizon Data Integrity to support POL in refuting claims by
Postmasters that Horizon is causing money to be lost.
I've put together an initial draft. I’ve ignored the “front bits” for now and am currently looking for comments
on the technical aspects and in particular the comments in yellow.
<< File: xxxSECxxxyyyy.HorizonDatalntegrity.doc >>
Also if anyone is aware of other material to feed in on this I would be grateful. I had a quick look through
PVCS and a search through the TED and found nothing useful there.
Note that this is NOT yet in a state to go to POL and once I’ve had feedback from all of you (and anyone else
you think is relevant), Jeremy needs to pass this through Commercial.
We have a conf call with POL 4pm on Friday, so I'd appreciate any feedback by lunch time Friday (which is
when I committed to providing this to Jeremy).
Jeremy: Do you have any thoughts as to where this should be lodged in Dimensions. I assume we need to
make this a formal document, but if not, I’m happy to remove the “front bits”.
Regards
Gareth
Gareth Jenkins
Distinguished Engineer
Applications Architect
Royal Mail Group Account
1 Please consider the environment - do you really need to print this email?
This e-mail is only for the use of its intended recipient. Its contents are subject to a duty of confidence and may be privileged.
Fujitsu Services does not guarantee that this e-mail has not been intercepted and amended or that it is virus-free.