POL00028428
POL00028428
many)
Shy
To: Simon Rilot feeaon spanke . cc: Catherine Fraser ———
Gail Morley
Sue Harding
Glands « Ruth Holleran
pong David Parnell
GRO : Janet Topham
Chris Young
From: Peter Jones Date : 7/10/98
I feel it necessary to write to you again as you appear unhappy with these
types of concerns being raised at the weekly testing meetings or in other
meetings.
Iknow we have a different perspective about what the status of the systems
should be by this stage of testing, and you may be getting a great deal of
benefit from the current phases in terms of being able to key transactions into
the system, but there is still a number of key areas which are vital and remain
unproven. We are leaving it very late to identify basic errors let alone more
complex one’s and then the required test phases in which to prove them as
being fixed.
My concerns are around the usual one’s (which I will expand later in this
document) I have raised many times before with you and are genuine
concerns in view of the date of 18/12 (end of MO & E2E testing) looming
ever closer.
Migration: You took the decision to remove migration from E2E testing, in
view of that, please can you confirm all the elements of reconciliation (as
documented in my previous letters to you) required to be proven during
migration are being covered. I am aware that the migration phase during
MOR2 did have some problems but due to extra or missing files we are
currently unable to give a factual view from our perspective. Iam aware
that no cash accounts balanced for the week involved, which may or may not
be due to migration problems.
Transactions: In view of your decision on migration I understand that the
detailed checking of outputs to detailed expected results would also be
carried out throughout MO to ensure the integrity of migration was fully
proven. We are having a great deal of difficulty in checking the files sent to
us as the files we have accepted differ from the scripts by upto 75% . Please
can you assure me that the differences are being fully identified to ensure that
itis not the system creating these differences . The large number of
differences between script and expected results are either being classed as
DRNAWord\lTIP\Testing\srilot01.doc 09/10/98
POL00028428
POL00028428
user input error or expected transaction records not reflecting the script .
Please can you inform me what improvement activities are being put in place
to ensure more accuracy in terms of carrying out the agreed scripts. As you
will be aware operator error was a major issue during DIT.
Rejected / Missing Files: The testing plans are all based on reconciliation
being carried out circa the day after or within a short time frame in order to
then reset the environments for the next run. We are rejecting a large number
of files as per the interface for a number of reasons some of which are
concerning and then the files are not being rectified and returned ina timely
manner . I would hope that the Pathway systems processes and procedures
are being tested to handle rejected files and would expect that they should be
dealt with on the same or next day. It appears to us that they only know there
are rejected files when we raise a PinICL. There are still files outstanding
from E2E first pass which finished some 3weeks ago. MOR2 is nearly
complete in terms of run days yet we are still waiting for files rejected on the
first day. We are currently awaiting 25 files covering cash accounts and
transactions for each of the MOR2 days. Which as you will appreciate means
we cannot reconcile any days including the migration situation. As per plan
we will have to commence preparation of the resetting the environments for
MOR 3 and will therefore be unable to accept these files after Friday 9 Oct.
Even if we were to get and accept these files we would have a major task in
trying to untangle them all and reconcile the information prior to the
commencement of E2E 2°4 pass and MOR 3. As occurred in previous phases
we also received files for dates pre data centre migration which should not be
possible.
Model office / Operational Running: During the planning phases everyone
was most insistent that the systems should run as if “live” which in TIP’s
case means we should receive files during a scheduled window of 8pm to
3am. I can understand and support this theory as it allows us to test that the
schedules operate correctly and allows us to replicate live running. As you
will be aware due to previous problems Pathway asked us to change our
operational schedules and extend the window to 3pm to 3am which we
obliged despite this a number of days are still arriving outside this extended
window eg: 6am which then are not processed until 3pm. Please can you
inform me when we should request the schedules to be reset to the proposed
live running and therefore test this area correctly.
Cash accounts: I am aware that in the grand scheme of things you do not
consider this area to be a major issue. As you know I consider it vital . We
have yet to see a valid cash account in either MO or E2E. Please can you
confirm that your team are identifying and logging all the errors on cash
accounts and when we can expect to have balanced accounts. It is not until
you have basic accounts balanced that you can start to look for the more
complex errors.
DRN:\WordllTIP\Testing\srilot01.doc 09/10/98
POL00028428
POL00028428
Reference Data: Another key area, to me, to be proven is that the systems
(Pathway and Tip) must remain in line when Ref Data updates are applied it
is unclear and mixed messages are being given about if Pathway are applying
the change drops as per the script. Examples have occurred during MOR2
which suggest not e.g. a product being sellable after it has expired , files
being rejected for invalid Org unit version number. Please can you inform me
if or when Pathway intend to fully apply the agreed change drops.
On another subject I feel that when a PinICL is raised but Pathway say it is
fixed and will be proven in a later test phase it should not be closed and if
required identified in the categories as perhaps “Fixed awaiting retest” my
reasons being the current process gives a distorted view on the number that
are being cleared / fixed and also would remove the chance for some not to
be re-tested.
If you do not intend to reply to the issues please can you tell me how and
with whom these should be raised and progressed.
Peter Jones
Tel Direct i
DRN:\Word\ITIP\Testing\srilot01.doc 09/10/98