People & context

SAP. ECC six,

point zero.

Nightly batch. There

is an extract that runs at two in the morning

and produces master so the...

That is the mechanism, and it is the only mechanism.

No, and there will not be one for this

project. Building one is a separate business so...

does not

He is back and says it is captured but not exposed. Getting it out means either an extract that nobody owns, or writing it back into SAP, which is a change request. I will not be raising a change request for something that is not in the current phase.

Answer on the historic files. OP_ID is present in the raw landing zone for every daily file since 14 February. It is not carried into the aggregated waste tables, which group by store and article. However it is present in the reconciliation extracts we produced in June and September for the data quality review, because those were row level. Two of those were emailed as attachments, once to Priya and once to a distribution list of nine people including three at Meridian. I can purge the landing zone. I cannot purge the attachments.

Hi Priya, Confirmed. OP_ID is dropped at the extract from the next run, which is Friday night. On files already delivered: they are in the raw landing zone and in the processed store. Kwame is checking whether the field made it into anything downstream. I will come back with a definitive answer rather than a reassuring one.

Nadia, Following Monday's call, confirming this as a condition rather than a request. The waste write-off extract currently includes OP_ID, a six digit operator identifier identifying the individual who scanned the item out. Examples from the November file: 447102 Marika Lindqvist, store 118 Vantaa, nights 448933 Osman Yildirim, store 118 Vantaa 451207 Heidi Salminen, store 094 Espoo This has been in every file since February. It must be excluded at source before the next extract runs. Please confirm the field is dropped and tell me what you have done with the files already delivered. Our works council will ask. I would rather answer them than be asked. If the answer to the second question is complicated, put it on the agenda for the January session rather than replying to this.

Extracts. Slower is acceptable, the other thing is not. Ivan, no offence intended. It is not about Meridian, it is that I cannot explain to my data protection officer why a third party had read access to everything.

It is not impossible, it is unowned, which in practice is the same thing this year. Proceed with your approximation.

Not yet. Half of them are on holiday. I will chase.

Hi, MHDHB is the total shelf life, not remaining. It is populated for part of the range. Remaining life at order time is not stored anywhere I know of. It would have to be derived from goods receipt date, and that is in the warehouse system rather than in SAP. I will ask the warehouse team.

File size check in place. Any file under 1KB now fails the job loudly.

Technical note so this is on the record properly. 6 March: middleware failure on DC-2. This is the known issue from the April 2024 assessment. 14 March: not the middleware. The job ran and moved three files, one of which was empty because an upstream export had not completed. Job reported success. These are two different failure modes. Please do not let them be merged into one line in the incident log, because the fixes are different. The second one is a file size check and I can do it this week.