Every stop has a condition
When M3 refuses a step, it does so because a condition is not met: a quantity, a status, a setting or a missing record. The message on the screen names the symptom. The cause is the value that made the condition fail.
An example: a delivery that isn't released for picking
- 1The question: why is this delivery not released for picking?
- 2Read the record: the delivery and its four lines.
- 3Find the condition at this step: release for picking needs all lines fully allocated.
- 4Find the value: line 30 is allocated 6 of 10.
- 5The fix: allocate the open quantity on line 30, or release the delivery manually.
Reproducing a problem in a test environment only works if the test data matches production. Reading the real record, read-only, avoids that question altogether.
A checklist for any stuck record
- Which step should have happened, and which status is the record in now?
- Which screen, API call or job performs that step?
- Which conditions does that step check?
- Which value on this record, or on related master data, fails a condition?
- Is that value wrong, or is the process being used in a way it wasn't set up for?
When the message is all you have
If a user only has the message text, the first question is where that message can occur and what triggers it. "Location must be entered", for example, appears in receipts when no receiving location is given and no default is found. That points to the request, or to the default location setup.
Fix the cause, not only the record
Once the failing value is found, check whether it is a one-off or a pattern. If the same master data gap stops other records too, a quality rule will find all of them, and fixing the source prevents the next ticket.
How M3CloudSlam helpsDebugger →Ask why a record didn't go through. It reads the record from your M3 and names the condition and the value.