Who moves your invoice out of the Peppol Waiting Queue, and how often that actually happens — transcript
Play this as a presentation — one slide per step, with the same narration. Every word of every step is on this page.
This presentation is for whoever has watched an invoice sit in the Peppol Waiting Queue for a day and wants to know what is supposed to move it. In about nine minutes you will be able to name the job, say whether your tenant has it, explain why pressing Process did not help, and ask your BigLedger contact for the one thing that would.
Step 1 — Name the one job that moves it
After this step you will know exactly which job to ask about. One scheduled job turns Waiting Queue rows into Peppol documents: the waiting-queue-to-document sweep. Each time it runs it builds the document, looks up your company’s default participant identifier and the customer’s, checks the result against the Malaysian Peppol profile, and — if it passes — hands the document straight to the sender, which transmits immediately. The sender needs no clock of its own. So on the sending side there is exactly one thing that has to be scheduled, and nothing on any screen tells you whether it is. That job is not in the standard set a new tenant is given; it exists on your tenant only if somebody at BigLedger added it by hand.
Reference: My Peppol Admin Applet — Before you can use it
Step 2 — Ask how many tenants actually have it
After this step you will stop assuming the queue empties itself. When we counted, the sweep had a clock on a handful of BigLedger’s tenants — seven of ninety — against about sixty-five for the equivalent LHDN job. On every other tenant, turning Peppol on does exactly what the previous presentation described and then nothing more: rows are written at finalise, the Waiting Queue grows, and no document is ever built. There is no error, no e-mail, and no screen that says why, because from the system’s point of view nothing has failed. Across the estate the queues held hundreds of thousands of rows, and most of them were between one and six months old. If your queue only ever grows, that is the first thing to establish.
Reference: My Peppol Admin Applet — The journey after the Waiting Queue, stop by stop
Step 3 — Ask how often it fires and how many rows it takes
After this step you will be able to do the arithmetic for your own company. On six of the seven tenants that have the sweep, its clock fires once a day, at midnight. Each run takes ten rows of each of the two kinds — the finalise-time rows and the after-LHDN rows — so twenty documents a day, and it takes them in whatever order the database returns, not oldest first. Suppose GadgetSphere Distribution finalises forty invoices a day to Peppol customers. The queue grows by twenty a day for ever, and any particular invoice may wait indefinitely. A daily clock and a cap of ten were reasonable for a pilot; they are the reason a live queue never comes down.
Reference: My Peppol Admin Applet — The journey after the Waiting Queue, stop by stop
Step 4 — Know what the Process button really does
After this step you will know why Process did not move your row. The Waiting Queue has a Process button on the listing and another on the row’s detail page. Both collect whatever you selected, and both then throw the selection away: the call they make runs one pass of the sweep — ten rows of each kind, in database order. No row can be selected anyway, because the checkbox is drawn only when that blank Status column reads a value it never has. On an empty test queue Process appears to work perfectly, which is why setup guides tell you to press it. On a queue of thousands it moves twenty documents that are probably not yours. Export and Withdraw on that screen do nothing at all.
Reference: My Peppol Admin Applet — The journey after the Waiting Queue, stop by stop
Step 5 — Ask for the right thing
After this step you will send one message instead of three. Nothing on your screens shows a clock, so the question goes to your BigLedger contact, and it is three questions in one: does my tenant have the waiting-queue-to-document sweep scheduled, what is its schedule, and can the cap be raised or the schedule shortened to match the volume we send. Put it in the same message as the request for the applet itself, before the first Peppol customer goes live — a queue that starts empty and is drained daily looks healthy for about a week. And when someone tells you it is scheduled, ask for the expression; once a day is what most tenants that have it are running.
Reference: My Peppol Admin Applet — Before you can use it
Check yourself
Three to five questions on what you just heard. Every correct answer links to the page that makes it correct, so you can check the source, not just the mark.
Answer key
- The waiting-queue-to-document sweep has no clock on your tenant — true of most tenants — My Peppol Admin Applet — The journey after the Waiting Queue, stop by stop
- Ten of each of the two kinds — twenty — in database order — My Peppol Admin Applet — The journey after the Waiting Queue, stop by stop
- One pass of the sweep for twenty rows of the queue's choosing — the selection is discarded — My Peppol Admin Applet — The journey after the Waiting Queue, stop by stop
- The waiting-queue-to-document sweep; the sender is run by it immediately — My Peppol Admin Applet — Before you can use it
Next: Reading To Peppol AP, Queue and History — what each Peppol screen is evidence of · Back to the series · Play this as a presentation