Sources fail sometimes — a shop goes down for maintenance, an address changes, a certificate expires. Here is exactly what Yaplet does about it, so nothing about your catalogue is a surprise.
The statuses
| Status | What it means |
|---|---|
| PENDING | Created, but nothing has been read from it yet. |
| PROCESSING | A read or a push is running right now. |
| ACTIVE | The last read succeeded. |
| WARNING | The last attempt failed. We keep retrying every hour. |
| INACTIVE | Three failures in a row. Scheduled reading has stopped. |
Hover the information icon beside the status for the reason in plain English, including the exact error where there is one.
What happens, in order
Within a single read we try three times, waiting between attempts — but only when the problem looks temporary, like a timeout or an error on your server. A flat rejection such as "not found" fails straight away.
- First and second failure in a row — status goes to WARNING. We keep retrying every hour. Nobody is told.
- Third failure in a row — status goes to INACTIVE and scheduled reading stops. Your organisation's owner gets a notification and an email.
The email says what happened, why, how to fix it, and what happens if you do nothing, with a button straight to the Products page. You get one email per hour at most, however many sources are involved, so it cannot nag you.
What happens to your products
That depends on the choice you made on the source under If the source keeps failing, and it only takes effect at the third failure:
- Keep serving the last products we read — your AI carries on recommending what it already knows. Safe for availability, but the prices and stock it quotes are from the last successful read.
- Hide the products until it works again — the products disappear from your AI and from product cards, and all come back on the next successful read.
Products you switched off by hand are never switched back on by this.
Starting it again
Fix the problem at your end, then press Fetch now (the circular arrow) on the source's row. A successful manual read clears the failures, sets the source back to ACTIVE, and brings back any products that were hidden.
Only Fetch now restarts a stopped source. If your shop pushes products to us, those pushes keep working and keep your catalogue current even while the source is INACTIVE — but they do not restart the schedule.
There is a five-minute cooldown per source. Inside it, the button tells you how long is left.
Things that are not failures
- A Fetch now you ran yourself. If it fails it sets WARNING, but it never counts towards the three strikes and never emails anyone.
- A push from your shop that fails. It shows in the run report only.
- An empty response. A read that comes back with no products at all is treated as a failure, not as "delete everything" — so a broken feed can never silently empty your catalogue.
Reading the run report
The chart icon on the row opens the report from the most recent run: how many items arrived, how many products were added, updated, left unchanged or removed, and anything skipped as invalid with the reason.
Only the latest run is kept for each source, so read it before you trigger another one.
Common causes
- The address changed or now redirects somewhere we cannot follow. Edit the source.
- A Product API address that is not
https. These are https-only, on the first request and on every redirect. - Your catalogue outgrew a limit — 20,000 products per source, or 20,000 across your whole organisation. Write to support to have the organisation limit raised.
- The response is too big (over 50 MB) or too slow (over 30 seconds per attempt).
- You are on the free plan and your content allowance is full. Products share that allowance with your articles, pages and files.
See also: connecting your products.