Understanding error messages
A transfer has failed, and a message appears in the traces. This page explains how such a message is structured, at which two places errors occur at all, and which cases sit behind the frequent messages. The goal is that you can place the cause yourself and need support only for the cases that genuinely belong with us.
Where the traces live and how to get to a failed row is covered under traces (reference). The route from the symptom to the row is shown under product does not appear in the shop.
Prerequisites
- A failed transfer in the traces, or a task that ends with an error.
Two places, two kinds of errors
Before interpreting a message it pays to look at where it stands, because the place says how large the damage is:
- On the individual transfer, in the traces. The error concerns exactly this record. The task as a whole may have succeeded, and the remaining records went through normally. That is the frequent case, and the entry stays until the record is through. Why it is built that way is under traces.
- On the task itself, in the task list. Then the run as a whole did not run, or did not run to the end, typically because the connected system was unreachable or because TRADElube stopped after too many errors in a row. Here it is not individual records that help, but a look at the connection and the configuration.
Structure of a message
In the traces list, on a failure the short message appears in place of the state, so that you recognize it without opening the row. As soon as you select the row, the full message appears above the list, and it has three parts:

| Part | What it carries |
|---|---|
| Short message | The classification in one term, the same one that appears in the state column |
| Plain text | What happened, with the concrete values of the record, the affected product number for instance |
| Note | What it is usually down to and where the remedy starts |
| Technical trail | Below, in the Details dialog: the raw text the connected system or TRADElube itself reported |
Nobody expects you to read the technical trail line by line. Often it can still be placed: the values in it are your own, and the technical field name next to them says what they are about. How to read it is under traces (guides). Where that leads nowhere, copy it with the button at the top right and send it to us. With it a case is settled in minutes rather than days.
We maintain the mapping from technical errors to these readable messages centrally, and it grows with every case we come across. If no mapping applies to an error yet, you see a generic message pointing to support. That is not a graver error than the others, only one without a translated explanation yet. In that case copy the technical trail and send it to us.
Note that the message texts are maintained in German: even with the interface set to English, the short message and plain text appear in German. The frequent ones below are therefore named in German, with a translation next to them.
Frequent messages
The messages grow out of day-to-day operation, so their exact wording can evolve. What stays are the cases behind them, and most errors belong to one of six:
Verbindung fehlerhaft (connection failed). The connected system was unreachable or rejected the login. Usually concerns the whole task, not individual records. Check the credentials on the channel and the system's address. After a temporary outage the case resolves itself with the next execution.
Something is used twice, such as Doppelte Produktnummer (duplicate product number) or Zuordnung nicht eindeutig (mapping not unique). Two records claim the same identifier: a line occurs twice in the source file, the number is already taken in the target system, or two numbers were swapped with each other. The remedy is to clean up the duplicate in the source or change the identifier.
A referenced record is missing, such as Elternprodukt nicht vorhanden (parent product not available) or a manufacturer that was not found. The record points at something that does not exist on the other side yet. Often this is a follow-up error: if the parent article fails, its variants fail along with it, and the number of errors is then larger than the number of causes. Fix the referenced record first, the rest goes through the next time.
A mapping in the channel is missing, such as Bestellstatus nicht zugeordnet (order status not mapped). In the channel, under Mappings, a value pairing is not set, for instance which external order status corresponds to which internal one. Remedy: fill the affected tab in the channel and save.
The target system rejects a value: a mandatory field is empty, a text too long, a character not allowed, a tax rate not set up, as with MwSt. Satz nicht konfiguriert (tax rate not configured). The requirement is the target system's, TRADElube passes it on. Remedy: correct the value on the record or the affected setting.
Zu viele Fehler nacheinander (too many errors in a row). TRADElube stopped the execution because errors were piling up. That points to a cause hitting all records alike, usually in the configuration. Look at the first individual errors of the execution, the actual cause is there.
Checking the result
A fixed error does not vanish from the list by itself: the entry stays on Failed until the record has been transferred again. The next execution picks it up by itself, a manual start immediately, and Restart on the row forces it regardless of whether anything changed. After that the row shows Created, Modified or Neutral, and the case is closed.
Further reading
- Traces (reference): the list, the Details dialog and the restart buttons
- Traces (guides): the path from the deviation to the cause, with examples
- Contacting support: which details speed a request up when the message does not get you further
- Product does not appear in the shop: the checklist in which the message is one step of several