Guides
Each guide takes you through one task, from the starting point to the point where you see that it works. The groups follow the life of an integration: take over, extend, refine, automate, change, monitor, repair, clean up. Look for the group that describes your situation, and within it the task.
What each field of a screen means is in the Reference. The first setup from scratch is shown in Getting started.
Taking over someone else's configuration
- Understand what a route does — a route that someone else built: from where to where, when, and what happens to the data along the way.
- Find out where a value in the target comes from — from the surprising value in the target field back to the mapping row that wrote it.
Setting up another integration
- Create a connection — a new external system with address, sign-in and field structure, until a route can use it.
- When the field selection stays empty — the mapping offers no source fields, although the connection is in place.
- When the external system responds differently than expected — the run ends with "Connection does not match the response" or brings only part of the dataset.
- Fetch from a queue and acknowledge — fetch each message and confirm to the partner that it has arrived.
- Pass incoming messages to the responsible route — one input, several kinds of messages, the responsible route for each kind.
- Match a response to the right document — a reply from the partner lands in exactly the document it belongs to.
- Transfer master data into a second company — write items from your own SAP company into a second one.
- Transfer data within SAP Business One — create a second document from one in the same company, without a connection.
- Transfer deletions or deliberately leave records in place — what happens in the target system when a record drops out of the transfer.
- Decide which side wins in a field conflict — who is right when a field was changed in the source and in the target system.
- Send several records in one call — a writing step bundles the records for a target system that accepts many at once.
Transferring only what is needed
- Transfer only what changed — a route reads only the records that have changed since the last run.
- Narrow down with a data filter — define which records a route knows at all.
- Limit the first run — first transfer a small excerpt, look at the result, then lift the limit.
Mapping fields and values
- Build a target field from several source values — a value made of fixed text and fields, and what happens when a part is missing.
- Convert and translate values — with a value table,
paypalin the shop becomesPPin SAP Business One. - Map header and lines when the target splits differently — a message with a list of lines becomes a document header and document lines.
- Load missing details — fetch details that are missing from the record from SAP Business One before the mapping.
- Write a field only on create — the shop's requested date goes into the new document only, and the committed delivery date stays.
- A field is missing from the mapping — the target system requires a field that is not in the selection.
Triggering and scheduling
- Choose the right trigger — by the clock or in response to an event, and when both together make sense.
- React to a change in SAP — a change goes out as soon as it is saved in SAP Business One.
- Run on a schedule — the changes daily, the whole excerpt with cleanup weekly.
- Put two routes in order — one route should only run when another has finished: staggered schedules, a dispatch or the same trigger for both.
Changing without disrupting operation
- Change a running configuration — change in the draft, compare and activate while everything keeps running.
- Fix the findings of a validation — work through the checklist in order until the draft is Validated.
- Two editors on the same object — what the warning before typing and the conflict when saving mean.
- Go back to an earlier snapshot — put the snapshot from before the last activation back into operation immediately.
- Bring a configuration into the production system — a snapshot is written to a file and imported on another system.
- When a connected system changes — a field or an interface changes, and a route stalls or is blocked.
Monitoring and proving
- Check whether a run went through cleanly — the morning round from the Overview through Execution down to the single job.
- Find and trace a single job — prove what was transferred when, and be able to say why something is still missing.
- See how much goes through the interface — volume, errors, duration and backlog, from the day down to the single route.
Fixing problems
- Wait or intervene? — how you can tell whether something resolves by itself.
- Stop a running run — abort a whole run or just a single job.
- It ran for months and stopped yesterday — narrow down the cause step by step, from the whole system down to the job.
- Retry failed jobs — return many similar cases to processing in one go.
- Resolve a clarification — read from the cause and state of a Clarification case whether Reprocess or Close is the right outcome.
- Avoid and investigate duplicates — before a second attempt, check whether the record is already in the target.
- Push manually — a run or a single record — transfer now what would otherwise only come with the next run.
Managing and cleaning up
- Adjust retention periods — define how long Xchange keeps jobs and its system log.
- Release or discard a suspended cleanup — whether Xchange may tell the target system that records no longer belong there.
- Take a route out of operation — pause a route or delete it for good.
- Remove test data before going live — remove test runs and test configuration before live operation begins.
- Change the credentials of a connection — enter and test a new password or client secret.
- Run a database update — after a version change, bring the database up to date and resume operation.