Wait or intervene?
Something does not look right. A run is red, a number is rising, a job has been at the same state for a while. Ask first: Do I need to do anything at all, or will this resolve itself?
Xchange answers the question itself
In two places, the answer is given in plain language:
- For a failed job — in History, in the job details under Status and history. There you find Attempts and, as long as another attempt is pending, Next retry with a time. If a time is shown there, the matter is still with Xchange.
- For a case in Clarification — on the detail page in the Cause area, below the description. There is exactly one sentence about what the case means for the record.
This one sentence exists in five versions, and they come down to three answers: a reprocessing is running right now — wait for it; the record runs along again in the next run or it is skipped until a stated time, after which it runs along again — it continues by itself; in the other versions, the record stays where it is until you decide. Which five they are and when each one appears is described under Clarification.
The rule behind it
- Xchange retries temporary errors itself. The target system was briefly unreachable or took too long — the next attempt can succeed. Such jobs stay outside Clarification; their only trace is Attempts: 2 in History.
- Permanent errors stay at one attempt. The answer would be the same on the tenth attempt as on the first. The job goes to Clarification, and there it waits for a person — unless the record changes in the source system or the case carries a deadline.
How Xchange tells the two apart and what that means for the next run is described under Execution.
A run that is currently working is unremarkable. Running now on the Overview and a progress bar mean that something is moving forward — even if it takes longer than usual.
Where to go if you intervene
| What you see | Where to go next |
|---|---|
| A case in Clarification that is waiting for a decision | Resolve a clarification |
| Many cases with the same cause, and the cause is fixed | Retry failed jobs |
| Nothing failed, and yet something is missing in the target | Push manually |
| Nothing arrived at all, and you do not know why | It ran for months and stopped yesterday |
| The suspicion that something was written twice | Avoid and investigate duplicates |
| Something is running right now and should stop immediately | Stop a running run |
The last row is the only urgent one. All the others can wait until you have finished reading what Xchange tells you about the case.
See also
- History (reference) — where Attempts and Next retry are shown on a job.
- Clarification (reference) — the five versions of the sentence and all causes.
- Overview (reference) — Action required, which brings you here in the first place.