Job statuses
| Status | Meaning and next action |
|---|---|
| PENDING | Job accepted and waiting to process; check back for progress |
| PROCESSING | Work is running; monitor counts and pipeline progress |
| COMPLETED | Job finished without reported errors; spot-check the resulting records |
| COMPLETED_WITH_ERRORS | Some work failed; download the available error report and inspect affected rows |
| FAILED | Job-level failure; read the displayed error message and check for partial writes before retrying |
| CANCELLED | Job recorded as cancelled; the current wizard does not provide a cancel action |
The status view polls while a job is active and stops at a terminal status. Start new import resets the wizard; it does not cancel or reverse the previous job. Import history lists earlier jobs.
For pipelines, the progress table contains Entity, Rows processed, Rows failed and Stage. Read error counts alongside the stage: a terminal job can show Done even when rows failed. Processed rows are not necessarily newly created records; imports can resolve or update existing records.
Error reports
For COMPLETED WITH ERRORS, use Download error report when the button is available. The CSV contains:
| Column | Meaning |
|---|---|
| Row Number | Source row reference used by the importer |
| Error | Validation or processing message |
| Raw Data | Original row data included for investigation |
| Step | Pipeline entity/step context when available |
| Step Id | The corresponding pipeline step identifier when available |
The report is sorted by row number. Match the raw values back to your original file instead of assuming spreadsheet line numbers always match, especially with blank lines or multiline CSV cells. The download URL is temporary; reopen the job to obtain a fresh one if it expires. Not every failure produces a report: a FAILED job may instead display a job-level message.
Correct and retry
| Problem | What to check |
|---|---|
| Required field missing | Map the required field and ensure affected rows supply a value; a static value is suitable only if true for every row |
| Invalid date | Confirm the selected date format and correct impossible or mixed-format dates |
| Unknown lookup value | Map to an existing value, correct spelling, or deliberately use initial-import lookup creation where supported |
| Reference cannot resolve | Check the source platform and external ID, email or name; complete the dependency import first |
| Conflicting identity | Check source IDs and supporter numbers against existing records before changing or retrying them |
| Upload or service failure | Preserve the job ID and error; check Import history before submitting again in case the earlier request succeeded |
Correct a copy of the source file and rerun only the intended records. Keep stable external keys for updates; a new job does not automatically deduplicate every entity. Verify successful writes before retrying a whole file. If progress stays unchanged or a service error repeats, give your administrator the job ID, status, timestamp and error message, avoiding unnecessary donor data in support messages.