Oracle CDC source succeeds with zero rows moved; per-table task never starts

Hi acitsme​,
Thank you for sharing the additional details. The latest results help provide more context.

According to the Microsoft documentation, Oracle CDC in Copy job uses LogMiner to capture changes from the Oracle redo logs. The documented configuration is to select the CDC-enabled tables, choose Incremental copy, and then Save + Run. Microsoft also states that the first run performs an initial full snapshot, while subsequent runs capture inserted, updated, and deleted records through CDC. 

You have already confirmed that the table is identified as CDC-enabled, it is the only table selected, the Oracle/LogMiner prerequisites have been verified, and a normal full copy of the same table succeeds.

Given the additional Save/Review + save validation error and the fact that the existing Copy job does not reload its previously configured Source, Destination, and Mapping fields, I would try creating the CDC Copy job again with the single CDC-enabled table and the documented Incremental copy configuration, and then check whether the configuration can be saved and the initial full snapshot starts.

If the same validation error continues, the details you have already collected — including the Copy job status, table-level Not started status, validation message, timestamps, and browser network information — should provide useful information for further investigation.
I hope this helps with the next step.

For more details, please refer to the below Official Microsoft documentation:
Change data capture from Oracle database using Copy job - Microsoft Fabric | Microsoft Learn
Change data capture (CDC) in Copy Job - Microsoft Fabric | Microsoft Learn

I hope this helps. Please feel free to reach out if you have any further questions.
Thank you.

Hi acitsme​,

That latest test changes the direction I would investigate.

If a brand-new Copy job reaches Save/Review, immediately returns Error saving, validation error, no save request is sent from the browser, and reopening Edit copy job shows Source, Destination and Mapping as blank, I would first treat this as an authoring/configuration-state problem rather than an Oracle LogMiner problem.

The documented Oracle CDC Copy job flow should allow you to select the CDC table, destination and incremental mode, then Save + Run. The first run should perform the initial full snapshot before subsequent CDC runs.

Since your normal full Copy job against the same Oracle source succeeds, I would use that as an A/B test and avoid changing further Oracle privileges until the CDC job can be saved consistently.

One useful diagnostic is the newer Copy job JSON editing capability. It became GA in May 2026. After reproducing the validation error, inspect the Copy job definition and check whether the source, destination and table/mapping configuration were actually persisted. I would use the JSON primarily for diagnosis rather than manually constructing undocumented CDC settings.

I would also reproduce once in an InPrivate window or clean browser profile to rule out cached client state/extensions. If the UI still loses the configuration and no save request reaches the backend, I think that is strong evidence to open a Fabric Support case with:

• the Copy job item ID
• workspace ID
• exact timestamp
• browser console output
• HAR/network capture
• screenshots of the configured wizard and blank Edit screen
• exported/viewed Copy job JSON

Oracle CDC in Copy job is still Preview, so if this reproduces consistently I would want Microsoft Support to confirm whether this is an authoring validation defect.

I would not expect supplemental logging or LogMiner grants to explain a save validation that occurs before the job definition is submitted.

AI-assisted drafting: AI was used to help structure and phrase this response. I reviewed and validated the technical content before posting.