How Blober Handles Duplicate Files in Long-Running Backups

A cloud library can take hours or days to copy. During that time, a computer may restart, a network may disconnect, or a provider may ask you to sign in again. You may also want to run the same export later to collect newer files.

Those two situations need different tools:

• Resume the existing task after an interruption. Blober keeps that task's progress and does not process its completed records again. This works on every provider and needs no setting.
• Use destination deduplication for a deliberate rerun. When a new task writes to the same destination, deduplication decides whether files that are already there are skipped or overwritten. This depends on the destination supporting it.

A workflow is the reusable setup in Blober: source, destination, selected folders, filters, and naming rules. Each time you start that workflow, Blober creates a task that records the progress of that run.

Within a task, Blober records each source path as it is processed or skipped. If that task is interrupted and later continues, records already marked complete are not processed again.

This task state is stored locally with your workflows. Closing Blober, restarting the computer, or installing a newer version of Blober keeps the existing workflows and task history.

1. Reopen Blober.
2. Find the existing task on the Progress page.
3. If it was paused, resume that same task.
4. If it was still pending or running when Blober closed, the app can rediscover it after launch and continue it automatically.
5. Complete any provider login prompt that appears.

Don't start a new task just because the original one stopped. A new run has its own progress records and does not continue the interrupted run.

On long cloud transfers, a provider login may expire during a run. Signing in again refreshes access, and the existing task keeps its record of completed files.

Deduplication is a destination setting that decides what happens when the exact destination path is already occupied:

• Skip if a file with the same name exists is the default. The file is not written, the record is marked skipped, and the run continues with the next file.
• Overwrite the existing file replaces the destination copy.
• Keep both (rename the new copy) is available on Local destinations, which save the new file as name (1).ext, then name (2).ext.

The comparison is the destination path: destination folder, path template, and filename. File contents are never compared.

The control appears only for destinations that can act on an existing file:

• Source-only providers have nothing to deduplicate.
• GoPro Cloud cannot replace or compare by name, so a rerun writes the file again.

Azure Blob Storage adds one rule of its own: archived blobs are always skipped, whichever policy is selected, because an archived blob cannot be rewritten. Rehydrate the blob first if you want to replace it.

For a repeat export where an existing destination path means "already transferred":

1. Open or create the workflow.
2. Select the destination.
3. Open the destination configuration.
4. Set Deduplication to Skip if a file with the same name exists.
5. Save and run the workflow.

Blober checks the destination before it opens the source, and checks again when it claims the final filename. The second check also stops two concurrent transfers from racing to create the same path.

See Deduplication for the policies and the destinations that support them.

Google Photos is a common case, because a large library can take a long time to download and may contain repeated camera filenames.

For an interrupted Google Photos download, continue the existing task. For a deliberate later export into the same local folder, use Skip if a file with the same name exists only when keeping one file per exact destination filename suits your library.

The full setup is in How to Back Up Google Photos Without Google Takeout, which uses the same setting as the Deduplication reference.

• Keep the computer connected to power and avoid moving or disconnecting the destination drive.
• Use a stable destination folder and path template for the whole task.
• Pause before intentionally shutting down if you want manual control over when the task continues.
• After an unexpected restart, reopen Blober before creating anything new and check the existing task first.
• Respond to provider login prompts instead of replacing the workflow.
• Review files marked failed or skipped after completion. A provider may temporarily make an individual file unavailable even though the rest of the transfer completed.

Will updating Blober erase my workflows or task progress? No. Installing an update keeps your workflows, tasks and per-file progress. Reopen the existing task after the update instead of creating a replacement.

Does skipping by name compare file contents? No. It checks whether the exact destination path already exists and does not calculate or compare content hashes. A file you edited since the last run is skipped too if its path is unchanged; Overwrite the existing file replaces it.

Does a skipped file count as complete in the same task? Yes. Blober saves the reason in the task details and continues with the remaining files, so one unavailable source file or an intentional destination duplicate doesn't block the rest of the transfer.

Can I use deduplication with Google Drive? Yes. Google Drive destinations offer skip and overwrite. If a destination doesn't show the Deduplication control, it cannot act on an existing file, and resuming the original task is how you avoid transferring the same files twice.