ProwessDataUnloader
Find recoverable documents in WiredTiger files, historical values and deep fragments.
What is PDU
Prowess Data Unloader (PDU) recovers MongoDB data from preserved physical files, even when the database cannot start.
Extract data even when the instance cannot start
Read a frozen WiredTiger directory copy and extract documents that can still be parsed, without starting the damaged instance.
Extract directly from collection files
Read collection-*.wt files. Single-file mode records where the file came from, but does not automatically identify its business collection name.
Read catalog and checkpoints
Read turtle, metadata and the catalog to identify collection entries and the selected checkpoint. This does not automatically replay the journal.
Extract historical deletion and update candidates
Extract old values from collection history, History Store and pre-images, and record their sources. These values still need to be assessed before they can represent a complete pre-incident state.
Scan images and deep fragments
Read data using supplied extent maps, or scan isolated complete blocks and plaintext BSON fragments. Save possible complete documents separately from field fragments.
Why Choose PDU?
Explore MongoDB recovery methods for the files, logs and fragments that remain.
Catalog & Checkpoint Recovery
Confirm the collection information in the catalog and the selected checkpoint, then export each collection as BSON or Canonical EJSON.
Learn moreHistorical Value Candidates
Extract possible deleted or updated records from surviving collection history. PDU does not automatically write the results back to a running database.
Learn moreSafe & Non-destructive
Use frozen source copies. Save task state and exported data in PDU's current working directory, outside the source files, to avoid writing to damaged data.
Learn moreFragment Scanning
Process mapped extents, isolated complete blocks and plaintext deep fragments separately. Individual field fragments are not counted as recovered documents.
Learn moreHistory Store & Pre-images
Save complete old values directly. Reversing modifications requires the base value and the full modification chain. Separate log evidence is needed to link an old value to a specific operation.
Learn moreCollection Export & Resume
Resume exports using PDU task records and file metadata. If one collection fails, record the cause and continue with the others.
Learn moreWhat PDU Brings
Preserve the incident files, then choose a recovery method based on the files and logs that remain.
Professional Document Extraction
Use interactive commands to read WiredTiger directories, collection files and checkpoints, and extract readable documents from the surviving files.
Find Historical Documents
Read old versions from collection history, History Store and pre-images separately. PDU records missing base values or modification chains and transactions that cannot be identified.
From Documents to Deep Fragments
Extract possible complete documents and save suspected document structures and individual field fragments. Count fragments separately from recovered documents.
Minimize Downtime
Maximize Recovery
After data loss, preserve the original files first. PDU extracts readable data and records parsing failures to help determine what can be recovered.
- 01Keep incident sources read-only and save results in a separate launch directory
- 02Choose recovery paths for files, logs and fragments
- 03Maintain full control over sensitive data
- 04Preserve native results, failure reasons and provenance
Ready to Recover Your Data?
Read the MongoDB documentation to confirm versions, surviving sources and applicable recovery methods.