Verified on release 3.4.5.21 "Trade Management for Kazakhstan" (edition 3.4).
You are transferring the database to a new server. Or collapsing the old configuration and loading data into a clean one. The upload takes half an hour, and at the end — an error about "a reference that cannot be created." You dig into the configuration structure to understand who is doing this, and stumble upon a processing with a long name "DataExportImportRecreatedReferenceWriteStream". Let me say the main thing right away so you don't search for something that doesn't exist: this is a service object of the export/import mechanism. It has no postings, no e-invoices, no printed forms, and it is not opened manually from the "Sales" or "Purchases" section. Below is what it is, why it's needed, and how to handle it correctly.
1. Purpose
The processing is part of the "Data Export and Import" subsystem (from the Standard Subsystems Library, SSL). It serves the technical recreated reference write stream when loading data from an export file. When an object is recreated in the new database (with its own internal identifier-reference), the processing helps write it and correctly re-link all references to it in other objects. It has no accounting meaning — it is data migration infrastructure.
2. Where to find it
This processing is not in the regular accountant interface menu — it is not displayed in the command interface, because it is called programmatically from the loading procedure. You can open it directly in two ways:
- Via a navigation link. Service functions menu (≡ icon or "More" in the top right corner) → "Go to navigation link" → insert:
e1cib/list/DataProcessor.DataExportImportRecreatedReferenceWriteStream - The migration mechanism itself, for which it exists: section "Master Data and Administration" → "Maintenance" → "Data Export and Import" (or in "Administration" — "Data export for migration to a service" / "Data import"). This is where export/import is launched, and the processing works internally.
Important: the directly opened processing form will most likely be empty or unavailable — this is normal. It only becomes useful within the loading process.
2a. How to find your release
"Help" → "About the program" (or the ⓘ icon on the panel). In the window that opens you will see the platform version (e.g., 8.3.24) and the configuration release — "Trade Management for Kazakhstan, edition 3.4 (3.4.5.21)." Always specify the release: the set of SSL service processings changes between versions, and the path/behavior may differ.
3. How to "fill in" — or rather, how to correctly launch the process
The processing itself has no fields for manual entry — you don't specify a counterparty, amount, or rate. Everything required from you is set at the level of the "Data Export and Import" wizard. Let's review the key parameters of this wizard, because they determine whether the recreated reference write stream will work correctly.
| Parameter (in the wizard) | What it's for and what happens on error |
|---|---|
| Export file (required) | Path to the data .zip/.xml file. If you specify the wrong or corrupted file, the loading will be interrupted, and the references will not be recreated. |
| Mode: export or import (required) | Determines the direction. When importing into a non-empty database, reference conflicts may occur — this processing resolves them. |
| Load into current / new database | Only load into a prepared database of the same edition (3.4). Different editions — references will not match. |
| Release compatibility | The source release and the target release must match or be compatible. Otherwise, some objects will not find their type and the "recreation" of the reference will fail with an error. |
| Exclusive mode | Perform the loading when there are no other users in the database. Otherwise, recreated reference writing conflicts with user activity. |
Practical conclusion: there's no need — and no way — to "configure" the processing. Your task is to ensure a correct file, matching editions, and exclusive access. The write stream will handle the rest itself.
4. Worked example
There cannot be postings here, so the "example" is a real migration scenario where the processing comes into play. The figures are given to show scale, not accounting.
Situation. The company "Astana-Trade" is migrating its Trade Management for Kazakhstan 3.4.5.21 database from a local installation to the cloud.
- In the old database, via "Data Export and Import," the file
data_dump.zipis generated — 1,850 objects: 320 items, 210 counterparties, 40 contracts, 1,100 sales and receipt documents, 180 items of other catalogs. - An empty database of the same release 3.4.5.21 is created in the cloud.
- The import is launched. The platform recreates the objects — each item/counterparty gets a new internal identifier (reference).
- Here's where our processing works: a sales document referenced the counterparty "Bereke LLP" by the old identifier. The processing writes the recreated object "Bereke LLP" with the new identifier and re-links to it all 47 documents where it was mentioned.
- Result: 1,850 objects loaded, not a single reference left "hanging."
What would happen without correct operation of the mechanism: a sales document for, say, 1,160,000 ₸ (including VAT 16% — 160,000 ₸) would remain with a "broken" counterparty, and settlements on account 1210 would not reconcile. That is why the recreated reference write stream is critical for data integrity.
Note: the processing itself does not calculate the amount of 1,160,000 ₸ or the 16% VAT, and does not write postings — it only ensures that the document which forms these amounts references the correct object after the transfer.
5. Operation types
There is no separate "Operation type" field here, as in sales documents. Functionally, the processing participates in two modes of the general mechanism:
- Data export — preparing a database snapshot file (objects are serialized, their references are recorded).
- Data import — receiving the file, recreating objects, and writing recreated references with dependency re-linking. This is the "core" mode for the processing.
6. What is generated upon posting
The processing has no posting. It is not a document, therefore:
- Postings (Dr/Cr) — not generated. It does not affect RK accounts (1210, 1030, 1330, 3310, 3130, 6010, 7010).
- Electronic documents (e-invoices in the e-invoice IS, waybills) — not generated. Sending to the e-invoice IS is done by sales/receipt documents, not the migration mechanism.
- Register movements — the processing itself does not create movements. But after loading into the new database, the register entries (settlements, batches, inventory balances) that were in the source documents are restored — thanks to correctly recreated references.
The only "result" of the operation is the written objects with new identifiers and integral links between them, as well as a service log/loading journal.
7. Printed forms
There are no printed forms. The processing is not designed for paper output. The only thing you can save is the loading log (journal) with a list of processed and problematic objects, if the export/import mechanism generates one. Look for errors after loading in the "Event Log" ("Master Data and Administration" → "Maintenance" → "Event Log").
8. Common errors
- "Failed to create reference to object … / Object not found by reference." The cause is a mismatch of source and target releases, or an incomplete export file. Fix: re-export from the source of the same release 3.4.5.21, load into a database of the same release.
- "Data loading is only possible in exclusive mode." There are active sessions in the database. End user sessions ("Administration" → "Active users") and try again.
- "Data integrity violation when writing an object." Some recreated references point to objects that are not in the file (the export was partial). Fix: perform a full export, not a selective one.
- "Export format version not supported." The file was created in a different edition/platform version. Bring the platform and release to compatible ones and try again.
- Loading "hung" on a large volume. Do not interrupt the process: a break during the recreated reference write stage will leave the database in an incomplete state. Let it finish, and if necessary, increase the server timeouts.
9. FAQ
Is this a document? Where are its postings? No, this is a service processing of the migration mechanism. It has no postings, movements, or printed forms by definition.
Can I open it and "generate" something manually? You can open it via the navigation link, but you shouldn't use the form directly — it is called automatically during data loading.
Does it affect the 16% VAT or document amounts? No. Amounts and VAT are calculated by the trade documents themselves. The processing only preserves correct references so that after the transfer these documents remain linked to the correct counterparties and items.
Does it generate e-invoices or waybills? No. Electronic documents in the e-invoice IS are generated by sales/receipt documents, not by the export/import mechanism.
What are "recreated references"? When loading into a new database, objects are recreated and get new internal identifiers. The processing writes such objects and re-links all dependent data to them, so that nothing "falls off."
Why can't I transfer between different editions? The data structure of editions differs, and the type of the recreated object may not be found. Transfer between the same/compatible releases — for you, that's 3.4.5.21.
The import returned an error on one reference — is the data corrupted? Loading is usually performed within a transaction: on error, the database rolls back to its original state. Investigate the cause via the event log and repeat the process fully.
Do I need to make a backup before loading? Absolutely. Before any data loading, make a backup copy of the target database — this protects against incomplete or erroneous loading.
How do I know the transfer went correctly? After loading, verify the control figures: number of counterparties and items, inventory balances (account 1330), settlement balances (1210/3310). If the references were recreated correctly, the figures will match the source.
Can it be removed from the configuration if migration is not needed? Don't touch it. It is a standard SSL object; removing it will break the entire export/import mechanism and take the database out of support.
10. Related objects
- Main mechanism: "Data Export and Import" (SSL) — it is what calls this processing. Related service objects of the same family are responsible for object serialization, reference mapping, and data writing.
- Based on what it is "entered": it is not entered based on documents. It is initiated manually from the export/import wizard.
- What is linked by data: after the transfer, correctness is verified against the "Sales of Goods and Services," "Receipt of Goods and Services" documents, settlement and balance registers — that is, against everything the recreated objects referenced.
How to find your release
"Help" → "About the program" — it shows the "1C:Enterprise" platform version and the configuration release. This guide was prepared for "Trade Management for Kazakhstan," edition 3.4, release 3.4.5.21. If you have a different release, the set and behavior of SSL service processings may differ — check the description for your release.
Prepared for release 3.4.5.21 "Trade Management for Kazakhstan" (edition 3.4).