Examples and use cases

Complete command examples for the supported migration paths, plus advanced scenarios. Each example assumes a credentials.json file with these entries (see the Quick Start Guide and Database Connections):

Auth ID	Role	ds_type	database
oracle_prod	Source	oracle	ORCLPDB1
postgres_prod	Source	postgres	analytics
mssql_source	Source	mssql	SourceDB
netezza_source	Source	netezza	PROD_DB
mssql_target	Target	mssql	target_db
pg_target	Target	postgres	dwh
ms_mig_log	Migration tracking	mssql (or postgres)	migration_log

The source and target database names are taken from these entries, so the commands do not pass --source_db_name or --target_db_name. The examples describe MigratorXpress v0.7.3.

---

Oracle to SQL Server

Two-phase approach: transfer data first, then add constraints.

Phase 1: Schema and data

Linux

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --source_schema_name FINANCE \
  --target_db_auth_id mssql_target \
  --target_schema_name finance \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --fasttransfer_p 2 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

Windows (PowerShell)

.\MigratorXpress.exe --auth .\credentials.json `
  --source_db_auth_id oracle_prod `
  --source_schema_name FINANCE `
  --target_db_auth_id mssql_target `
  --target_schema_name finance `
  --migration_db_auth_id ms_mig_log `
  --n_jobs 4 `
  --fasttransfer_p 2 `
  --drop_tables_if_exists true `
  --task_list translate create transfer diff

Phase 2: Constraints

Constraint copy (copy_pk, copy_ak, copy_fk) is available for Oracle, PostgreSQL and SQL Server sources with a SQL Server target. Foreign keys are created according to --fk_mode (default disabled).

Linux

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --source_schema_name FINANCE \
  --target_db_auth_id mssql_target \
  --target_schema_name finance \
  --migration_db_auth_id ms_mig_log \
  --task_list copy_pk copy_ak copy_fk

Windows (PowerShell)

.\MigratorXpress.exe --auth .\credentials.json `
  --source_db_auth_id oracle_prod `
  --source_schema_name FINANCE `
  --target_db_auth_id mssql_target `
  --target_schema_name finance `
  --migration_db_auth_id ms_mig_log `
  --task_list copy_pk copy_ak copy_fk

---

PostgreSQL to SQL Server

Linux

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id postgres_prod \
  --source_schema_name analytics \
  --target_db_auth_id mssql_target \
  --target_schema_name analytics \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --fasttransfer_p 2 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

Windows (PowerShell)

.\MigratorXpress.exe --auth .\credentials.json `
  --source_db_auth_id postgres_prod `
  --source_schema_name analytics `
  --target_db_auth_id mssql_target `
  --target_schema_name analytics `
  --migration_db_auth_id ms_mig_log `
  --n_jobs 4 `
  --fasttransfer_p 2 `
  --drop_tables_if_exists true `
  --task_list translate create transfer diff

---

SQL Server to PostgreSQL

Target schema, table and column names are created in lower case on PostgreSQL (dbo.Orders lands in sales.orders below). copy_fk is not available for a PostgreSQL target, so constraints are limited to copy_pk copy_ak.

Linux

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id mssql_source \
  --source_schema_name dbo \
  --target_db_auth_id pg_target \
  --target_schema_name sales \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff copy_pk copy_ak

Windows (PowerShell)

.\MigratorXpress.exe --auth .\credentials.json `
  --source_db_auth_id mssql_source `
  --source_schema_name dbo `
  --target_db_auth_id pg_target `
  --target_schema_name sales `
  --migration_db_auth_id ms_mig_log `
  --n_jobs 4 `
  --drop_tables_if_exists true `
  --task_list translate create transfer diff copy_pk copy_ak

The same command works from an Oracle (oracle_prod) or PostgreSQL (postgres_prod) source. For an Oracle source, check the result of diff and spot-check date, timestamp and binary columns: the exactness of Oracle to PostgreSQL transfers was not verified.

---

Netezza to SQL Server

A Netezza source can only be migrated to SQL Server, and without constraint copy (copy_pk, copy_ak and copy_fk are rejected at startup with exit code 10).

Linux

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id netezza_source \
  --source_schema_name WAREHOUSE \
  --target_db_auth_id mssql_target \
  --target_schema_name warehouse \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --fasttransfer_p 2 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

Windows (PowerShell)

.\MigratorXpress.exe --auth .\credentials.json `
  --source_db_auth_id netezza_source `
  --source_schema_name WAREHOUSE `
  --target_db_auth_id mssql_target `
  --target_schema_name warehouse `
  --migration_db_auth_id ms_mig_log `
  --n_jobs 4 `
  --fasttransfer_p 2 `
  --drop_tables_if_exists true `
  --task_list translate create transfer diff

---

SQL Server to SQL Server

Clone or migrate schemas between SQL Server instances. SQL Server sources have no intra-table parallel method, so --fasttransfer_p has no effect; use --n_jobs for parallelism.

Linux

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id mssql_source \
  --source_schema_name dbo \
  --target_db_auth_id mssql_target \
  --target_schema_name dbo \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

Windows (PowerShell)

.\MigratorXpress.exe --auth .\credentials.json `
  --source_db_auth_id mssql_source `
  --source_schema_name dbo `
  --target_db_auth_id mssql_target `
  --target_schema_name dbo `
  --migration_db_auth_id ms_mig_log `
  --n_jobs 4 `
  --drop_tables_if_exists true `
  --task_list translate create transfer diff

---

Advanced scenarios

Incremental data sync

Re-run only the transfer and diff tasks to refresh data without recreating tables. Useful for periodic syncs during a phased migration.

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --source_schema_name SALES \
  --target_db_auth_id mssql_target \
  --target_schema_name sales \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --fasttransfer_p 2 \
  --task_list transfer diff

tip

transfer empties each target table before reloading it (default --load_mode truncate); use --load_mode append to add rows instead. --drop_tables_if_exists only affects the create task, so it is not needed here.

For Oracle and Netezza sources, a table whose maximum ROWID (Oracle) or creation transaction ID (Netezza) has not changed since the previous transfer is skipped (nothing to transfer). This only detects new rows: add --without_xid to force a full reload of every table.

Schema-only migration

Translate and create the schema without transferring any data. Useful for reviewing the translated column types (stored in the targetcolumns tracking table) before committing to a full migration.

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --source_schema_name INVENTORY \
  --target_db_auth_id mssql_target \
  --target_schema_name inventory \
  --migration_db_auth_id ms_mig_log \
  --drop_tables_if_exists true \
  --task_list translate create

Parallel large tables

For schemas with very large tables, increase both parallelism levels:

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --source_schema_name EVENTS \
  --target_db_auth_id mssql_target \
  --target_schema_name events \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 8 \
  --fasttransfer_p 4 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

• --n_jobs 8 migrates up to 8 tables concurrently.
• --fasttransfer_p 4 splits each table whose estimated row count is at least --ft_large_table_th (default 250,000,000) into 4 parallel streams. Lower the threshold to parallelize smaller tables.

warning

Higher parallelism increases load on both the source and target databases. Monitor CPU and I/O during large migrations and adjust values as needed.

Multi-schema patterns

Migrate multiple schemas in one run with SQL LIKE patterns or a comma-separated list. Omit --target_schema_name so that each source schema keeps its name on the target:

# All schemas starting with "APP_", each recreated under its own name
./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --source_schema_name "APP_%" \
  --target_db_auth_id mssql_target \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

# Every user schema in the database (--source_schema_name omitted)
./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --target_db_auth_id mssql_target \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

To consolidate several source schemas into one target schema, give --target_schema_name a plain name. Tables with the same name in different source schemas would collide, so use this only when table names are unique:

# SALES_EU and SALES_US merged into the single target schema sales_all
./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --source_schema_name "SALES_EU, SALES_US" \
  --target_db_auth_id mssql_target \
  --target_schema_name sales_all \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --drop_tables_if_exists true \
  --task_list translate create transfer diff

Resume a run

If a run is interrupted or some tables failed (see the RUN SUMMARY at the end of the log), resume it with its run ID. Schema names and the task list come from the tracking database:

./MigratorXpress --auth ./credentials.json \
  --source_db_auth_id oracle_prod \
  --target_db_auth_id mssql_target \
  --migration_db_auth_id ms_mig_log \
  --n_jobs 4 \
  --resume 20260203-cf5be0ac-159f-4dad-861e-8b2b6f3868d7

---

Best practices

1. Migrate data before constraints. Run translate create transfer diff first, then copy_pk copy_ak copy_fk in a separate run. This avoids constraint-violation errors during bulk loading.

2. Start with a small schema. Test the full workflow on a schema with a few tables before migrating a large production database.

3. Use diff to validate. Always include diff in your task list to compare row counts, column counts and column statistics between source and target.

4. Tune parallelism gradually. Begin with --n_jobs 2 --fasttransfer_p 2 and increase based on observed throughput and system load.

5. Secure your credentials file. Passwords are stored in clear text in the credentials file: restrict its permissions to the migration user and keep it out of version control.

6. Review translated types. Run translate create first and inspect the created tables (or the targetcolumns tracking table) before transferring data.

7. Check the exit code and the run summary. The process exits with code 40 when some tables failed and 41 when all of them failed; read the RUN SUMMARY: n/m tables succeeded block and resume failed runs with --resume.

Next steps

• Quick Start Guide -- step-by-step setup
• CLI Reference -- all available options and flags
• Command Wizard -- build your command interactively