ssis2fabric

Convert SSIS DTSX packages to Microsoft Fabric Data Factory items.

ssis2fabric is a Python command-line tool that reads a SQL Server Integration Services (SSIS) .dtsx package file and creates the following Fabric items in a target workspace:

SSIS artifact	Fabric artifact
Package control flow	Data Pipeline
Data Flow task	Dataflow Gen2 (Power Query M)
Connection Manager	Fabric Shareable Connection

Activities that cannot be fully auto-converted are created with state: InActive so the pipeline can still be saved and opened in Fabric — they just need follow-up manual editing.

---

Prerequisites

• Python 3.9+
• A Microsoft Fabric workspace where you have Contributor or higher access
• A browser (for interactive Microsoft Entra ID sign-in)

---

Installation

Option A — Install the latest release wheel (recommended)

Download the .whl from the Releases page, then:

pip install ssis2fabric-0.1.0-py3-none-any.whl

This installs the ssis2fabric command directly onto your PATH and pulls in all dependencies automatically.

Option B — Install from source

git clone https://github.com/kromerm/markssis.git
cd markssis
pip install .

Option C — Editable / development install

git clone https://github.com/kromerm/markssis.git
cd markssis
pip install -e .

After any of the above you can verify the install with:

ssis2fabric --help

---

Quick Start

1. Dry run — review the conversion without touching Fabric

Parse your package, convert it, and write the JSON output locally. No Fabric API calls are made and no authentication is required.

ssis2fabric \
  --dtsx  MyPackage.dtsx \
  --workspace-id  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
  --dry-run \
  --output-dir  output/

Expected output:

============================================================
  Parsing DTSX: MyPackage.dtsx
============================================================

============================================================
  SSIS Package Summary
============================================================
  Package name   : MyPackage
  Variables      : 3
  Connections    : 2
  Top-level tasks: 5
  Data flows     : 1
  Precedence     : 4

  Connection Managers:
    - DW  [OLEDB:SQL]  server=myserver  db=DW
    - FlatFiles  [FILE]

  Control Flow Tasks:
    - Load Staging  [DataFlow]
    - Send Status Mail  [SendMail]
    - Cleanup  [ExecuteSQL]
    - Archive Files  [FileSystem]
    - Notify  [SendMail]

============================================================
  Dry Run – building pipeline definition only
============================================================
  [saved] output/connections.json
  [saved] output/dataflow_Load Staging.json
  [saved] output/pipeline_MyPackage.json

  [dry-run] No Fabric API calls made.

2. Full migration — create items in Fabric

A browser window will open for Microsoft Entra ID sign-in on first use.

ssis2fabric \
  --dtsx  MyPackage.dtsx \
  --workspace-id  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
  --folder  "SSIS Migration"

3. Override the pipeline display name

ssis2fabric \
  --dtsx  MyPackage.dtsx \
  --workspace-id  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
  --folder  "SSIS Migration" \
  --pipeline-name  "Daily Sales Load"

4. Skip dataflow creation (pipeline only)

Useful if you only need the control-flow pipeline and want to wire up dataflows manually.

ssis2fabric \
  --dtsx  MyPackage.dtsx \
  --workspace-id  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
  --no-dataflows

5. Skip connection creation (use dummy connection IDs)

Useful when connections already exist in the workspace and you only want the pipeline/dataflow items.

ssis2fabric \
  --dtsx  MyPackage.dtsx \
  --workspace-id  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
  --no-connections

6. Save artifacts locally AND create in Fabric

ssis2fabric \
  --dtsx  MyPackage.dtsx \
  --workspace-id  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
  --folder  "SSIS Migration" \
  --output-dir  output/

7. Verbose mode — debug HTTP calls

ssis2fabric \
  --dtsx  MyPackage.dtsx \
  --workspace-id  xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
  --dry-run \
  --verbose

---

Options

Flag	Description
--dtsx PATH	Required. Path to the .dtsx file
--workspace-id GUID	Required. Target Fabric workspace GUID
--folder NAME	Optional folder name to place items in
--pipeline-name NAME	Override the Fabric pipeline display name (default: SSIS package name)
--dry-run	Parse + convert but do not call Fabric APIs
--verbose / -v	Print HTTP request/response details
--output-dir DIR	Save converted JSON artifacts to this directory
--no-connections	Skip Fabric connection creation
--no-dataflows	Skip Fabric Dataflow Gen2 creation

---

Authentication

The tool uses Microsoft Entra ID interactive browser authentication (InteractiveBrowserCredential from azure-identity). On first run a browser window will open to https://login.microsoftonline.com — sign in with the account that has access to the target Fabric workspace.

The token is cached in memory for the duration of the run.

---

SSIS → Fabric Mapping Details

Control Flow → Pipeline Activities

SSIS Task	Fabric Activity	Notes
Execute SQL Task	Script or SqlServerStoredProcedure	SP name detected automatically
Data Flow Task	RefreshDataFlow	References the Dataflow Gen2 created for that task
ForEach Loop	ForEach	Inner tasks recursively converted
For Loop Container	Until	⚠ Set InActive – loop expressions converted to TODO placeholder (SSIS expression syntax differs from Fabric)
Sequence Container	IfCondition (always-true wrapper)	Inner tasks recursively converted
Execute Package Task	ExecutePipeline	⚠ Set InActive – referenced pipeline may not exist yet
Script Task	Script	⚠ Set InActive – logic must be manually ported
Send Mail Task	Office365Email	⚠ Set InActive – connection required; fields From/To/CC/BCC/Subject/Body/Priority/Attachments populated
Web Service Task	WebActivity (HTTP POST)	⚠ Set InActive – SOAP endpoint URL, headers, and envelope body must be updated
FTP Task (Receive)	Copy (FTP → ADLS/lakehouse)	⚠ Set InActive – linked service and dataset configuration required
FTP Task (Send)	Copy (ADLS/lakehouse → FTP)	⚠ Set InActive – linked service and dataset configuration required
FTP Task (DeleteRemoteFile / DeleteLocalFile)	Delete	⚠ Set InActive – linked service configuration required
FTP Task (directory ops / rename)	Script	⚠ Set InActive – no direct Fabric equivalent
Execute Process Task	WebActivity	⚠ Set InActive – no direct equivalent
File System Task	Script	⚠ Set InActive – rework using Lakehouse file APIs
Bulk Insert Task	Copy	⚠ Set InActive – source/sink configuration required
All others	Wait (1 s)	⚠ Set InActive – manual replacement required

Disabled tasks: Any SSIS task with DTS:Disabled="True" is emitted with state: InActive and its description prefixed with [Disabled in original SSIS package].

SSIS Variables → Pipeline Parameters: User:: namespace variables are converted to Fabric pipeline parameters (type mapping: String/DateTime → string, Int32/Int64 → int, Double/Decimal → float, Boolean → bool). System:: variables are skipped.

Data Flow → Dataflow Gen2 (Power Query M)

Components are mapped to M expression steps:

SSIS Component	M Expression
OLE DB / ADO Source	Value.NativeQuery(Source, sql)
Flat File Source	Csv.Document(File.Contents(...))
Derived Column	Table.AddColumn
Aggregate	Table.Group
Sort	Table.Sort
Merge Join	Table.NestedJoin
Lookup	Table.NestedJoin (left outer)
Union All	Table.Combine
Data Conversion	Table.TransformColumnTypes
Pivot	Table.Pivot
Unpivot	Table.UnpivotOtherColumns
Conditional Split	Table.SelectRows
Destination	Comment only – data loading done via Copy activity

All steps contain // TODO annotations where manual adjustment is needed.

Connections → Fabric Connections

SSIS connection types are mapped to Fabric connectivity types:

SSIS Type	Fabric Type
OLEDB / ADO.NET	SQL (ShareableCloud)
File / Flat File	File
HTTP	Web (Anonymous)
FTP	FTP
SMTP / others	SQL (dummy)

Connections are created with skipTestConnection: true and dummy/placeholder credentials so the API call succeeds. Credentials must be updated in Fabric after migration.

---

Output Files

When --output-dir is specified the following files are written:

File	Contents
connections.json	Array of Fabric connection creation payloads
dataflow_<name>.json	Fabric Dataflow Gen2 item definition
pipeline_<name>.json	Fabric pipeline-content.json (base64-wrapped)

---

Post-Migration Checklist

1. Open the pipeline in Fabric and find activities shown with a ⚠ badge (InActive).
2. For each InActive activity:
   • Review the description field for original SSIS task details.
   • Update connection references, SQL statements, or script logic.
   • Set state back to Active once the activity is ready.
3. Open each Dataflow Gen2 item and:
   • Set real data source connections.
   • Fix // TODO expressions throughout the M query.
4. Update connection credentials via Fabric > Manage connections and gateways.
5. Run the pipeline in Debug mode and iterate.

---

Project Structure

SSIS3/
├── ssis2fabric/
│   ├── __init__.py
│   ├── __main__.py        # python -m ssis2fabric
│   ├── cli.py             # CLI argument parsing & orchestration
│   ├── models.py          # SSIS dataclass models
│   ├── parser.py          # DTSX XML parser
│   ├── converters/
│   │   ├── __init__.py
│   │   ├── connections.py # Connection Manager → Fabric Connection
│   │   ├── dataflow.py    # Data Flow → Dataflow Gen2 (M query)
│   │   └── pipeline.py    # Control Flow → Fabric pipeline
│   └── fabric/
│       ├── __init__.py
│       └── client.py      # Fabric REST API client (user-auth)
├── requirements.txt
├── pyproject.toml
└── README.md