> For the complete documentation index, see [llms.txt](https://ecfmg.gitbook.io/azure-serverless-quickstart/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ecfmg.gitbook.io/azure-serverless-quickstart/setup/infrastructure-setup/optional-items/azure-data-factory.md).

# Azure Data Factory

In the Azure Portal - search for Data Factories  and select it - it will take yoou to the [Data Factories blade](https://portal.azure.com/#blade/HubsExtension/BrowseResource/resourceType/Microsoft.DataFactory%2FdataFactories).

Click **Create data factory** button, you will be display the Create Data Factory  blade, 

Subscription: <\<your subscription>>\
Resource Group: sharethrift-rg\
Region: east US 2\
Name: sharethrift-data-factory\
Version: v2\
\
Click Next: Git Configuration\
Check "Configure Git Later"

Click: Review + create (wait for the validationo to complete), then click "Create", you will be taken to the overview page, click the "Go to resource" button<br>

On the data factory instance page,Click the **author and monitor** button, the Azure Data Factory portal will open.

Click Create Pipeline:

in the properties pane on the right of the screen:

* Name: account-sync
* Description: Copies accounts from legacy database to cloud database.
* Concurrency: <\<leave empty>>

Under Activities, choose Move & transform,  drag Copy data to the Design Surface

At bottom of screen:

* **General** Tab
  * Name: copy-accounts
  * (leave other settings the default)
* **Source** Tab:
  * Source dataset:
    * Click **+ New** (chooser pane appears)
      * Select appropriate Databases Connection type (e.g. **PostgreSQL**)
    * Choose **Open** next to the source connection that you selected
      * Linked service:&#x20;
        * Click **+ New** ( connection pane shows) *or choose an existing connection*&#x20;
          * Name: **accountsdb**
          * (other appropriate details.. may need to set "not validate sever certificate")
        * Select the appropriate table and click preview data to ensure data is correct
        * Select **"account-sync"** tab at top to return to the pipeline
  * &#x20;Use Query: **table**
* **Sink** Tab:
  * Sink dataset:
    * Click **+ New** (chooser pane appears0
      * Select **Azure CosmosDB (MongoDB API)**
    * Choose Open next to the sink dataset that you selected
      * Linked service:
        * Click **+ New** (connection pane shows)
        * Name: **sharetriftdb**
        * (other appropriate details..)
        * Account selection method : **From Azure subscription**
          * Subscription: **<\<your subscription>>**&#x20;
          * Account: **sharethrift**
          * DBName: **sharethrift**
      * Collection Name: **accounts**
      * Select **"account-sync"** tab at top to return to the pipeline
  * Write behavior: **upsert**
  * Write batch size: **1000**
* **Mapping** Tab:
  * Click **Import Schemas**
  * Click **New Mapping**
    * Select source / destination columns as appropriate
* Click **Validate Pipeline** at top of screen
* Choose **Debug** - at top of screen

Cosmos:

Firewall tab -> all networks

: