You can migrate your databases and data to SQL Server vNext CTP1 running on Linux. The method you choose to use depends on the source data and your specific scenario. The following sections provide best practices for various migration scenarios.

Migrate from SQL Server on Windows

1. Create a backup of the database on the Windows machine.
2. Transfer the backup file to the target SQL Server Linux machine.
3. Restore the backup on the Linux machine.

Migrate from other database servers

Migrate structured data

Restore a SQL Server database from Windows to Linux

• Download the AdventureWorks backup file on a Windows machine
• Transfer the backup to your Linux machine
• Restore the database using Transact-SQL commands

Download the AdventureWorks database backup

Transfer the backup file to Linux

1. For Windows, install a Bash shell. There are several options, including the following:
   • Download an open source Bash shell, such as PuTTY.
   • Or, on Windows 10, use the new built-in Bash shell (beta).
   • Or, if you work with Git, use the Git Bash shell.
2. Open a Bash shell (terminal) and navigate to the directory containing AdventureWorks2014.bak.
3. Use the scp (secure copy) command to transfer the file to the target Linux machine. The following example transfers AdventureWorks2014.bak to the home directory of user1 on the server named linuxserver1.

sudo scp AdventureWorks2014.bak user1@linuxserver1:./

Move the backup file

sudo su

mkdir -p /var/opt/mssql/backup

mv /home/user1/AdventureWorks2014.bak /var/opt/mssql/backup/

exit

Restore the database backup

sqlcmd -S localhost -U SA

RESTOREDATABASE AdventureWorks
FROM DISK = '/var/opt/mssql/backup/AdventureWorks2014.bak'
WITHMOVE'AdventureWorks2014_Data'TO'/var/opt/mssql/data/AdventureWorks2014_Data.mdf',
MOVE'AdventureWorks2014_Log'TO'/var/opt/mssql/data/AdventureWorks2014_Log.ldf'
GO

USE AdventureWorks
GO

Run the following query that lists the top 10 products in the Production.Products table.

SELECT TOP 10Name, ProductNumber FROM Production.Product ORDERBYName
GO

Export and import a database on Linux with sqlpackage

This topic shows how to the sqlpackage tool to export and import databases to SQL Server vNext CTP1 on Linux.

Overview

The sqlpackage command line utility can be used to perform a number of operations on databases such as:

• export a database (database schema and user data) to a .bacpac file
• import a database (database schema and/or user data) from a .bacpac file
• create a database snapshot (.dacpac) file for a database
• incrementally update a database schema to match the schema of a source .dacpac file
• create a Transact-SQL incremental update script that updates the schema of a target database to match the schema of a source database

You can use the sqlpackage command line utility with Microsoft SQL Server running on-premises or in the cloud, on Linux, Windows or Docker and from Azure SQL Database. The sqlpackage command line utility is particularly useful when you wish to import a database from Azure SQL Database to SQL Server on Linux or vice-versa.

sqlpackage location

The sqlpackage command line utility is in the following location after you install SQL Server on Linux. Running sqlpackage with no command line parameters displays help about the available parameters:

/opt/mssql/bin/sqlpackage

You can specify command line parameters for a number of actions along with action specific parameters and properties as follows:

/opt/mssql/bin/sqlpackage {parameters}{properties}{SQLCMD Variables} 
/opt/mssql/bin/sqlpackage /Action:Extract /SourceServerName:tcp: /SourceDatabaseName: /TargetFile:

/opt/mssql/bin/sqlpackage /a:Export /ssn:tcp: /sdn: /tf:

/opt/mssql/bin/sqlpackage /a:Export /ssn:tcp: /sdn: /su: /sp: /tf:

Import a database (schema and user data) from a .BACPAC file

/opt/mssql/bin/sqlpackage /a:Import /tsn:tcp: /tdn: /tu: /tp: /sf:

Export and import a database on Linux with SSMS

Export a database with SSMS

• Connect to your source database in Object Explorer. The source database can be in Microsoft SQL Server running on-premises or in the cloud, on Linux, Windows or Docker and Azure SQL Database or Azure SQL Data Warehouse.
• Right-click the source database in the Object Explorer, point to Tasks, and click Export Data-Tier Application...
• In the export wizard, click Next, and then on the Settings tab, configure the export to save the BACPAC file to either a local disk location or to an Azure blob.
• By default, all objects in the database are exported. Click the Advanced tab and choose the database objects that you wish to export.
• Click Next and then click Finish.

The *.BACPAC file is successfully created at the location you chose and you are ready to import it into a target database.

Import a database with SSMS

• Start SSMS by typing Microsoft SQL Server Management Studio in the Windows search box, and then click the desktop app.

• Connect to your target server in Object Explorer. The target server can be Microsoft SQL Server running on-premises or in the cloud, on Linux, Windows or Docker and Azure SQL Database or Azure SQL Data Warehouse.
• Right-click the Databases folder in the Object Explorer and click Import Data-tier Application...
• To create the database in your target server, specify a BACPAC file from your local disk or select the Azure storage account and container to which you uploaded your BACPAC file.
• Provide the New database name for the database. If you are importing a database on Azure SQL Database, set the Edition of Microsoft Azure SQL Database (service tier), Maximum database size, and Service Objective (performance level).
• Click Next and then click Finish to import the BACPAC file into a new database in your target server.

The *.BACPAC file is imported to create a new database in the target server you specified.

Use SSMA to migrate databases to SQL Server on Linux

This topic introduces SQL Server Migration Assistant (SSMA) that helps you easily migrate databases to SQL Server vNext CTP1 on Linux from Microsoft Access, DB2, MySQL, Oracle and Sybase. SSMA is a Windows application, so use SSMA when you have a Windows machine that can connect to a remote SQL Server instance on Linux.

SSMA supports a variety of source databases including Oracle, MySQL, Sybase, DB2 and Microsoft Access to SQL Server vNext CTP1 on Linux and helps automate migration tasks such as:

• assess your source database
• convert the source database schema to Microsoft SQL Server schema
• migrate the schema
• migrate the data
• test the migration

To get started, download SQL Server Migration Assistant (SSMA) for your source database from the list below:

• SSMA for Access
• SSMA for DB2
• SSMA for MySql
• SSMA for Oracle
• SSMA for Sybase ASE

Bulk copy data with bcp to SQL Server on Linux

This topic shows how to use the bcp command line utility to bulk copy data between an instance of SQL Server vNext CTP1 on Linux and a data file in a user-specified format.

You can use bcp to import large numbers of rows into SQL Server tables or to export data from SQL Server tables into data files. Except when used with the queryout option, bcp requires no knowledge of Transact-SQL. The bcp command line utility works with Microsoft SQL Server running on-premises or in the cloud, on Linux, Windows or Docker and Azure SQL Database and Azure SQL Data Warehouse.

This topic will show you how to:

• Import data into a table using the bcp in command
• Export data from a table uisng the bcp out command

Install the SQL Server command-line tools

bcp is part of the SQL Server command-line tools, which are not installed automatically with SQL Server on Linux. If you have not already installed the SQL Server command-line tools on your Linux machine, you must install them.

Import data with bcp

In this tutorial, you will create a sample database and table on the local SQL Server instance (localhost) and then use bcp to load into the sample table from a text file on disk.

Create a sample database and table

Let's start by creating a sample database with a simple table that will be used in the rest of this tutorial.

1. On your Linux box, open a command terminal.
2. Copy and paste the commands below into the terminal window. These commands use the sqlcmd command line utility to create a sample database (BcpSampleDB) and a table (TestEmployees) on the local SQL Server instance (localhost). Remember to replace the username and  as necessary before running the commands.

Create the database BcpSampleDB:

sqlcmd -S localhost -U sa -P  -Q "CREATE DATABASE BcpSampleDB;"




Create the table TestEmployees in the database BcpSampleDB:

sqlcmd -S localhost -U sa -P -d BcpSampleDB -Q "CREATE TABLE TestEmployees (Id INT IDENTITY(1,1) NOT NULL PRIMARY KEY, Name NVARCHAR(50), Location NVARCHAR(50));"

Create the source data file

cat > ~/test_data.txt << EOF
1,Jared,Australia
2,Nikita,India
3,Tom,Germany
EOF

cat ~/test_data.txt

1,Jared,Australia
2,Nikita,India
3,Tom,Germany

Import data from the source data file

bcp TestEmployees in ~/test_data.txt -S localhost -U sa -P -d BcpSampleDB -c -t  ','

• -S: specifies the instance of SQL Server to which to connect
• -U: specifies the login ID used to connect to SQL Server
• -P: specifies the password for the login ID
• -d: specifies the database to connect to
• -c: performs operations using a character data type
• -t: specifies the field terminator. We are using comma as the field terminator for the records in our data file

sqlcmd -S localhost -d BcpSampleDB -U sa -P  -I -Q "SELECT * FROM TestEmployees;"

Id          Name                Location
----------- ------------------- -------------------
          1 Jared               Australia
          2 Nikita              India
          3 Tom                 Germany

(3 rows affected)

Export data with bcp

bcp TestEmployees out ~/test_export.txt -S localhost -U sa -P -d BcpSampleDB -c -t ','

cat ~/test_export.txt

1,Jared,Australia
2,Nikita,India
3,Tom,Germany

Conclusion