SQL Server

The SQL Server source reads tables from one Microsoft SQL Server database and loads each table in full, once. Later sync runs skip tables that are already loaded, so new and changed rows are not picked up and deletes are not captured. Use it for migrations and one-off copies of a database into a warehouse.

Skippr, your data department

Speak to an expert. Upfront insights in days, not months.

Book an intro call →

20 minutes · No sales pitch

Before you begin

• Network access. The machine running Skipprd must reach the server on its port (default 1433).
• No TLS. Skipprd connects to SQL Server without encryption, including the login. The server must not force encryption (Force Encryption off in SQL Server Configuration Manager), and you should run Skipprd on a private network or through a tunnel. Azure SQL Database always requires encryption, so this source cannot connect to it.
• A SQL login with read access. Skipprd lists tables from INFORMATION_SCHEMA and runs SELECT * on each one. db_datareader covers both:

CREATE LOGIN skippr WITH PASSWORD = 'Change-me-123!';
USE MyDB;
CREATE USER skippr FOR LOGIN skippr;
ALTER ROLE db_datareader ADD MEMBER skippr;

Configure

The connection string holds the password, so keep the whole string in an environment variable. Skipprd also reads .env and .env.local next to skippr.yml.

export MSSQL_CONNECTION_STRING="server=tcp:sql.internal,1433;database=MyDB;user id=skippr;password=Change-me-123!"

Python

from skippr import Config, DataSourceMssql, EnvRef, Pipeline

cfg = Config.discover()
src = cfg.data_source(
    "sqlserver",
    DataSourceMssql(
        connection_string=EnvRef("MSSQL_CONNECTION_STRING"),
        tables=["dbo.customers", "dbo.orders"],
    ),
)
pipe = cfg.pipeline("sqlserver", Pipeline(data_source=src))
cfg.save()

CLI

skipprd connect data-source mssql \
  --pipeline sqlserver \
  --name sqlserver \
  --connection-string '${MSSQL_CONNECTION_STRING}' \
  --tables dbo.customers \
  --tables dbo.orders

YAML

pipelines:
  sqlserver:
    data_source: data_sources.sqlserver

data_sources:
  sqlserver:
    Mssql:
      connection_string: ${MSSQL_CONNECTION_STRING}
      tables: ["dbo.customers", "dbo.orders"]

Omit tables to load every base table in the database.

Add a destination to the pipeline (see the connector catalog), then discover the schema and run one sync pass:

Python

from skippr import Session

session = Session(pipe)
session.discover()
session.sync(once=True)

CLI

skipprd discover --pipeline sqlserver --log
skipprd sync --pipeline sqlserver --once --log

To try the source locally, start SQL Server in Docker and point the connection string at it:

docker run -e 'ACCEPT_EULA=Y' -e 'MSSQL_SA_PASSWORD=YourStrong!Passw0rd' \
  -p 1433:1433 --name mssql-dev \
  -d mcr.microsoft.com/mssql/server:2022-latest

export MSSQL_CONNECTION_STRING="server=tcp:127.0.0.1,1433;database=master;user id=sa;password=YourStrong!Passw0rd"

Options

Key	Type	Default	Description
connection_string	${ENV} reference	Required	ADO.NET-style connection string: server=tcp:<host>,<port>;database=<db>;user id=<login>;password=<password>.
tables	list of strings	every base table in the database	Tables to read, as schema.table, or table for the dbo schema.
batch_size_rows	integer	10000	Rows per batch handed to Skipprd.
query_timeout_seconds	integer	none	Not used by this source.
format	string	json	Leave unset. Rows are always sent as JSON.
batch_size_bytes	integer	none	Not used by this source.
batch_size_seconds	integer	none	Not used by this source.

What gets synced

Tables. The tables in tables, or every base table in INFORMATION_SCHEMA.TABLES when you omit it. Views are not included in discovery. Skipprd reads each table with one SELECT * and holds the result in memory before writing it, so size the machine for your largest table.

Table names. Each source table lands in a destination table named after the table alone, without the schema: dbo.customers becomes customers. Skipprd lowercases names and replaces characters other than letters, digits, and _ with _. Tables with the same name in different schemas land in the same destination table, so split them across pipelines if you need them apart.

Types.

SQL Server type	Arrives as
tinyint, smallint, int, bigint, real, float	number
decimal, numeric	string with the column's scale, for example "120.50"
money, smallmoney	string with four decimal places, for example "9.9900"
bit	boolean
char, varchar, nchar, nvarchar, text, ntext, xml	string
uniqueidentifier	string
date	string, YYYY-MM-DD
time	string, HH:MM:SS with fractional seconds
datetime, datetime2, smalldatetime	string, YYYY-MM-DDTHH:MM:SS with fractional seconds, no time zone
datetimeoffset	string in RFC 3339 with the offset, for example 2025-01-03T09:15:00+01:00
binary, varbinary, image	array of byte values

Decimal values arrive as strings so no precision is lost. Run skipprd discover to infer column types; see Schema discovery and evolution.

Incremental behaviour. Skipprd records each table as loaded once its rows are durably written. Later runs skip it, even if rows changed. To reload the database, run the source under a new pipeline name, which starts with fresh progress.

Deletes. Not captured.

Troubleshooting

Run sync with --log to see these messages. A connection failure is logged, and the run ends without loading anything.

Symptom	Cause	Fix
Failed to connect to MSSQL in the log	Wrong server, port, or credentials, or no network path.	Check the connection string and that port 1433 (or yours) is reachable from the machine running Skipprd.
Login failed for user	Wrong login or password, or SQL authentication is disabled.	Check the login, and enable SQL Server and Windows Authentication mode on the server.
The connection is refused during login with an encryption error	The server forces encryption, or it is Azure SQL Database.	Turn off Force Encryption for this server on a private network. Encrypted connections are not supported by this source.
Failed to query table ... in the log, and that table is missing	The name is wrong or the login lacks SELECT.	Fix the entry in tables, or add the login to db_datareader. Other tables still load.
Two source tables end up in one destination table	They share a name in different schemas.	Put each schema in its own pipeline.
Changed rows never arrive	Each table loads once.	Run the source under a new pipeline name to take a fresh copy.

Next steps

• Connector catalog — pick a destination
• skippr.yml — the full project file, with a SQL Server to Snowflake example
• skipprd sync — run once or on a schedule