Convert SQL Server to MariaDB
Drop the script SSMS's Generate Scripts writes with schema and data, as .sql or .sql.gz, UTF-16 or UTF-8, and download a script for MariaDB with the same tables, keys, indexes and data. Every value MariaDB can't hold as it is, and every object it doesn't carry, is listed in the report.
18 rows converted, 3 fixes, none rejected
Tables in shop_2024_sqlserver.sqlSQL Server dump · written by SQL Server Management Objects · utf-8
orders10 rows1
11 columns id int not null customer_id int not null status varchar(10) not null total decimal(19,4) not null items tinyint unsigned not null ordered_at datetime not null shipped_at datetime(6) note text details text order_ref uuid not null total_with_tax decimal(12,2) customers8 rows2
8 columns id int not null name varchar(100) not null email varchar(200) not null country char(2) not null is_active tinyint(1) not null signed_up date created_at datetime not null row_version blob
Not converted
- View paid_orders
- Trigger orders_no_negative
Each database writes these in its own language. Rewrite them for MariaDB.
-- Written by IntoSQL from a SQL Server dump. Load it with: mariadb --default-character-set=utf8mb4 <database> < <file>
SET NAMES utf8mb4;
SET FOREIGN_KEY_CHECKS = 0;
SET UNIQUE_CHECKS = 0;
SET SQL_MODE = 'NO_AUTO_VALUE_ON_ZERO,ALLOW_INVALID_DATES';
CREATE TABLE `orders` (
`id` int NOT NULL AUTO_INCREMENT,
`customer_id` int NOT NULL,
`status` varchar(10) NOT NULL DEFAULT 'pending',
`total` decimal(19,4) NOT NULL,
`items` tinyint unsigned NOT NULL DEFAULT 1,
`ordered_at` datetime NOT NULL,
`shipped_at` datetime(6),
`note` text,
`details` text,
`order_ref` uuid NOT NULL DEFAULT (uuid()),
`total_with_tax` decimal(12,2) GENERATED ALWAYS AS (cast(`total` * (1.16) AS DECIMAL(12,2))) STORED,
PRIMARY KEY (`id`),
KEY `idx_orders_customer` (`customer_id`),
KEY `idx_orders_status_date` (`status`,`ordered_at`),
CONSTRAINT `check2` CHECK (`status` = 'cancelled' OR `status` = 'shipped' OR `status` = 'paid' OR `status` = 'pending'),
CONSTRAINT `check3` CHECK (`total` >= (0)),
FOREIGN KEY (`customer_id`) REFERENCES `customers` (`id`)
) AUTO_INCREMENT=1 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_bin;
CREATE TABLE `customers` (
`id` int NOT NULL AUTO_INCREMENT,
`name` varchar(100) NOT NULL,
`email` varchar(200) NOT NULL,
`country` char(2) NOT NULL DEFAULT 'US',
`is_active` tinyint(1) NOT NULL DEFAULT 1,
`signed_up` date,
`created_at` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP,
`row_version` blob,
PRIMARY KEY (`id`),
UNIQUE KEY (`email`)
) AUTO_INCREMENT=1 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_bin COMMENT='People who bought at least once';
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (1,'Ana Martínez','ana@example.com','MX',1,'2023-01-15','2023-01-15 10:24:00.000000');
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (2,'Zoë Müller','zoe@example.de','DE',1,'2023-02-03','2023-02-03 08:11:42.000000');
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (3,'José Peña','jose@example.es','ES',0,'2023-03-21','2023-03-21 19:05:17.000000');
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (4,'李雷','lilei@example.cn','CN',1,'2023-05-09','2023-05-09 02:40:00.000000');
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (5,'Noah O\'Brien','noah@example.ie','IE',1,NULL,'2023-06-30 14:00:00.000000');
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (6,'Amélie Laurent','amelie@example.fr','FR',1,'2024-01-02','2024-01-02 09:30:00.000000');
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (7,'Ravi Patel','ravi@example.in','IN',0,'2024-02-29','2024-02-29 23:59:59.000000');
INSERT INTO `customers` (`id`,`name`,`email`,`country`,`is_active`,`signed_up`,`created_at`) VALUES (8,'Sofia Rossi 🌻','sofia@example.it','IT',1,'2024-04-12','2024-04-12 12:12:12.000000');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (1,1,'shipped',129.9000,3,'2024-03-01T10:00:00.000000','2024-03-02T16:30:00.123456',NULL,'{"gift": false}','0b7a4c2e-3f1d-4e8a-9c55-1a2b3c4d5e6f');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (2,2,'paid',54.0000,1,'2024-03-02T11:15:00.000000',NULL,'Gift wrap 🎁','{"gift": true, "wrap": "red"}','1c8b5d3f-4a2e-4f9b-8d66-2b3c4d5e6f70');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (3,1,'cancelled',12.5000,1,'2024-03-03T09:45:00.000000',NULL,'Customer changed their mind',NULL,'2d9c6e4a-5b3f-4a0c-9e77-3c4d5e6f7081');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (4,4,'shipped',980.0000,12,'2024-03-05T02:10:00.000000','2024-03-07T08:00:00.000000','请小心轻放','{"fragile": true}','3e0d7f5b-6c4a-4b1d-8f88-4d5e6f708192');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (5,5,'pending',45.7500,2,'2024-03-08T17:20:00.000000',NULL,'Leave at the door; ring twice',NULL,'4f1e806c-7d5b-4c2e-9a99-5e6f708192a3');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (6,6,'shipped',310.4000,5,'2024-03-10T13:00:00.000000','2024-03-11T09:00:00.000000','Fragile: "glass"','{"fragile": true}','50f2917d-8e6c-4d3f-8baa-6f708192a3b4');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (7,8,'paid',76.2000,2,'2024-03-11T18:45:00.000000',NULL,NULL,NULL,'6103a28e-9f7d-4e40-9cbb-708192a3b4c5');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (8,2,'shipped',19.9900,1,'2024-03-12T08:00:00.000000','9999-12-31T23:59:59.999999','Shipped date set to the maximum by an old import',NULL,'7214b39f-a08e-4f51-8dcc-8192a3b4c5d6');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (9,3,'shipped',64.0000,2,'2024-03-14T15:30:00.000000','2024-03-15T10:00:00.000000','Two parcels','{"parcels": [1, 2]}','8325c4a0-b19f-4062-9edd-92a3b4c5d6e7');
INSERT INTO `orders` (`id`,`customer_id`,`status`,`total`,`items`,`ordered_at`,`shipped_at`,`note`,`details`,`order_ref`) VALUES (10,7,'paid',1499.0000,1,'2024-03-15T21:05:00.000000',NULL,'C:\\Users\\ravi\\invoice.pdf',NULL,'9436d5b1-c2a0-4173-8fee-a3b4c5d6e7f8');
SET UNIQUE_CHECKS = 1;
SET FOREIGN_KEY_CHECKS = 1;
Fixes
- "shipped_at" of orders is datetime2(7), with more digits of a second than MariaDB keeps, so it is datetime(6) and the last digits are cut.
- "created_at" of customers holds moments with a time zone; MariaDB has no type that keeps one, so they are stored in UTC as datetime.
- "row_version" of customers is a row version SQL Server sets by itself; the dump has no values for it, so it is blob and empty.
Needs a look
- 6 text columns compare ignoring case in the source; MariaDB compares exactly, so unique keys and searches may behave differently. Use a case-insensitive collation or column type on the columns that need it.
- 2 views, triggers, routines or events aren't converted; the report lists each one. Recreate them by hand in MariaDB.
- The check check1 on orders can't be written for MariaDB (function isjson), so it isn't created. Write the check by hand if it matters.
- The index idx_orders_pending on orders uses an expression that can't be written for MariaDB (neither MySQL 5.7 nor MariaDB index an expression), so it isn't created. Create the index by hand if queries need it.
Your file is converted on this computer and never uploaded.
Other samples:
What changes from SQL Server to MariaDB
Each row is a rule the converter applies, with the report line it writes when the rule changes something in your data.
bit
tinyint(1)
MariaDB's BOOLEAN is TINYINT(1): 1 and 0, as bit holds them.
char(n), varchar(n), nchar(n), nvarchar(n), varchar(max), text
char(n), varchar(n) up to 1024 characters, text types beyond
Text keeps every character as utf8mb4. A column in a key is varchar(768) at most, the longest a key holds.
decimal(p,s), money, smallmoney
decimal(p,s); money as decimal(19,4)
decimal holds 65 digits, more than SQL Server's 38, so every value is exact.
datetime, datetime2, smalldatetime
datetime(p)
A datetime is a wall-clock reading in both.
datetime2(7), time(7) and datetimeoffset(7)
six digits of a second, the seventh cut
MariaDB keeps microseconds; the seventh digit (100 nanoseconds) is cut, never rounded, so no value moves to the next second or day.
Report line: fraction_cut
datetimeoffset
datetime(p) in UTC
MariaDB's TIMESTAMP only reaches 2038, so moments are stored as datetime in UTC, the same instant whatever offset each value had.
Report line: moments_as_utc
time
time(p)
The same times of day.
dates before the year 1000
NULL
MariaDB stores years 1000 to 9999, and SQL Server goes back to the year 1. Those values become NULL and the report counts them.
Report line: date_out_of_range
json (SQL Server 2025)
json, a longtext with a JSON check
The same JSON values, which MariaDB's JSON functions read.
uniqueidentifier
uuid
The same identifiers, written in lowercase.
binary, varbinary, image
binary(n), varbinary(n) and blob types
Bytes stay bytes.
geometry, geography
blob with SQL Server's own bytes
SSMS writes spatial values in SQL Server's serialization, which is kept byte for byte.
Report line: spatial_as_native
rowversion, timestamp
blob, empty
SQL Server sets a row version by itself and SSMS scripts no values for it.
Report line: rowversion_empty
xml, sql_variant, hierarchyid, vector
longtext
MariaDB has no equivalent type; xml and sql_variant keep each value's text and hierarchyid its path, such as /1/2/.
Report line: type_replaced
collations that ignore case (_CI_)
exact comparison (utf8mb4_bin)
Tables use utf8mb4_bin so unique keys keep apart what SQL Server kept apart; columns that ignored case lose that. Accent rules differ between versions, so no _ci collation matches SQL Server's on every one.
Report line: case_insensitive_collation
IDENTITY
AUTO_INCREMENT, continuing after the highest id
The column that leads the primary key counts; the counter continues after the highest id.
IDENTITY columns that don't lead the primary key
a plain integer
MariaDB counts one column per table, at the start of a key; any other keeps its values without a counter.
Report line: counter_dropped
CREATE SEQUENCE
CREATE SEQUENCE with the same options
MariaDB sequences work the same way. SSMS doesn't script where a sequence has reached, so it starts where it was created.
computed columns
generated columns, STORED when PERSISTED, VIRTUAL otherwise
MariaDB computes them the same two ways, with CONCAT written so NULL counts as empty as in SQL Server.
computed columns with functions MariaDB lacks
a plain column
A formula that can't be written leaves a plain column and a report line; SSMS scripts no values for computed columns.
Report line: generated_not_translated
defaults with functions MariaDB lacks, and NEXT VALUE FOR
no default
A default that can't be written is left out and reported.
Report line: default_not_translated
CHECK constraints
translated
The same checks, tested as each row loads.
CHECK constraints with functions MariaDB lacks
not created
A check that can't be written is left out and reported.
Report line: check_skipped
CHECK constraints added WITH NOCHECK or disabled
not created
SQL Server doesn't apply these checks, so the rows may break them.
Report line: check_not_enforced
filtered and columnstore indexes
not created
IntoSQL writes b-tree indexes on whole columns; create the others again after loading.
Report line: index_skipped
foreign keys
foreign keys, loaded with FOREIGN_KEY_CHECKS=0
The rows load in any order, and the keys check new rows afterwards.
foreign keys to tables the dump doesn't create
not created
A key can't name a table that isn't there.
Report line: foreign_key_skipped
tables on a partition scheme
one table with every partition's rows
The rows are the same; partition the table again in MariaDB if it needs it.
Report line: partitions_dropped
temporal tables (SYSTEM_VERSIONING)
WITH SYSTEM VERSIONING, starting from the current rows; the history table as a plain table
MariaDB keeps the history by itself from now on; SQL Server's period columns and history table are copied as they were.
schemas
one database, tables named after their schema first
MariaDB's schemas are databases, which a script loaded into one database can't create; tables in dbo keep their names.
Report line: schemas_merged
NOT NULL columns with values that become NULL
NOT NULL dropped
A column whose values had to become NULL can't keep NOT NULL, or the load would stop.
Report line: not_null_relaxed
views, functions, procedures, triggers, synonyms, types and partition functions
listed in the report
Their code is written in T-SQL, so they are listed for you to recreate.
Report line: not_converted_objects
scripts without CREATE TABLE (data only)
types read from the values
A data-only script has only rows, so the types come from the values.
Report line: no_schema
expressions in INSERT values
NULL
A converter that doesn't run a database can't evaluate them.
Report line: expression_value
types IntoSQL doesn't know
longtext
Text holds any value; change the type after loading.
Report line: unknown_type
NUL characters in text
kept
The value is copied with the NUL character escaped.
text that isn't valid in its encoding
the bad bytes as the replacement character
The script is UTF-8 and every value has to be valid text.
Report line: invalid_utf8_replaced
text decoded twice (ñ)
copied as it is
The script holds exactly these characters, so they are copied; the report points them out.
Report line: mojibake_suspected
| In SQL Server | In MariaDB | Report line |
|---|---|---|
| bit | tinyint(1)MariaDB's BOOLEAN is TINYINT(1): 1 and 0, as bit holds them. | |
| char(n), varchar(n), nchar(n), nvarchar(n), varchar(max), text | char(n), varchar(n) up to 1024 characters, text types beyondText keeps every character as utf8mb4. A column in a key is varchar(768) at most, the longest a key holds. | |
| decimal(p,s), money, smallmoney | decimal(p,s); money as decimal(19,4)decimal holds 65 digits, more than SQL Server's 38, so every value is exact. | |
| datetime, datetime2, smalldatetime | datetime(p)A datetime is a wall-clock reading in both. | |
| datetime2(7), time(7) and datetimeoffset(7) | six digits of a second, the seventh cutMariaDB keeps microseconds; the seventh digit (100 nanoseconds) is cut, never rounded, so no value moves to the next second or day. | fraction_cut |
| datetimeoffset | datetime(p) in UTCMariaDB's TIMESTAMP only reaches 2038, so moments are stored as datetime in UTC, the same instant whatever offset each value had. | moments_as_utc |
| time | time(p)The same times of day. | |
| dates before the year 1000 | NULLMariaDB stores years 1000 to 9999, and SQL Server goes back to the year 1. Those values become NULL and the report counts them. | date_out_of_range |
| json (SQL Server 2025) | json, a longtext with a JSON checkThe same JSON values, which MariaDB's JSON functions read. | |
| uniqueidentifier | uuidThe same identifiers, written in lowercase. | |
| binary, varbinary, image | binary(n), varbinary(n) and blob typesBytes stay bytes. | |
| geometry, geography | blob with SQL Server's own bytesSSMS writes spatial values in SQL Server's serialization, which is kept byte for byte. | spatial_as_native |
| rowversion, timestamp | blob, emptySQL Server sets a row version by itself and SSMS scripts no values for it. | rowversion_empty |
| xml, sql_variant, hierarchyid, vector | longtextMariaDB has no equivalent type; xml and sql_variant keep each value's text and hierarchyid its path, such as /1/2/. | type_replaced |
| collations that ignore case (_CI_) | exact comparison (utf8mb4_bin)Tables use utf8mb4_bin so unique keys keep apart what SQL Server kept apart; columns that ignored case lose that. Accent rules differ between versions, so no _ci collation matches SQL Server's on every one. | case_insensitive_collation |
| IDENTITY | AUTO_INCREMENT, continuing after the highest idThe column that leads the primary key counts; the counter continues after the highest id. | |
| IDENTITY columns that don't lead the primary key | a plain integerMariaDB counts one column per table, at the start of a key; any other keeps its values without a counter. | counter_dropped |
| CREATE SEQUENCE | CREATE SEQUENCE with the same optionsMariaDB sequences work the same way. SSMS doesn't script where a sequence has reached, so it starts where it was created. | |
| computed columns | generated columns, STORED when PERSISTED, VIRTUAL otherwiseMariaDB computes them the same two ways, with CONCAT written so NULL counts as empty as in SQL Server. | |
| computed columns with functions MariaDB lacks | a plain columnA formula that can't be written leaves a plain column and a report line; SSMS scripts no values for computed columns. | generated_not_translated |
| defaults with functions MariaDB lacks, and NEXT VALUE FOR | no defaultA default that can't be written is left out and reported. | default_not_translated |
| CHECK constraints | translatedThe same checks, tested as each row loads. | |
| CHECK constraints with functions MariaDB lacks | not createdA check that can't be written is left out and reported. | check_skipped |
| CHECK constraints added WITH NOCHECK or disabled | not createdSQL Server doesn't apply these checks, so the rows may break them. | check_not_enforced |
| filtered and columnstore indexes | not createdIntoSQL writes b-tree indexes on whole columns; create the others again after loading. | index_skipped |
| foreign keys | foreign keys, loaded with FOREIGN_KEY_CHECKS=0The rows load in any order, and the keys check new rows afterwards. | |
| foreign keys to tables the dump doesn't create | not createdA key can't name a table that isn't there. | foreign_key_skipped |
| tables on a partition scheme | one table with every partition's rowsThe rows are the same; partition the table again in MariaDB if it needs it. | partitions_dropped |
| temporal tables (SYSTEM_VERSIONING) | WITH SYSTEM VERSIONING, starting from the current rows; the history table as a plain tableMariaDB keeps the history by itself from now on; SQL Server's period columns and history table are copied as they were. | |
| schemas | one database, tables named after their schema firstMariaDB's schemas are databases, which a script loaded into one database can't create; tables in dbo keep their names. | schemas_merged |
| NOT NULL columns with values that become NULL | NOT NULL droppedA column whose values had to become NULL can't keep NOT NULL, or the load would stop. | not_null_relaxed |
| views, functions, procedures, triggers, synonyms, types and partition functions | listed in the reportTheir code is written in T-SQL, so they are listed for you to recreate. | not_converted_objects |
| scripts without CREATE TABLE (data only) | types read from the valuesA data-only script has only rows, so the types come from the values. | no_schema |
| expressions in INSERT values | NULLA converter that doesn't run a database can't evaluate them. | expression_value |
| types IntoSQL doesn't know | longtextText holds any value; change the type after loading. | unknown_type |
| NUL characters in text | keptThe value is copied with the NUL character escaped. | |
| text that isn't valid in its encoding | the bad bytes as the replacement characterThe script is UTF-8 and every value has to be valid text. | invalid_utf8_replaced |
| text decoded twice (ñ) | copied as it isThe script holds exactly these characters, so they are copied; the report points them out. | mojibake_suspected |
Load the script into MariaDB
The script creates the tables, copies the rows, then adds keys, indexes and foreign keys, so it loads fast and stops at the first error.
mariadb --default-character-set=utf8mb4 mydb < sql_server.mariadb.sqlForeign keys are created as the dump declares them, with FOREIGN_KEY_CHECKS=0 while the rows load, as mysqldump does.
For a large dump, run the client on the same machine as the server, and raise max_allowed_packet if a single row is very large.
IntoSQL Pro loads the file straight into MariaDB from the desktop app or the command line, without writing a script.
Dump SQL Server for the best result
The converter reads what the dump holds. These options keep more of your data in it:
- In SSMS, right-click the database, Tasks, Generate Scripts, and pick the tables. dbatools, mssql-scripter and other tools built on the same library write the same script.
- Under Advanced, set Types of data to script to Schema and data. A script of the schema alone has tables without rows.
- Save it as Unicode, SSMS's default, or as UTF-8: both are read the same way, and gzip works too.
What isn't converted
Views, functions, procedures, triggers, synonyms, rules and security policies are written in T-SQL. They are listed in the report by name, never dropped silently, so you can rewrite them for MariaDB.
Questions about SQL Server to MariaDB
Not answered here? Write to us from the contact page and a person replies. Contact page
Is my dump uploaded?
No. The converter runs in your browser and reads the dump from your disk in pieces, so it never leaves your computer. The desktop app and the command line convert on your computer too.
How large can the dump be?
The converter never holds the whole dump in memory: it reads the file twice, by ranges, and writes the script as it goes. Multi-gigabyte dumps convert in the browser, in the desktop app and with the command line.
Does it read the UTF-16 file SSMS saves?
Yes. SSMS saves scripts as UTF-16 by default, and the converter reads UTF-16 and UTF-8 alike, gzipped or not, with GO between batches.
Which SQL Server versions can the script come from?
Scripts that SSMS, dbatools or mssql-scripter write from SQL Server 2017 on. Every test script comes from SQL Server 2017, 2022 and 2025 and is compared with its source cell by cell.
Are IDENTITY values kept?
Yes, for one column per table: the IDENTITY column that leads the primary key becomes AUTO_INCREMENT and continues after the highest id. Sequences become MariaDB sequences with the same options.
What happens to money, datetimeoffset and other SQL Server types?
bit becomes tinyint(1), money decimal(19,4), uniqueidentifier UUID, and datetimeoffset datetime in UTC. Temporal tables keep system versioning. Columns that ignored case compare exactly, and the report says which ones.
What about procedures, triggers and views?
They are listed in the report by name and not converted, because each database writes them in its own language.
Which MariaDB versions does the script load on?
MariaDB 10.11 and later. Every test script loads on 10.11 and 11.4, stopping at the first error if anything goes wrong.