Synopsis

Use the ROLLBACK TO SAVEPOINT statement to revert the state of the transaction to a previously established savepoint. This can be particularly useful to handle and unwind errors like key/index constraint violations.

Syntax

savepoint_rollback ::= ROLLBACK [ WORK | TRANSACTION ] TO 
                       [ SAVEPOINT ] name

savepoint_rollback

ROLLBACKWORKTRANSACTIONTOSAVEPOINTname

Semantics

begin

ROLLBACK [ WORK | TRANSACTION ] TO [ SAVEPOINT ] name

NAME

The name of the savepoint to which you wish to roll back.

Examples

Create a sample table and add one row to start.

CREATE TABLE sample(k int PRIMARY KEY, v int);
INSERT INTO sample(k, v) VALUES (1, 2);

Begin a transaction and insert some rows.

BEGIN TRANSACTION;
INSERT INTO sample(k, v) VALUES (3, 4);

Now, create a savepoint before inserting a duplicate row for k=1:

SAVEPOINT test;
INSERT INTO sample(k, v) VALUES (1, 3);

You should get the following error:

ERROR:  duplicate key value violates unique constraint "k_pkey"

Any other operations should error, since the transaction is now in a bad state:

SELECT * FROM sample;
ERROR:  current transaction is aborted, commands ignored until end of transaction block

However, you can roll back to our earlier savepoint and continue with the transaction without losing our earlier insert:

ROLLBACK TO test;
INSERT INTO sample(k, v) VALUES (5, 6);
COMMIT;

If you check the rows in the table, you will see the row you inserted before the primary key violation, as well as the one you inserted after roll back:

SELECT * FROM sample;
 k  | v
----+----
  1 |  2
  3 |  4
  5 |  6
(3 rows)

See also