Usage Instructions

Last updated:2025-06-06 15:01:40

Usage Instructions

Last updated: 2025-06-06 15:01:40

Category
Description
Supported Versions
1. Source database
Self-built database PostgreSQL 10-16
Third-party cloud vendors (Alibaba Cloud RDS PostgreSQL/AWS RDS PostgreSQL/Huawei Cloud RDS for PostgreSQL) 10-16
TencentDB for PostgreSQL 10-16 (Migration between same master accounts / Migration between different main accounts)
2. Target database
TencentDB for PostgreSQL 10-16
Source database impact
1. DTS occupies certain source instance resources when executing full data synchronization, which may lead to an increase in the load of the source instance and add pressure on the database itself. If your database configuration is too low, it is recommended that you perform operations during business off-peak hours. 
2. Defaults to a lock-free method. No global lock is applied to the source database during the sync process. 
3. When performing data synchronization, DTS uses the account executing the synchronization task to create postgres.public.identify and {database}.__tencentdb__sync_*.* in the source database to record task status and configuration information. The target instance uses the {database}.__tencentdb__ schema to record synchronization status information. postgres.public.identify, {database}.__tencentdb__sync_*.*, and {database}.__tencentdb__ schema occupy very little space, about one-thousandth to one-ten-thousandth of the source database storage space. Therefore, they have no impact on the performance of the source database and do not preempt resources.
4. The sync task depends on the max_replication_slots parameter (one replication_slot is occupied when configuring a sync task for each database). You can view the usage status by using the SQL statement "select * from pg_replication_slots". If the number of configured sync databases + replication_slots > max_replication_slots, creating a new sync task will fail. Users are advised to end the task proactively before the system releases the replication_slots (replication_slots will not be released if the failed task is not ended).
sync object
1. Unsupported object synchronization: ROLE, PROCEDURE, TRIGGER, FUNCTION, RULE.
2. No support for synchronization of public objects and permission changes.
3. Foreign key import is not supported during structure initialization in the full-load phase. Foreign key constraints will be automatically removed before import.
4. Data sync does not support the FOREIGN TABLE table type. Export and import do not support TEMPORARY TABLE in the full-load phase. Only UNLOGGED TABLE data structure sync is supported in the incremental synchronization phase.
5. The source database updates SEQUENCE objects in the incremental phase. After synchronization to the target database, refreshing the last_value of the SEQUENCE is not supported. Users may need to manually check the sync result and modify the parameter value on the target database.
6. After the objects in the source database are synchronized to the target end, the object owner will be changed to the target database account. For example, if the owner of table 1 was A before, and account B is used for synchronization in the target database, the owner of table 1 will be changed to B after being written into the target database.
7. Cannot select the entire source instance for synchronization. If there is an operation to create a new database (CREATE DATABASE) on the source side in subsequent synchronization tasks, it is recommended to create the repository object on the target side before starting the synchronization task.
8. If there are CREATE TABLE or RENAME TABLE operations on the source during the sync task process, please select the entire database where the table is located when selecting the sync object in the configuration phase. Otherwise, the newly created table or the renamed table will not be synchronized to the target end.
Data Type
1. Data types supported
Aclitem,Bit,Bool,Box,Bpchar,Bytea,Char,Cid,Cidr,Circle,Date,Float4,Float8,Inet,Int2,Int4,Int8,Interval,Json,Jsonb,Jsonpath,Line,Lseg,Macaddr,Name,Numeric,Oid,Path,Point,Polygon,Record,Text,Tid,Time,Timestamp,Timestamptz,Unknown,Uuid,Varbit,Varchar,XID,Daterange,Int4range,Int8range,Numrange,Tsrange,Tstzrange,Datemultirange,Int4multirange,Int8multirange,Nummultirange,
Tsmultirange,Tstzmultirange,XML,Money,Time with time zone,Tsvector,Tsquery,pg_lsn,txid_snapshot.
2. Time zone
Fields with time zone (timestamp with time zone) in the source instance will automatically convert to the corresponding value of the target instance time zone after synchronizing to the target instance.
Supported SQL
1. DML
INSERT,UPDATE,DELETE
2. DDL
DROP DATABASE,ALTER DATABASE
CREATE SCHEMA,DROP SCHEMA
CREATE TABLE,ALTER TABLE,DROP TABLE
CREATE VIEW,DROP VIEW
CREATE INDEX,DROP INDEX
CREATE SEQUENCE,DROP SEQUENCE
Note:
CREATE TABLE table name AS SELECT statement is not currently supported.
SQL Filter Policy with Where condition filtering is not currently supported.
Limits of Operation
Do not perform the following operations during the sync process, otherwise the synchronization task will fail.
1. Do not modify or delete user information (including username, password and permissions) and port number in the source database and target database.
2. During the incremental synchronization phase, concurrent DDL and DML operations may cause data conflicts. It is recommended to reduce DDL during the incremental phase.
3. Do not perform the operation of purging WAL on the source database.
4. Do not delete the system database and schema postgres.public.identify, __tencentdb__sync_*.*, {database}.__tencentdb__ schema during the synchronous increment stage.
5. When the source and target database types are Tencent Cloud Database, do not modify the network of the database instance. If the network of the source instance and the target instance changes, causing the original VIP to be deleted, the synchronization task to fail, or to remain running unchanged (at this point, the synchronization is no longer working), in this case, the synchronization task cannot be retried. You need to manually end the task and then reestablish a new synchronization task.
Data Synchronization Description
1. When DTS initiates a data synchronization task, it sets the REPLICA IDENTITY FULL attribute for tables awaiting synchronization and newly created tables (for tasks with incremental DDL enabled). However, if there are long-running transactions on the table, this may result in a timeout during the setup process, ultimately causing the task to fail.
2. When DTS initiates a data synchronization task, if the database instance contains uncommitted long-running transactions, the creation of the replication slot may time out, thereby leading to task failure.
Source database HA 
HA occurs in the source database and migration of the primary node of the source instance occurs, causing changes in the WAL log. The original synchronization task will fail. The synchronized data in the target database needs cleaning, and then a new synchronization task needs to be established.

Was this page helpful?

You can also Contact Sales or Submit a Ticket for help.

Yes

Feedback

Category	Description
Supported Versions	1. Source database Self-built database PostgreSQL 10-16 Third-party cloud vendors (Alibaba Cloud RDS PostgreSQL/AWS RDS PostgreSQL/Huawei Cloud RDS for PostgreSQL) 10-16 TencentDB for PostgreSQL 10-16 (Migration between same master accounts / Migration between different main accounts) 2. Target database TencentDB for PostgreSQL 10-16
Source database impact	1. DTS occupies certain source instance resources when executing full data synchronization, which may lead to an increase in the load of the source instance and add pressure on the database itself. If your database configuration is too low, it is recommended that you perform operations during business off-peak hours. 2. Defaults to a lock-free method. No global lock is applied to the source database during the sync process. 3. When performing data synchronization, DTS uses the account executing the synchronization task to create postgres.public.identify and {database}.__tencentdb__sync_. in the source database to record task status and configuration information. The target instance uses the {database}.__tencentdb__ schema to record synchronization status information. postgres.public.identify, {database}.__tencentdb__sync_., and {database}.__tencentdb__ schema occupy very little space, about one-thousandth to one-ten-thousandth of the source database storage space. Therefore, they have no impact on the performance of the source database and do not preempt resources. 4. The sync task depends on the max_replication_slots parameter (one replication_slot is occupied when configuring a sync task for each database). You can view the usage status by using the SQL statement "select * from pg_replication_slots". If the number of configured sync databases + replication_slots > max_replication_slots, creating a new sync task will fail. Users are advised to end the task proactively before the system releases the replication_slots (replication_slots will not be released if the failed task is not ended).
sync object	1. Unsupported object synchronization: ROLE, PROCEDURE, TRIGGER, FUNCTION, RULE. 2. No support for synchronization of public objects and permission changes. 3. Foreign key import is not supported during structure initialization in the full-load phase. Foreign key constraints will be automatically removed before import. 4. Data sync does not support the FOREIGN TABLE table type. Export and import do not support TEMPORARY TABLE in the full-load phase. Only UNLOGGED TABLE data structure sync is supported in the incremental synchronization phase. 5. The source database updates SEQUENCE objects in the incremental phase. After synchronization to the target database, refreshing the last_value of the SEQUENCE is not supported. Users may need to manually check the sync result and modify the parameter value on the target database. 6. After the objects in the source database are synchronized to the target end, the object owner will be changed to the target database account. For example, if the owner of table 1 was A before, and account B is used for synchronization in the target database, the owner of table 1 will be changed to B after being written into the target database. 7. Cannot select the entire source instance for synchronization. If there is an operation to create a new database (CREATE DATABASE) on the source side in subsequent synchronization tasks, it is recommended to create the repository object on the target side before starting the synchronization task. 8. If there are CREATE TABLE or RENAME TABLE operations on the source during the sync task process, please select the entire database where the table is located when selecting the sync object in the configuration phase. Otherwise, the newly created table or the renamed table will not be synchronized to the target end.
Data Type	1. Data types supported Aclitem,Bit,Bool,Box,Bpchar,Bytea,Char,Cid,Cidr,Circle,Date,Float4,Float8,Inet,Int2,Int4,Int8,Interval,Json,Jsonb,Jsonpath,Line,Lseg,Macaddr,Name,Numeric,Oid,Path,Point,Polygon,Record,Text,Tid,Time,Timestamp,Timestamptz,Unknown,Uuid,Varbit,Varchar,XID,Daterange,Int4range,Int8range,Numrange,Tsrange,Tstzrange,Datemultirange,Int4multirange,Int8multirange,Nummultirange, Tsmultirange,Tstzmultirange,XML,Money,Time with time zone,Tsvector,Tsquery,pg_lsn,txid_snapshot. 2. Time zone Fields with time zone (timestamp with time zone) in the source instance will automatically convert to the corresponding value of the target instance time zone after synchronizing to the target instance.
Supported SQL	1. DML INSERT,UPDATE,DELETE 2. DDL DROP DATABASE,ALTER DATABASE CREATE SCHEMA,DROP SCHEMA CREATE TABLE,ALTER TABLE,DROP TABLE CREATE VIEW,DROP VIEW CREATE INDEX,DROP INDEX CREATE SEQUENCE,DROP SEQUENCE Note: CREATE TABLE table name AS SELECT statement is not currently supported. SQL Filter Policy with Where condition filtering is not currently supported.
Limits of Operation	Do not perform the following operations during the sync process, otherwise the synchronization task will fail. 1. Do not modify or delete user information (including username, password and permissions) and port number in the source database and target database. 2. During the incremental synchronization phase, concurrent DDL and DML operations may cause data conflicts. It is recommended to reduce DDL during the incremental phase. 3. Do not perform the operation of purging WAL on the source database. 4. Do not delete the system database and schema postgres.public.identify, __tencentdb__sync_., {database}.__tencentdb__ schema during the synchronous increment stage. 5. When the source and target database types are Tencent Cloud Database, do not modify the network of the database instance. If the network of the source instance and the target instance changes, causing the original VIP to be deleted, the synchronization task to fail, or to remain running unchanged (at this point, the synchronization is no longer working), in this case, the synchronization task cannot be retried. You need to manually end the task and then reestablish a new synchronization task.
Data Synchronization Description	1. When DTS initiates a data synchronization task, it sets the REPLICA IDENTITY FULL attribute for tables awaiting synchronization and newly created tables (for tasks with incremental DDL enabled). However, if there are long-running transactions on the table, this may result in a timeout during the setup process, ultimately causing the task to fail. 2. When DTS initiates a data synchronization task, if the database instance contains uncommitted long-running transactions, the creation of the replication slot may time out, thereby leading to task failure.
Source database HA	HA occurs in the source database and migration of the primary node of the source instance occurs, causing changes in the WAL log. The original synchronization task will fail. The synchronized data in the target database needs cleaning, and then a new synchronization task needs to be established.

tencent cloud