Chatta subito con l'assistenza
Chat con il supporto

SharePlex 12.1 - Release Notes

System Requirements and Conditions of Support When Replicating from PostgreSQL

This chapter contains the requirements when capturing from a PostgreSQL database and replicating to supported target databases.

PostgreSQL source basics

This section contains the requirements for a PostgreSQL database as a source when capturing from a PostgreSQL database and replicating to supported target databases.

Supported operating systems

For information regarding supported source and target operating systems with specific database versions, see the Compatibility Matrix for Supported Operating Systems and Target Platforms section. To view compatibility across operating systems for remote replication, see Compatibility Across Operating Systems (Remote Collection and Posting).

Supported versions and ODBC requirements

  • PostgreSQL 13.x, 14.x, 15.x, 16, and 17

  • Enterprise Database (EDB) 15.x, 16

  • Fujitsu Enterprise PostgreSQL 15 SP2, 16, and 17

  • AlloyDB Omni 15.5.x

ODBC driver requirement:

ODBC drivers for Community Edition: postgresql13-odbc-13.02.0000, postgresql14-odbc-13.02.0000,postgresql15-odbc-16.00.0000, and postgresql16-odbc-16.00.0000

ODBC drivers for Enterprise Edition (EDB): edb-odbc-13.02.0000 and edb-odbc-16.00.0000.01

Supported cloud platforms and databases

Amazon Web Services:

  • Amazon EC2 (IaaS) Virtual Machine

  • Amazon RDS for PostgreSQL

  • Amazon Aurora

Microsoft Azure:

  • Azure Virtual Machine (IaaS)

  • Azure Flexible Server

Google Cloud Platform:

  • Cloud SQL for PostgreSQL

  • AlloyDB for PostgreSQL

PostgreSQL source basic conditions of support

Supported / non-supported SharePlex features for PostgreSQL as a source

The following table shows whether specific SharePlex features are supported for replication from PostgreSQL to Oracle and open targets .

SharePlex feature Supported targets
reconcile command (target instantiation) Oracle, PostgreSQL
compare/compare using and repair/repair using commands PostgreSQL
copy/copy using and append/append using commands Not supported
Hash horizontally partitioned replication PostgreSQL
Column-based horizontally partitioned replication PostgreSQL
Vertically partitioned replication PostgreSQL, Oracle
Column mapping PostgreSQL, Oracle, SQL Server, Kafka, Snowflake
Key definition PostgreSQL, Oracle
Build configuration with scripts Not supported
Named queues PostgreSQL, Oracle, SQL Server, Kafka, Snowflake
Commit Reduction (feature of Post Enhanced Performance) PostgreSQL, Oracle, SQL Server, Kafka, Snowflake
Dependency Checking (feature of Post Enhanced Performance) Not supported
Transformation Oracle
Conflict resolution

PostgreSQL, Oracle

Peer-to-peer replication (bi-directional) PostgreSQL, Oracle
Consolidated replication (many to one) PostgreSQL, Oracle, SQL Server, Kafka, Snowflake
Broadcast replication (one to many) PostgreSQL, Oracle, SQL Server, Kafka, Snowflake
High availability replication (active/passive bi-directional) Not supported
Change tracking target (CDC) Not supported
Data encryption Not supported
Data compression Not supported
SSH PostgreSQL, Oracle, SQL Server, Kafka, Snowflake
auth_hosts file Not supported
Monitoring scripts Not supported
SNMP monitoring Not supported
Continue posting on error (SP_OPX_CONT_ON_ERR) PostgreSQL, SQL Server, Kafka, Snowflake
Suspend on out of sync errors (SP_OPX_OUT_OF_SYNC_SUSPEND) PostgreSQL, SQL Server, Kafka, Snowflake
Reduced key (SP_OPX_REDUCED_KEY) PostgreSQL, SQL Server, Kafka, Snowflake
Logical Transaction Rollback on out-of-sync transactions Not supported
Tables without key1 PostgreSQL, Oracle, SQL Server, Kafka, Snowflake
Special characters2 PostgreSQL, Oracle, SQL Server, Kafka, Snowflake

Conditions of support:

  1. Data inconsistency is likely to occur in replication or compare-repair processes when SharePlex replication involves tables containing non-key columns with duplicate data.

  2. SharePlex supports special characters like hyphen (-), underscore (_), and plus (+) in PostgreSQL database names, as well as a dot (.), but other PostgreSQL objects, such as column names, do not support all special characters.

Supported and non-supported key features for physical and logical slots

The table below presents information on the metrics that are supported and not supported for physical and logical slots in on-premises and PostgreSQL Database as a Service environments.

Note: SharePlex supports logical replication with pgoutput plugin only.

SharePlex Features PostgreSQL On-Prem/Cloud VM/EDB (Enterprise Database) PGDB as a Service AWS- RDS and Aurora, Azure Flexi server
Physical Replication Logical Replication (pgoutput plugin) Logical Replication (pgoutput plugin)
Activate with LSN Supported Not supported Not supported
HA Cluster Replication with failover Applicable with On-prem PostgreSQL database CrunchyData* Not supported Applicable only with Azure Flexi server using the pg_failover_slots extension; it is not supported with AWS-RDS and Aurora.
Horizontal and Vertical Partitioning Supported Supported Supported
Utilities and Commands Supported Supported Supported
Database level Table Partitioning Supported Supported Supported
Bi-Directional Replication Supported Supported Supported
DDL Support Not supported Not supported Not supported
Replication Performance No Lag Minimal Lag Minimal Lag
Insert using PostgreSQL Copy command Not supported Not supported Not supported

* The HA Cluster Replication with Failover feature is not supported with PostgreSQL Enterprise Database (EDB).

LIMITATIONS:

The following limitations are observed related to the TRUNCATE operation:

  • Applicable only for physical replication. If the truncate operation is performed on a table in replication, the Capture process ignores the DML operations performed on the table until the commit for the truncate operation is received or processed by the Capture process. Therefore, TRUNCATE should either be performed with auto-commit enabled or both TRUNCATE and COMMIT should be executed without any DML operations in between on the table.

  • Applicable only for logical replication. In bi-directional replication, the bulk delete operations performed immediately after a TRUNCATE operation are processed with some delay by the Capture processes. There is no data inconsistency once the bulk delete operations are processed.

PostgreSQL to Kafka replication

SharePlex supports replication from PostgreSQL to Kafka, according to the details below.

Supported operating systems

For information regarding supported source and target operating systems with specific database versions, see the Compatibility Matrix for Supported Operating Systems and Target Platforms section. To view compatibility across operating systems for remote replication, see Compatibility Across Operating Systems (Remote Collection and Posting).

Supported Kafka targets

  • Apache Kafka 3.6.1 and 3.8

  • Confluent Kafka 7.6

We test and reproduce issues against:

  • Apache Kafka

  • Confluent Kafka

NOTE: If you would like to use a different vendor who provides their own Kafka distribution or provides a Kafka compliant interface, you are welcome to do so. From a support perspective, our support and development teams will share any Kafka broker errors that SharePlex receives and also ensure SharePlex itself is working properly to capture data, the target configuration formats in SharePlex are correct, and the SharePlex poster is posting or attempting to post according to the Kafka versions under support. If there are SharePlex issues with Kafka support that can be reproduced on Apache Kafka or otherwise demonstrated to be definitely associated with SharePlex, our team will address those as standard Kafka issues.

Supported formats

XML format supported data types

SharePlex supports the following data types for replication from PostgreSQL to Kafka in XML format:

PostgreSQL Kafka (XML format)
SMALLINT Decimal
INT Decimal
BIGINT Decimal
REAL Decimal
DOUBLE PRECISION Decimal
NUMERIC Decimal
CHAR String
VARCHAR String
TEXT String
BYTEA Base64Binary
DATE DateTime
TIME Time
TIME WITH TIME ZONE Time
TIMESTAMP DateTime
TIMESTAMP WITH TIME ZONE DateTimeStamp
BOOLEAN String
JSON/JSONB String
JSON format supported data types

SharePlex supports the following data types for replication from PostgreSQL to Kafka in JSON format:

PostgreSQL Kafka (JSON format)
SMALLINT Number
INT Number
BIGINT Number
REAL Number
DOUBLE PRECISION Number
NUMERIC Number
CHAR String
VARCHAR String
TEXT String
BYTEA String containing base64 encoded data
DATE String containing date
TIME String containing time
TIME WITH TIME ZONE String containing time
TIMESTAMP String containing timestamp
TIMESTAMP WITH TIME ZONE String containing timestamp
BOOLEAN String
JSON/JSONB String

Supported DML operations

SharePlex supports DML operations on tables. DML can be replicated from a PostgreSQL source to a kafka target.

Tables

SharePlex supports the following DML operations on tables:

  • Singular and bulk operation - Insert
  • Singular and bulk operation - Update with key(s)/ non key(s)
  • Singular and bulk operation - Delete with key(s)
  • Above operations with rollback/savepoint

Supported operations and objects for DDL replication

Currently, SharePlex does not support DDL operations when replicating data from a PostgreSQL source to a Kafka target. It only supports tables as objects for data replication.

PostgreSQL to PostgreSQL replication

SharePlex supports replication from PostgreSQL to PostgreSQL , according to the details below.

Supported target operating systems

For information regarding supported source and target operating systems with specific database versions, see the Compatibility Matrix for Supported Operating Systems and Target Platforms section. To view compatibility across operating systems for remote replication, see Compatibility Across Operating Systems (Remote Collection and Posting).

Supported target versions and ODBC requirements

  • PostgreSQL 13.x, 14.x, 15.x, 16, and 17

  • Enterprise Database (EDB) 15.x, 16

  • Fujitsu Enterprise PostgreSQL 15 SP2, 16, and 17

  • AlloyDB Omni 15.5.x

ODBC driver requirement:

ODBC drivers for Community Edition: postgresql13-odbc-13.02.0000, postgresql14-odbc-13.02.0000,postgresql15-odbc-16.00.0000, and postgresql16-odbc-16.00.0000

ODBC drivers for Enterprise Edition (EDB): edb-odbc-13.02.0000 and edb-odbc-16.00.0000.01

Supported target cloud platforms

Amazon Web Services:

  • Amazon EC2 (IaaS) Virtual Machine

  • Amazon RDS for PostgreSQL

  • Amazon Aurora

Microsoft Azure:

  • Azure Virtual Machine (IaaS)

  • Azure Flexible Server

Google Cloud Platform:

  • Cloud SQL for PostgreSQL

  • AlloyDB for PostgreSQL

Supported characterset

UTF 8

Supported data types

SharePlex PostgreSQL supports the following data types for replication from PostgreSQL to PostgreSQL:

  • CHAR
  • VARCHAR
  • TEXT
  • SMALLINT
  • INT
  • BIGINT
  • NUMERIC
  • REAL
  • DOUBLE PRECISION
  • DATE
  • TIMESTAMP (0)
  • TIMESTAMP
  • TIMESTAMP WITH TIME ZONE
  • BOOLEAN
  • BYTEA
  • TIME
  • TIME WITH TIME ZONE
  • JSON
  • JSONB

Limitations:

  • The JSON & JSONB data types are supported only with logical replication.

  • Columns with the JSON and JSONB data types are not supported when defining a unique key.

  • Partition keys defined on JSON and JSONB data type columns are not supported for replicating partitioned table data.

  • A table containing columns with data types such as VARCHAR > 2k, VARCHAR without specified length, TEXT, BYTEA, JSON, and JSONB, without a defined key, and where basic data type columns are non-key columns with duplicate data across multiple rows, can result in data inconsistency during the Repair process.

Supported DML operations

SharePlex supports DML operations on tables. DML can be replicated from a PostgreSQL source to a PostgreSQL target.

Tables

SharePlex supports the following DML operations on tables:

  • Singular and bulk operation - Insert
  • Singular and bulk operation - Update with key(s)/ non key(s)
  • Singular and bulk operation - Delete with key(s)
  • Above operations with rollback/savepoint

The target table must support the column types that are being replicated from the source PostgreSQL table.

Supported operations and objects for DDL replication

Currently, SharePlex does not support DDL operations when replicating data from a PostgreSQL source to a PostgreSQL target. It only supports tables as objects for data replication.

Related Documents

The document was helpful.

Seleziona valutazione

I easily found the information I needed.

Seleziona valutazione