This topic describes the release notes for the minor versions, console, Ganos, and API of PolarDB for PostgreSQL (Compatible with Oracle) and provides links to the relevant references.
You can execute the following statement to view the minor version that is used by PolarDB for PostgreSQL (Compatible with Oracle):
SHOW polardb_version;
The revision version numbers of PolarDB for PostgreSQL (Compatible with Oracle) clusters changed to a 6-digit format on February 26, 2025 (UTC+8). For more information, see Revision version numbers changed to a 6-digit format.
December 2020
Minor version 1.1.7
Category | Feature | Description | References |
New feature | Partitioned table | The Merge Partitions feature is added to merge two partitioned tables. | |
Compatibility with decoding in Oracle | Base64 decoding used in Oracle is supported. The BASE64_DECODE function can automatically identify an invalid encoded string and decode the first valid part of the string. | ||
Date format | Dates in the | None | |
SYS.***_TAB_COLS view | The | ||
polar_utility plug-in | The | ||
Stored procedure | Holdable refcursors can be returned by stored procedures. | None | |
DBMS_JOB package | The DBMS_JOB package is added. You can invoke the package or schedule it across databases by using a privileged account. | ||
Optimized feature | UTL_ENCODE | The UTL_ENCODE plug-in can be used by newly created standard users. | |
polar_agg plug-in | By default, the polar_agg plug-in is enabled. | None | |
fdw plug-in | The stability of the fdw plug-in is improved. | None | |
Fixed issue | The following issue is fixed: An error occurs when you execute | ||
The following issue is fixed: The constraints become invalid when an index scan is performed after you execute |
Console
Category | Feature | Description | References |
New feature | DBLink | A database link can be established from a PolarDB for PostgreSQL (Compatible with Oracle) database to a self-managed PostgreSQL database hosted on an Elastic Compute Service (ECS) instance. This way, you can connect a PolarDB for PostgreSQL (Compatible with Oracle) database to a PostgreSQL database. You can also connect a PostgreSQL database to a PolarDB for Oracle database by using a database link in PostgreSQL. This way, the PolarDB for Oracle database and the PostgreSQL database can communicate with each other. To create a database link in the console, choose on the cluster details page. | Create a database link from PolarDB for PostgreSQL (Compatible with Oracle) to PostgreSQL |
Ganos V3.2
Category | Feature | Description | References |
New feature | ST_AsPNG | BYTEA-type images can be returned for vector pyramids. This feature can be used to visualize vector data. | |
Raster data | The JPEG 2000 compression algorithm is supported. This algorithm is used to compress 16-bit raster data before the data is stored. | None | |
Optimized feature | st_dwithin | The ST_DWithin function is optimized to improve the performance of distance queries. | None |
Fixed issue | The following issue is fixed: The intersection of trajectories on the t axis is improper. | ||
The issues related to the memory topology indexes that are supported by Ganos are fixed. |
API
Category | Feature | Description |
Optimized feature | Database links can be created to connect PolarDB for PostgreSQL (Compatible with Oracle) databases to self-managed PostgreSQL databases that are hosted on ECS instances. |
November 2020
Minor version 1.1.6
Category | Feature | Description | References |
New feature | Compatibility | This minor version is compatible with PostgreSQL 11.9. | |
Failover | By default, the OnlinePromote feature is enabled. This feature reduces the amount of time that is consumed by a failover and improves cluster availability. | ||
DBLink | Database links can be used to connect databases across PolarDB for PostgreSQL (Compatible with Oracle) clusters. | ||
Fixed issue | The following issue is fixed: A node does not restart smoothly in some scenarios. | ||
The following issue is fixed: In some scenarios, clusters do not respond due to OOM errors. | |||
The following issue is fixed: In some scenarios, read-only nodes do not respond. | |||
The following issue is fixed: An error occurs when the UTF-8 encoding format is converted into GBK after you run the | |||
The following issue is fixed: An error occurs when the first parameter of the NVL function is left empty. |
Console
Category | Feature | Description | References |
New feature | DBLink | A database link can be created to connect a PolarDB for PostgreSQL (Compatible with Oracle) database to another PolarDB for PostgreSQL (Compatible with Oracle) database. This allows you to access data across PolarDB for PostgreSQL (Compatible with Oracle) databases. To create a database link in the console, choose on the cluster details page. |
API
Category | Feature | Description |
New feature | This operation can be called to create a database link for a PolarDB for PostgreSQL (Compatible with Oracle) cluster. | |
This operation can be called to query database links for a PolarDB for PostgreSQL (Compatible with Oracle) cluster. | ||
This operation can be called to delete a database link for a PolarDB for PostgreSQL (Compatible with Oracle) cluster. |
October 2020
Minor version 1.1.5
Category | Feature | Description | References |
New feature | Audit log | Precompiled SQL statements in audit logs can be displayed as executable SQL statements. | None |
Hash partitioning | The syntax that is used to create a specified number of hash partitioned tables is added. | ||
Partitioned table | The name of a partition or a subpartition can be specified when you query data of a specific partition. | ||
HOST parameter | The HOST parameter is supported. This parameter is used to obtain the HOST value of a client. | None | |
Optimized feature | View | The query_id field is added to optimize the polar_stat_activity view. | None |
Compatibility | The minor version is updated to 11.8. Several stability issues are fixed. | None | |
Fixed issue | The following issue is fixed: A semi-synchronization cannot time out. | ||
The following issue is fixed: The CONNECT BY clause cannot be used together with the WITH clause. |
Ganos V3.1
Category | Feature | Description | References |
New feature | Trajectory data type | GiST indexes are supported for trajectories. In addition, six operator families of different dimensions are provided to meet multi-dimensional analysis requirements. | None |
The BoxND data type is supported. This data type is used to describe the bounding box of a spatio-temporal extent or a storage trajectory. | None | ||
The following operators are supported: the && operator, the @> operator, and the <@ operator. These operators are used to compare the spatio-temporal relationships between two specified objects from various dimensions. The && operator is used to check whether the two specified objects intersect. The @> operator is used to check whether the left operand-specified object includes the right operand-specified object. The <@ operator is used to check whether the left operand-specified object is included in the right operand-specified object. | None | ||
The ST_ndIntersects, ST_ndDWithin, ST_ndContains, and ST_ndWithin functions are supported to process trajectories. | |||
Trajectory statistics can be collected. Scan costs can be estimated based on these trajectory statistics. | None | ||
TrajGist indexes are supported. This provides an optimized indexing method. | None | ||
ganos_update function | The ganos_update function is supported. You can run the | None | |
Optimized feature | Spatio-temporal range query | Spatio-temporal range queries are optimized. The query performance is improved by using optimized GiST index-based two-phase queries. | None |
Vector pyramid | Source data from all the coordinate systems that provide spatial reference identifiers (SRIDs) is supported. The source data can be tiles that are based on the EPSG:3857 or EPSG:4326 coordinate reference system. | None | |
The pixelSize parameter is supported. This parameter is used to aggregate point data and reduce the number of tiles. | None | ||
Fixed issue | The following issue is fixed: Ganos rasters fail to be updated in some scenarios. | ||
The following issue is fixed: An unexpected exit may occur after the binary files of Ganos are upgraded to later versions. | |||
The following issue is fixed: The data volume of top-level tiles is excessively large after the point data of vector pyramids is built by using the default parameters. |
September 2020
Minor version 1.1.4
Category | Feature | Description | References |
New feature | GROUP_ID function | The GROUP_ID() function is supported. | |
Partitioned table | The CREATE INDEX CONCURRENTLY statement is supported. This way, you can concurrently create indexes on partitioned tables. | None | |
Support for making indexes of standard tables invisible | The ALTER INDEX [NAME] INVISIBLE statement is supported. You can execute this statement to make an index of a standard table invisible. | None | |
ali_decoding plug-in | The ali_decoding plug-in is supported. This plug-in is used to synchronize incremental DML statements between databases. | None | |
pg_cron plug-in | The pg_cron plug-in is supported. This plug-in is used to query the records of scheduled tasks. | ||
to_char() | Strings can be specified in the TO_CHAR() function. | None | |
Data type | The following operators are added to compare the prid_ data type of the ROWID column with a string data type such as TEXT. These operators are also used to compare a string data type such as TEXT with the prid_ data type of the ROWID column.
| None | |
View | The views that display process status statistics are added. | None | |
The views that display time consumption statistics for SQL statements are added. | None | ||
Audit log | SQL errors are recorded in audit logs. | None | |
Fixed issue | The following issue is fixed: In specific scenarios, page prefetching may cause deadlocks. | ||
The following issue is fixed: A JOIN query fails to retrieve the accurate partition during partition pruning. |
Ganos V3.0
Category | Feature | Description | References |
New feature | Support for the subset raster in a NetCDF raster file that has multiple subsets | The subset raster in a NetCDF raster file that has multiple subsets can be imported. You can specify the subset name to import the subset raster. | None |
Custom metadata and time series information of rasters | The ST_MetaItems function is supported. This function is used to obtain the names of all custom metadata items. | ||
The ST_MetaData function is supported. This function is used to obtain custom metadata items of a raster and return custom metadata items in the JSON format. | |||
The ST_SetMetaData function is supported. This function is used to specify a metadata item. | |||
The ST_BeginDateTime function is supported. This function is used to obtain the start time of a raster. | |||
The ST_EndDateTime function is supported. This function is used to obtain the end time of a raster. | |||
The ST_SetBeginDateTime function is supported. This function is used to configure the start time of a raster. | |||
The ST_SetEndDateTime function is supported. This function is used to configure the end time of a raster. | |||
The ST_SetDateTime function is supported. This function is used to configure the start time and end time of a raster and the time when a band is obtained. | |||
Support for converting a raster into a BYTEA-type image | The ST_AsImage function is supported. This function is used to convert a raster into a BYTEA-type image. | ||
The ST_AsPng function is supported. This function is used to convert a raster into a BYTEA-type PNG image. | |||
The ST_AsJPEG function is supported. This function is used to convert a raster into a BYTEA-type JPEG image. | |||
Support for the GeomGrid data type and operations | The GeomGrid data type is supported. | ||
The ST_AsText function is supported. This function is used to convert a grid into a text representation. | |||
The ST_AsGeometry function is supported. This function is used to obtain the geometric representation of a grid. | |||
The ST_AsBinary function is supported. This function is used to convert a grid into a binary representation. | |||
The ST_AsBox function is supported. This function is used to obtain the box representation of a grid. | |||
The ST_AsGrid function is supported. This function is used to query the grids that intersect with a geometric object. | |||
The ST_GridFromText function is supported. This function is used to convert a grid from a text representation into a GeomGrid representation. | |||
The ST_GridFromBinary function is supported. This function is used to convert a grid from a binary representation into a GeomGrid representation. | |||
The ST_Intersects function is supported. This function is used to identify the spatial relationship between two raster objects or between a raster object and a geometric object. This function allows you to check whether the first specified object intersects with the second specified object. | |||
The ST_Contains function is supported. This function is used to identify the spatial relationship between two raster objects or between a raster object and a geometric object. This function allows you to check whether the first specified object contains the second specified object. | |||
The ST_Within function is supported. This function is used to identify the spatial relationship between two grid objects or between a grid object and a geometric object. This function allows you to check whether the first specified object is within the second specified object. | |||
Rapid display and storage of vector data | The ST_BuildPyramid function is supported. This function is used to create a vector pyramid. | ||
The ST_DeletePyramid function is supported. This function is used to delete a vector pyramid. | |||
The ST_Tile function is supported. This function is used to obtain vector data in the MVT format. | |||
Fixed issue | The following issue is fixed: In some cases, OOM errors occur when you create pyramids. | ||
The following issue is fixed: Trajectory objects on January 1, 2000 cannot be created. | |||
The following issue is fixed: In some scenarios, an invalid trajectory is returned when the ST_Intersection function is used for trajectory objects. | |||
The following issue is fixed: The system stops responding during a Ganos upgrade. | |||
This minor version is compatible with PostGIS 2.5.4. |
August 2020
Minor version 1.1.3
Category | Feature | Description | References |
New feature | View | The views that display SQL execution statistics are added. | None |
The views that display queries per second (QPS) statistics are added. | None | ||
The views that display CGroup status statistics are added. | None | ||
Privileged account | The SELECT permissions on dba_* views are granted to the accounts that have the polar_superuser permissions. | None | |
Fixed issue | The following issue is fixed: A display error is returned for the polar_stat_io_info view. | ||
The following issue is fixed: The AWR system wait information is missing. | |||
The following issue is fixed: An error occurs during partition pruning for a partitioned table. | |||
The following issue is fixed: An error occurs when the Segmented Least Recently Used (SLRU) algorithm is used to calculate the memory size. | |||
The following issue is fixed: Read-only nodes are recreated due to a playback latency. |
Ganos V2.9
Category | Feature | Description | References |
New feature | Support for Cloud Optimized GeoTIFF (COG) files | The COG files are supported. The pyramid information that is stored in COG files can be read. | None |
ST_AddZ function | The ST_AddZ function is supported. This function is used to specify the Z coordinate of a geometry object based on the pixel value of the raster that is converted from the geometry object. | ||
Enhanced capabilities for obtaining the spatial extent information about raster objects | The ST_Extent function is supported. This function is used to retrieve the spatial extent of a raster object and return the box representation of the raster object. | ||
The ST_Envelope function is supported. This function is used to retrieve the spatial extent of a raster object and return the spatial extent in the format of a geometry object. | |||
The ST_ConvexHull function is supported. This function is used to retrieve the spatial extent of a raster object and return the spatial extent in the format of a geometry object. | |||
The ST_Height function is supported. This function is used to retrieve the pixel height of a raster object. | |||
The ST_Width function is supported. This function is used to retrieve the pixel width of a raster object. | |||
Fixed issue | The following issue is fixed: Performance bottlenecks occur when raster objects that are stored in external storage locations use a chunking dimension of 1 × n. You can configure storage parameters to customize the size of the chunking dimension. | ||
The following issue is fixed: The returned values are different from the coordinate values when you use the ST_Values function to obtain values in some directions. | |||
The following issue is fixed: The ST_BestPyramidLevel function returns a negative value. | |||
The following issue is fixed: The ST_BuildPyramid function repeatedly creates pyramids. | |||
The following issue is fixed: The block table of a raster table is not deleted when you truncate the raster table. | |||
The following issue is fixed: The ST_ExportTo function is invalid for CreateOption in some cases. | |||
The following issue is fixed: If a table name contains uppercase and lowercase letters, the ST_ClearChunks function fails. | |||
The following issue is fixed: An external pyramid cannot create an overview. | |||
The following issue is fixed: Raster objects that have external pyramids cannot create internal pyramids. | |||
The following issue is fixed: Invalid results are returned when you calculate the raster data that contains NaN values. |
July 2020
Minor version 1.1.2
Category | Feature | Description | References |
New feature | Pivot syntax | An alias can be specified for a table or a subquery in the Pivot syntax. | None |
ROWID pseudocolumn | The ROWID pseudocolumn and the ROWIDTOCHAR and CHARTOROWID functions are supported. | None | |
Built-in functions | The LNNVL, DUMP, SYS_CONTEXT, USERENV, TZ_OFFSET, and VSIZE functions are supported. | None | |
DBMS_JOB and DBMS_SCHEDULER | DBMS_JOB and DBMS_SCHEDULER are supported. They are used to maintain and manage jobs. | None | |
Temporary tables | Temporary tables can be written to local storage. This improves the read and write performance of temporary tables. | None | |
Event trigger | Logical replication can be performed on events captured by the event trigger. Accounts that have the polar_superuser permissions can perform logical replication. | None | |
Custom resource group | Requests can be initiated by custom resource groups. | None | |
Optimized feature | Process of recovery from a failure | The process to recover from a failure is optimized. This reduces the amount of time that is required to recover from a severe failure. | None |
Fixed issue | The following issue is fixed: Accounts that have the polar_superuser permissions cannot use the Index Advisor. |
Ganos V2.8
Category | Feature | Description | References |
New feature | Enhanced interfaces that are used to access raster metadata | The ST_XMin function is supported. This function is used to obtain the minimum value in a set of rasters on the X-axis. | |
The ST_YMin function is supported. This function is used to obtain the minimum value in a set of rasters on the Y-axis. | |||
The ST_XMax function is supported. This function is used to obtain the maximum value in a set of rasters on the X-axis. | |||
The ST_YMax function is supported. This function is used to obtain the maximum value in a set of rasters on the Y-axis. | |||
The ST_ChunkHeight function is supported. This function is used to obtain the height of a raster tile. | |||
The ST_ChunkWidth function is supported. This function is used to obtain the width of a raster tile. | |||
The ST_ChunkBands function is supported. This function is used to obtain the number of bands of a raster tile. | |||
ST_SrFromEsriWkt function | The ST_SrFromEsriWkt function is supported. This function is used to convert the format of a spatial reference string from Esri into Open Geospatial Consortium (OGC). | ||
Raster data type | Raster data files can be compressed by using the Zstandard (zstd) compression algorithm or the Snappy compression library. | None | |
Point cloud data type | Point cloud files can be copied in binary mode. | None | |
Environment variables | The PROJ_LIB and GDAL_DATA environment variables can be configured, and the relevant data can be deployed. | None | |
Fixed issue | The following issue is fixed: Databases do not respond due to invalid OSS paths. | ||
The following issue is fixed: After raster data is imported to databases, the SRIDs of some raster data are inconsistent with the original SRIDs. |
June 2020
Minor version 1.1.1
Category | Feature | Description | References |
New feature | NVL2 function | The NVL2 function is supported. This function is used to return the value of the second or third parameter based on whether the value of the first parameter is NULL or an empty string. | None |
polar_proxy_utils plug-in | The polar_proxy_utils plug-in is supported. This plug-in is used to manage proxy-related features. This plug-in allows you to configure read-only Universal Disk Format (UDF) files and read-only tables. This plug-in also allows you to use a cluster endpoint to route requests for accessing read-only UDF files and read-only tables to the read-only nodes in the cluster. | None | |
polar_resource_group plug-in | The polar_resource_group plug-in is supported. This plug-in is used to customize resource groups. You can use the custom resource groups to isolate resources such as CPUs and memory on a per-user, per-database, or per-session basis. | None | |
Optimized feature | Reliability and availability | The compute nodes of databases are decoupled from file systems. This allows you to perform O&M operations on the file systems. This improves the reliability and availability of databases. | None |
Transaction processing | The snapshot of the active transaction list is replaced with a commit sequence number (CSN). Each CSN is a monotonically increasing identifier. This improves the transaction processing performance of databases. | None | |
Execution plan | Execution plans are optimized to prevent obsolete statistics from being used. | None | |
Fixed issue | The following issue is fixed: A database process failure occurs when the timescaledb plug-in fails to request memory resources. | ||
The following issue is fixed: The I/O monitoring feature does not provide data statistics after a process exits. | |||
The following issue is fixed: A null pointer exception may occur after lock_debug is enabled. | |||
The following issue is fixed: The pg_cron plug-in becomes unavailable. | |||
Known issues of distributed shared memory (DSM) deadlocks in the community edition are fixed. | |||
The following issue is fixed: The number of connections exceeds the upper limit. |
May 2020
Minor version 1.1.0
Category | Feature | Description | References |
New feature | TO_SINGLE_BYTE function | The TO_SINGLE_BYTE function is supported. This function is used to convert multi-byte characters in a string into single-byte characters. | |
TO_MULTI_BYTE function | The TO_MULTI_BYTE function is supported. This function is used to convert single-byte characters in a string into multi-byte characters. | ||
REGEXP_LIKE function | The REGEXP_LIKE function is supported. You can use this function in a way similar to how the LIKE conditional expression is used. The REGEXP_LIKE function complies with the POSIX regular expression standard. | None | |
RATIO_TO_REPORT function | The RATIO_TO_REPORT function is supported. This function is used to calculate the ratio of a value to the sum of a set of values. | None | |
DBMS_UTILITY.FORMAT_ERROR_STACK function | The DBMS_UTILITY.FORMAT_ERROR_STACK function (macro) is supported. This function is used to return stack information and error messages layer by layer. | None | |
UTL_I18N.ESCAPE_REFERENCE and UTL_I18N.UNESCAPE_REFERENCE functions | The UTL_I18N.ESCAPE_REFERENCE and UTL_I18N.UNESCAPE_REFERENCE functions are supported. These functions provide escape and unescape mechanisms for HTML or XML characters. | None | |
INSERT ALL statement | The INSERT ALL statement is supported. You can execute the INSERT ALL statement to insert data into multiple tables at a time. | None | |
polar_concurrency_control plug-in | The polar_concurrency_control plug-in is supported. This plug-in is used to configure a concurrency limit on transaction executions, SQL queries, stored procedures, and DML operations. You can customize large queries and configure a concurrency limit on large queries to improve the execution efficiency in high concurrency scenarios. | None | |
pldebugger plug-in | The pldebugger plug-in is supported. This plug-in is used to debug local stored procedures. You can use the client pgAdmin4 V4.19 or later. | ||
oss_fdw plug-in | The oss_fdw plug-in is supported. This plug-in is used to support external tables of Alibaba Cloud OSS. You can use OSS external tables to write data of databases to OSS or load data from OSS to databases. You can use multiple external tables in parallel or compress data during imports and exports. This improves the performance of data imports and exports. You can also use this feature to separate hot data from cold data that are stored in a wide range of storage media. | ||
Optimized feature | ROUND function | The ROUND function is optimized to support the Interval data type. The returned values are measured in days. | |
polar_stat_activity view | The polar_stat_activity view is optimized. The wait_info and wait_time columns are added. The wait_info column displays the object for which the current process is waiting. The wait_time column displays the amount of time for which the process needs to wait. If the object is a process, the process identifier (PID) is displayed. If the object is a file, the file descriptor (FD) is displayed. | None | |
Index | The pre-extension feature is supported. This feature automatically adds additional index pages when index pages are insufficient for new indexes. This improves the execution performance when data is inserted into a table that has indexes. | None |
Ganos V2.7
Category | Feature | Description | References |
New feature | MD5 functions | The ST_MD5Sum function is supported. This function is used to obtain the MD5 hash value of a raster object. | |
The ST_SetMD5Sum function is supported. This function is used to configure the MD5 hash value of a raster object. | |||
Authentication method of spatial raster objects that are stored in OSS | The ST_AKId function is supported. This function is used to obtain the AccessKey ID that is used to access a raster object stored in OSS. | ||
The ST_SetAccessKey function is supported. This function is used to configure the AccessKey ID and the AccessKey secret that are used to access a raster object stored in OSS. | |||
The ST_SetAKId function is supported. This function is used to configure the AccessKey ID that is used to access a raster object stored in OSS. | |||
The ST_SetAKSecret function is supported. This function is used to configure the AccessKey secret that is used to access a raster object stored in OSS. | |||
Functions that are used to manage spatial raster metadata | The ST_ScaleX function is supported. This function is used to obtain the pixel width of a raster object on the X-axis of the spatial reference system. | ||
The ST_ScaleY function is supported. This function is used to obtain the pixel width of a raster object on the Y-axis of the spatial reference system. | |||
The ST_SetScale function is supported. This function is used to configure the pixel width of a raster object in the spatial reference system. | |||
The ST_SkewX function is supported. This function is used to obtain the skew of a raster object on the X-axis of the spatial reference system. | |||
The ST_SkewY function is supported. This function is used to obtain the skew of a raster object on the Y-axis of the spatial reference system. | |||
The ST_SetSkew function is supported. This function is used to specify the skew of a raster object in the spatial reference system. | |||
The ST_UpperLeftX function is supported. This function is used to obtain the upper-left X-coordinate of a raster object in the spatial reference system. | |||
The ST_UpperLeftY function is supported. This function is used to obtain the upper-left Y-coordinate of a raster object in the spatial reference system. | |||
The ST_SetUpperLeft function is supported. This function is used to configure the upper-left coordinates of a raster object in the spatial reference system. | |||
The ST_PixelWidth function is supported. This function is used to obtain the pixel width of a raster object in the spatial reference system. | |||
The ST_PixelHeight function is supported. This function is used to obtain the pixel height of a raster object in the spatial reference system. | |||
Fixed issue | The following issue is fixed: Aggregate functions result in extension upgrade failures. |