List of new features of Oracle 19c

Pankaj kushwaha
13 min readJul 14, 2021

--

List of new features of Oracle 19c
Labels (space separated): Oracle 19c

Article Directory
List of new features of Oracle 19c
Availability
general
Simplified the management of database parameters in DG Broker
Dynamically modify the Fast-Start Failover (FSFO) target library
Broker’s FSFO supports watch-only mode
When the main database flashes back, the standby database will also flash back
Propagate the main library restore point
DG multi-instance REDO application supports IM
DML redirection in ADG
PDB supports recovery catalog
Regularly clear the flashback log to improve the predictability of the FRA size
The introduction of new parameters in DG is used to adjust the automatic resolution of the interruption plan
More fine-grained supplementary logs
Fragmentation
Propagate parameter values ​​across shards
Support multiple PDB shards in the same CDB
System-Managed Sharding supports multi-table family
Supports multi-shard queries on alternate sharded catalog databases
Generate unique sequences across shards
Big data and data warehouse
general
Improve SQL diagnosis and repair capabilities
Automatic index
SQL function of count distinct based on Bitmap
Big data and In-Memory external table performance improvement
Automatically resolve SQL plan regressions
Real-time statistics
High frequency optimizer statistics collection
Mixed partition table
Overall new features of the database
Automated installation, configuration and patching
DBCA silently copy a database
DBCA silently clone a remote PDB
DBCA silently migrate one PDB to another CDB
Simplify mirror-based Oracle client installation
Install Oracle database root script to support automatic execution
Support Dry-Run Validation for Oracle cluster upgrade.
Automated upgrades, migrations and tools
Data Pump supports excluding encrypted words when importing
The data pump allows the table space to remain read-only during the TTS import process
Test mode of transmission table space in data pump
Data Pump supports resource limits
General new features
Data pump command line parameters: ENABLE_SECURE_ROLES
Data Pump Import supports wildcard dump file names for URL-based dump files maintained in object stores
Data Pump command-line parameter CREDENTIAL allows Import from object stores
performance
General new features
SQL isolation
Automatically enable resource management for IM
While filling, IM will wait
Memoptimized Rowstore-Fast Ingest
Automatic Database Diagnostic Monitor (ADDM) supports PDB
Real-time SQL monitoring
Load capture and replay in PDB
RAC and GRID
General new features
Parity Protected Files — Parity Protected Files
Automated PDB migration
Automated Transaction Draining for Oracle Grid Infrastructure Upgrades
Oracle supports re-upgrade and patching
Oracle Grid supports zero downtime patching
Safety
General new features
New words in the ALTER SYSTEM command FLUSH PASSWORDFILE_METADATA_CACHE
Transparent online conversion in non-OMF mode supports automatic renaming
Key Management of Encrypted Oracle-Managed Tablespaces in Transparent Data Encryption
Support additional algorithms for offline tablespace encryption
Support for Host Name-Based Partial DN Matching for Host Certificates
Privilege Analysis Now Available in Oracle Database Enterprise Edition
Support for Oracle Native Encryption and SSL Authentication for Different Users Concurrently
Ability to grant or revoke administrative privileges from mode-only accounts
Automatic Support for Both SASL and Non-SASL Active Directory Connections
Unified audit TOP statement
Password removed from Oracle database account
Signature-Based Security for LOB Locators
New EVENT_TIMESTAMP_UTC Column in the UNIFIED_AUDIT_TRAIL View
New PDB_GUID Audit Record Field for SYSLOG and the Windows Event Viewer
Database Vault Operations Control for Infrastructure Database Administrators
Database Vault Command Rule Support for Unified Audit Policies
Availability
general
Simplified the management of database parameters in DG Broker
The user can manage and set all DG related parameters through the ALTER SYSTEM command or the EDIT DATABASE… SET PARAMETER command in DGMGRL. And you can use ALL to modify a certain parameter of the database in all DG environments at one time, instead of modifying them one by one.

Dynamically modify the Fast-Start Failover (FSFO) target library
Currently, the DBA must disable Fast-Start Failover (FSFO) to change the FSFO target standby database. Starting from 19c, the command allows users to dynamically change the FSFO target standby database to another standby database in the target list without first disabling FSFO.

Broker’s FSFO supports watch-only mode
When the database administrator configures the FSFO function of DG Broker, it can now be configured as an observation-only mode to create a test mode to see when a failover or other interaction occurs during normal production processing. This allows users to more precisely adjust the FSFO parameter attributes and discover which conditions in their environment will cause automatic failover to occur. This makes it easier to prove the use of automatic failover to reduce the recovery time of failover.

This configuration allows users to test the automatic failover configuration without any actual impact on the production database. This improves the existing failover verification that already exists in Broker and helps users understand the FSFO automatic failover process more easily.

When the main database flashes back, the standby database will also flash back
Flashback Database moves the entire database to an older point in time and uses RESETLOGS to open the database. In DG, if the primary database flashback, the standby database is no longer synchronized with the primary database. In previous versions, the need to set the standby database to the same point in time as the primary database requires a manual process to flash back the standby database. A new parameter was introduced in 19c, which enables the standby database to automatically flashback when the flashback database is executed on the main database.

By automatically flashing back to the standby database when the primary database flashes back, time, effort, and human error are reduced, resulting in faster synchronization and reduced recovery time objectives (RTO).

Propagate the main library restore point
Prior to this, a normal restore point or a guaranteed restore point was defined on the main library to achieve fast point-in-time recovery in the event of any logical damage. However, this restore point is stored in the control file and will not be propagated to the standby database. If a failover occurs, the standby database becomes the primary database and the restore point information will be lost. And this new feature ensures that restore points are propagated from the primary database to the standby database so that the restore points are available even after a failover event.

DG multi-instance REDO application supports IM
Prior to this, multi-instance REDO applications and IM columnar storage could not be enabled at the same time. Starting from 19c, it can be enabled at the same time.

DML redirection in ADG
ADG DML redirection allows DML to be executed on the ADG standby database. When DML is executed, the operation will be passed to its related main library for execution, and the transaction’s REDO will be applied to the standby library. In short, it is

PDB supports recovery catalog
Pluggable database (PDB) is supported as the target database, and virtual private directory (VPC) users can be used to more finely control the authority to perform backup and restore operations at the PDB level. The metadata view is also limited, so VPC users can only view data for which the user has been granted permissions. In previous versions, the connection to the recovery catalog was not supported when the target database was a PDB.

Oracle 19c provides complete backup and recovery flexibility for container database (CDB) and PDB-level backup and restore, including recovery catalog support.

Regularly clear the flashback log to improve the predictability of the FRA size
When there are many databases that use the fast recovery area (FRA). They usually use the recovery_dest_size initialization parameter to set FRA. When enough FRA space is needed, the flashback log will not be cleared, which will cause FRA pressure. In many cases, the only remedy is to turn off flashback logging and turn it back on. In 19c, this feature makes the use of flashback space predictable from a storage management perspective, because flashback does not occupy the space required for reservation. This feature also allows users to control space pressure by adjusting the flashback log retention time.

FRA is essential to the database because it stores backups, online redo logs, archived redo logs and flashback logs. When the FRA space is used up, it will affect the normal use of the database, and the consequences will be very serious.

The introduction of new parameters in DG is used to adjust the automatic resolution of the interruption plan
DG has multiple processes on the main database and standby database for processing redo transmission and archiving. These processes communicate with each other through the network. In certain failure situations, network hangs, disconnections, and disk I/O problems, these processes may hang, which may cause delays in redo transmission and GAP resolution. DG has an internal mechanism to detect these hung processes and terminate them, thereby allowing normal interruption solutions to occur. In Oracle 19c, DBA can use two new parameters DATA_GUARD_MAX_IO_TIME and DATA_GUARD_MAX_LONGIO_TIME to adjust the waiting time of this detection cycle. These parameters allow to adjust the waiting time for a specific DG configuration based on the user’s network and disk I/O behavior.

More fine-grained supplementary logs
Designed and implemented supplementary logging for logical standby or full database replication requirements. This will add unnecessary overhead in an environment where only a subset of the tables are replicated. Fine-grained supplementary logging provides a way for some database replication users to disable supplementary logging for tables that are not of interest, so that even if supplementary logging is enabled at the database or schema level, it will not be provided for tables that are not of interest Supplementary logging overhead.

Using this function can significantly reduce the overhead of resource usage and redo generation, in case only some tables in the database require supplementary logging, such as in the GoldenGate partial replication configuration.

Fragmentation
Propagate parameter values ​​across shards
Before 19c, the DBA had to modify the parameter values ​​of each shard one by one. Starting from 19c, it only needs to be executed on the fragmented catalog database.

Support multiple PDB shards in the same CDB
Prior to this, only one PDB in the same CDB was supported as a fragment. Of course, there are still some restrictions, such as: different PDBs in the CDB must be shards in different shard databases.

System-Managed Sharding supports multi-table family
Before that, Sharding supported a table family in any way.

Supports multi-shard queries on alternate sharded catalog databases
Prior to this, it could only be supported on the main sharded catalog database.

Generate unique sequences across shards
Prior to this, the only way to ensure the uniqueness of the sequence was in all shard databases manually. Starting from 19c, all of this can be handed over to Oracle.

Big data and data warehouse
general
Improve SQL diagnosis and repair capabilities
SQL diagnosis and repair tools (such as SQL Test Case Builder and SQL Repair Advisor) have been enhanced to provide better diagnosis and repair functions for managing problematic SQL statements.

Automatic index
The automatic index function can automatically perform index management tasks, such as creating, rebuilding, and deleting indexes in the Oracle database according to changes in application workload.

SQL function of count distinct based on Bitmap
Count distinct is introduced in 12c, which is used to roughly count the number of different values ​​in a column. Continuously enhanced in 19c, performance and accuracy are greatly improved.

Big data and In-Memory external table performance improvement
IM external tables have added support for ORACLE_HIVE and ORACLE_BIGDATA drivers, parallel query, RAC, DG and on-demand filling.

Automatically resolve SQL plan regressions
SQL plan management searches for SQL statements in AWR. Prioritization is determined by the highest load, it looks for alternative plans among all available sources, and adds better-performing plans to the SQL plan benchmark. Oracle database also provides plan comparison tools and improved prompt reports.

Real-time statistics
Oracle will automatically collect statistics during the execution of the DML.

High frequency optimizer statistics collection
Users can specify higher frequency statistics collection for certain objects. So as to achieve, powerfully generate a more accurate execution plan. . .

Mixed partition table
The hybrid partitioned table function extends Oracle partitions by allowing the partitions to reside in the Oracle database segment and external files and sources. This feature significantly enhances the partitioning function of Big Data SQL, where most of the tables can reside in external partitions.

Overall new features of the database
Automated installation, configuration and patching
DBCA silently copy a database
A database can be copied through the createDuplicateDB command in DBCA.

DBCA silently clone a remote PDB
The PDB can be cloned through the createFromRemotePDB command in DBCA.

DBCA silently migrate one PDB to another CDB
PDB can be migrated through the relocatePDB command in DBCA.

Simplify mirror-based Oracle client installation
Starting with Oracle Database 19c, Oracle Database Client software can be used as an image file for download and installation. You must unzip the imaging software to the directory where you want the Oracle home directory to be, and then run the runInstaller script to start the Oracle Database Client installation. Of course, binary files are still provided and continue to be provided as non-zip files in the traditional format.

Install Oracle database root script to support automatic execution
Starting with Oracle Database 19c, the database installation program or setup wizard provides some options for setting the permissions to automatically run the root configuration script as needed during database installation. You can continue to run the root configuration script manually.

Support Dry-Run Validation for Oracle cluster upgrade.
Starting from 19c, it supports dry operation (simulation upgrade) to verify whether the upgrade requirements are met, rather than a real upgrade.

Automated upgrades, migrations and tools
Data Pump supports excluding encrypted words when importing
The new parameter OMIT_ENCRYPTION_CLAUSE can be used to ignore objects with encrypted columns.

The data pump allows the table space to remain read-only during the TTS import process
That is, in the TTS process, the table space can be reduced on both the source database and the target database. As for the read-only mode, it can provide read services. Instead of only ensuring that the source database is read-only, the target database tablespace cannot be used normally.

Test mode of transmission table space in data pump
The test mode for transportable tablespaces uses transportable tablespaces or fully transportable export/import to perform metadata-only export tests. It also eliminates the requirement that the source database tablespace is in read-only mode.

Now, DBAs can more easily determine the time required to export and discover unforeseen problems that were not reported by the closure check.

Data Pump supports resource limits
In the export and import process of the data pump, the use of its resources can be restricted. This can be achieved through two new parameters: MAX_DATAPUMP_JOBS_PER_PDB and MAX_DATAPUMP_PARALLEL_PER_JOB.

General new features
Data pump command line parameters: ENABLE_SECURE_ROLES
By default, Data Pump no longer enables password-protected security roles. Starting in 19c, you must explicitly enable password-protected roles for a single export or import job. A new command line parameter has been added, ENABLE_SECURE_ROLES =YES|NO, which can be used to
explicitly enable or disable these types of roles for a single export or import job.

Data Pump Import supports wildcard dump file names for URL-based dump files maintained in object stores
Data Pump command-line parameter CREDENTIAL allows Import from object stores
performance
General new features
SQL isolation
SQL statements terminated by Oracle Resource Manager due to excessive consumption of CPU and I/O resources can be automatically isolated. The execution plan associated with the terminated SQL statement will be isolated to prevent them from being executed again.

Automatically enable resource management for IM
When INMEMORY_SIZE is not 0, then the resource manager will be automatically enabled.

While filling, IM will wait
The new DBMS_INMEMORY_ADMIN.POPULATE_WAIT function will keep the object in a waiting state and cannot be accessed until the object of the specified priority has been filled to the specified percentage.

The new function ensures that the specified In-Memory object is filled before allowing application access. For example, a database may contain many in-memory tables with various priority settings. In a restricted session, you can use the POPULATE_WAIT function to ensure that each In-Memory table is completely filled. After that, you can disable restricted sessions to ensure that the application only queries the table’s In-Memory representation.

Memoptimized Rowstore-Fast Ingest
Automatic Database Diagnostic Monitor (ADDM) supports PDB
Real-time SQL monitoring
Load capture and replay in PDB
Prior to this, load and replay could only be captured at the CDB root container level. Starting from 19c, PDB level is supported.

RAC and GRID
General new features
Parity Protected Files — Parity Protected Files
The REDUNDANCY file type attribute specifies the redundancy of the file group. The PARITY value specifies redundant single parity. Parity is suitable for write-once files, such as archive logs and backup sets.

When traditional two or three ASM mirrors are used for files associated with database backup operations, they consume a lot of space. The backup file is a write-once file, this function allows to protect parity instead of traditional mirroring. This saves a lot of space.

Automated PDB migration
In Oracle Grid, Fleet Patching and Provisioning can be used to automatically relocate PDB from one CDB to another.

Automated Transaction Draining for Oracle Grid Infrastructure Upgrades
Oracle supports re-upgrade and patching
Use Fleet Patching and Provisioning to patch and upgrade Oracle Restart. In previous versions, the Oracle Restart
environment required users to perform patch and upgrade operations, which usually required manual intervention. Fleet Patching and Provisioning automate these
processes.

Oracle Grid supports zero downtime patching
Safety
General new features
New words in the ALTER SYSTEM command FLUSH PASSWORDFILE_METADATA_CACHE
The new clause FLUSH PASSWORDFILE_METADATA_CACHE in the ALTER SYSTEM command refreshes the metadata cache with the latest details of the database password file. You can retrieve the latest detailed information of the database password file by querying the V $ PASSWORDFILE_INFO view.

This feature is very useful when you change the database password file name or location, and you need to refresh the metadata cache with the details of the updated database password file.

Transparent online conversion in non-OMF mode supports automatic renaming
Starting from 19c, in the transparent data encryption online conversion in non-OMF mode, it is no longer necessary to include the FILE_NAME_CONVERT clause in the ADMINISTER KEY MANAGEMENT SQL statement. The file name retains its original name.

Key Management of Encrypted Oracle-Managed Tablespaces in Transparent Data Encryption
Support additional algorithms for offline tablespace encryption
Support for Host Name-Based Partial DN Matching for Host Certificates
Privilege Analysis Now Available in Oracle Database Enterprise Edition
Support for Oracle Native Encryption and SSL Authentication for Different Users Concurrently
Ability to grant or revoke administrative privileges from mode-only accounts
Automatic Support for Both SASL and Non-SASL Active Directory Connections
Unified audit TOP statement
Password removed from Oracle database account
Signature-Based Security for LOB Locators
New EVENT_TIMESTAMP_UTC Column in the UNIFIED_AUDIT_TRAIL View
New PDB_GUID Audit Record Field for SYSLOG and the Windows Event Viewer
Database Vault Operations Control for Infrastructure Database Administrators
Database Vault Command Rule Support for Unified Audit Policies
For more details, please check the official document:

https://docs.oracle.com/en/database/oracle/oracle-database/19/newft/new-features.html#GUID-06A15128-1172-48E5-8493-CD670B9E57DC

--

--

Pankaj kushwaha
Pankaj kushwaha

Written by Pankaj kushwaha

Database/System Administrator | DevOPS | Cloud Specialist | DevOPS

No responses yet