oracle12g_adg gap修复.pdf

杨凯同学

215

7页

0次

2023-09-15

5墨值下载

PowerView 已关闭

上次登录：July 31, 2023 2:25 PM CST

(0)

yang (可用)

与我们联系帮助

In this Document

Goal

Solution

Stop Managed Recovery and Redo Transport

Save Existing Logfile Names

Refresh Standby Control File From Primary

Document the Current SCN for the Standby

Save Current RMAN Configuration Settings

Refresh the Standby Control File

Replace RMAN Configuration Settings and Remove Orphaned Online and Standby Redo Log Files

Catalog Standby Database Files to the Refreshed Control File

Restore Any Missing Files

Switch Database to Copy

Clean Up Orphaned Files

Clear Online Redo Logs and Standby Redo Logs

Roll the Standby Database Forward

Restart All Instances to Mount

Re-Enable Redo Transport

Determine Section Size to be Used

Run Recover From Service

Recover Until Consistent

Re-Enable Flashback Database

Restart the Standby and Managed Recovery

Take a New Backup (If Needed)

References

APPLIES TO:

Oracle Database - Enterprise Edition - Version 12.1.0.1 to 21.6 [Release 12.1 to 21.0]

Information in this document applies to any platform.

GOAL

Utilize RECOVER DATABASE FROM SERVICE to skip over a large gap in redo or a gap in redo which is

unresolvable(missing logs).

SOLUTION

In cases where there is a large redo gap between the primary and standby or when there is an unresolvable gap of

missing logs, the standby database can be 'rolled forward' using the RECOVER DATABASE FROM SERVICE command

introduced in 12c. This process recovers blocks from the primary to the standby which have changed since the

standby database SCN. This is more efficient in catching up the standby rather than applying every individual change

to the database as is done with managed recovery.

Follow the steps to roll forward a standby database using RECOVER DATABASE FROM SERVICE.

NOTE: Cleaning up any backups from orphaned incarnations (e.g. snapshot standby incarnations) on the standby

before executing this process is recommended.

Stop Managed Recovery and Redo Transport

via Data Guard Broker connected to any database in the configuration:

DGMGRL> edit database <primary> set state=TRANSPORT-OFF;

Succeeded.

此文档是否有帮助?

文档详细信息

类型:

状态:

上次主更

新:

上次更新:

语言:

HOWTO

PUBLISHE

2023-2-22

2023-4-18

English

最近查看

How to Roll Forward a Standby Database Using Recover Database From Service (文档 ID 2850185.1) 转到底部

是

否

11g Steps to perform

for Rolling Forward a

Physical Standby

Database using RMAN

Incremental Backup.

[836986.1]

Roll Forward Physical

Standby Using RMAN

Incremental Backup in

Single Command

[2431311.1]

面板服务请求补丁程序和更新程序社区认证知识

提供反馈...

DGMGRL> edit database <standby> set state=APPLY-OFF;

Succeeded.

via SQLPLUS (if Data Guard Broker is not configured):

From the Primary database:

SQL> alter system set LOG_ARCHIVE_DEST_STATE_<#>=DEFER;

System altered.

SQL> alter system archive log current;

From the Standby database:

SQL> recover managed standby database cancel;

Media recovery complete.

Save Existing Logfile Names

This roll forward process will create new online redo logs and standby redo logs, leaving the existing logs as orphans

using up space. Save the logfile names with the following query to be removed later in the process.

SQL> set heading off linesize 999 pagesize 0 feedback off trimspool on

SQL> spool /tmp/delete_logfiles.log

SQL> select member from v$logfile;

SQL> spool off

Refresh Standby Control File From Primary

The standby database control file must be refreshed to advance the control file SCN. This new standby control file will

contain information about any new files added to the database since the current standby SCN. Additional steps are

described to update the locations of the standby files in the control file and restore any of those newly added files.

Document the Current SCN for the Standby

Run the command in the example below on the standby database and save the SCN for later use.

RMAN> select current_scn from v$database;

CURRENT_SCN#

------------------

644203931

Save Current RMAN Configuration Settings

Refreshing the standby control file from the primary control file overwrites the RMAN configuration items specific to

the standby database.

Create a script to restore the current RMAN configuration settings. This will be used after the control file is restored

from the primary.

NOTE: This step is when using target database control file instead of recovery catalog

On the standby:

$ $ rman target / nocatalog log=/tmp/RMAN_settings.log <<EOF

show all;

EOF

To remove the unnecessary commands output and the RETENTION POLICY command, which cannot be executed on a

standby, run the following:

$ grep ^CONFIGURE /tmp/RMAN_settings.log | grep -v 'RETENTION POLICY' >/tmp/RMAN_settings.rman

$ rm /tmp/RMAN_settings.log

Refresh the Standby Control File

Start one instance in nomount and restore the control file from the primary database.

$ srvctl stop database -d <db> -o immediate

$ rman target / nocatalog

RMAN> startup nomount

Assistant: Download

Reference for Oracle

Database/GI Update,

Revision, PSU,

SPU(CPU), Bundle

Patches, Patchsets and

Base Releases

[2118136.2]

Error Correction

Support Dates for

Oracle WebLogic Server

[950131.1]

Security Alert CVE-

2019-2725 Patch

Availability Document

for Oracle WebLogic

Server [2535708.1]

显示更多

PowerView 已关闭

上次登录：July 31, 2023 2:25 PM CST

(0)

yang (可用)

与我们联系帮助

In this Document

Goal

Solution

Stop Managed Recovery and Redo Transport

Save Existing Logfile Names

Refresh Standby Control File From Primary

Document the Current SCN for the Standby

Save Current RMAN Configuration Settings

Refresh the Standby Control File

Replace RMAN Configuration Settings and Remove Orphaned Online and Standby Redo Log Files

Catalog Standby Database Files to the Refreshed Control File

Restore Any Missing Files

Switch Database to Copy

Clean Up Orphaned Files

Clear Online Redo Logs and Standby Redo Logs

Roll the Standby Database Forward

Restart All Instances to Mount

Re-Enable Redo Transport

Determine Section Size to be Used

Run Recover From Service

Recover Until Consistent

Re-Enable Flashback Database

Restart the Standby and Managed Recovery

Take a New Backup (If Needed)

References

APPLIES TO:

Oracle Database - Enterprise Edition - Version 12.1.0.1 to 21.6 [Release 12.1 to 21.0]

Information in this document applies to any platform.

GOAL

Utilize RECOVER DATABASE FROM SERVICE to skip over a large gap in redo or a gap in redo which is

unresolvable(missing logs).

SOLUTION

In cases where there is a large redo gap between the primary and standby or when there is an unresolvable gap of

missing logs, the standby database can be 'rolled forward' using the RECOVER DATABASE FROM SERVICE command

introduced in 12c. This process recovers blocks from the primary to the standby which have changed since the

standby database SCN. This is more efficient in catching up the standby rather than applying every individual change

to the database as is done with managed recovery.

Follow the steps to roll forward a standby database using RECOVER DATABASE FROM SERVICE.

NOTE: Cleaning up any backups from orphaned incarnations (e.g. snapshot standby incarnations) on the standby

before executing this process is recommended.

Stop Managed Recovery and Redo Transport

via Data Guard Broker connected to any database in the configuration:

DGMGRL> edit database <primary> set state=TRANSPORT-OFF;

Succeeded.

此文档是否有帮助?

文档详细信息

类型:

状态:

上次主更

新:

上次更新:

语言:

HOWTO

PUBLISHE

2023-2-22

2023-4-18

English

最近查看

How to Roll Forward a Standby Database Using Recover Database From Service (文档 ID 2850185.1) 转到底部

是

否

11g Steps to perform

for Rolling Forward a

Physical Standby

Database using RMAN

Incremental Backup.

[836986.1]

Roll Forward Physical

Standby Using RMAN

Incremental Backup in

Single Command

[2431311.1]

面板服务请求补丁程序和更新程序社区认证知识

提供反馈...

DGMGRL> edit database <standby> set state=APPLY-OFF;

Succeeded.

via SQLPLUS (if Data Guard Broker is not configured):

From the Primary database:

SQL> alter system set LOG_ARCHIVE_DEST_STATE_<#>=DEFER;

System altered.

SQL> alter system archive log current;

From the Standby database:

SQL> recover managed standby database cancel;

Media recovery complete.

Save Existing Logfile Names

This roll forward process will create new online redo logs and standby redo logs, leaving the existing logs as orphans

using up space. Save the logfile names with the following query to be removed later in the process.

SQL> set heading off linesize 999 pagesize 0 feedback off trimspool on

SQL> spool /tmp/delete_logfiles.log

SQL> select member from v$logfile;

SQL> spool off

Refresh Standby Control File From Primary

The standby database control file must be refreshed to advance the control file SCN. This new standby control file will

contain information about any new files added to the database since the current standby SCN. Additional steps are

described to update the locations of the standby files in the control file and restore any of those newly added files.

Document the Current SCN for the Standby

Run the command in the example below on the standby database and save the SCN for later use.

RMAN> select current_scn from v$database;

CURRENT_SCN#

------------------

644203931

Save Current RMAN Configuration Settings

Refreshing the standby control file from the primary control file overwrites the RMAN configuration items specific to

the standby database.

Create a script to restore the current RMAN configuration settings. This will be used after the control file is restored

from the primary.

NOTE: This step is when using target database control file instead of recovery catalog

On the standby:

$ $ rman target / nocatalog log=/tmp/RMAN_settings.log <<EOF

show all;

EOF

To remove the unnecessary commands output and the RETENTION POLICY command, which cannot be executed on a

standby, run the following:

$ grep ^CONFIGURE /tmp/RMAN_settings.log | grep -v 'RETENTION POLICY' >/tmp/RMAN_settings.rman

$ rm /tmp/RMAN_settings.log

Refresh the Standby Control File

Start one instance in nomount and restore the control file from the primary database.

$ srvctl stop database -d <db> -o immediate

$ rman target / nocatalog

RMAN> startup nomount

Assistant: Download

Reference for Oracle

Database/GI Update,

Revision, PSU,

SPU(CPU), Bundle

Patches, Patchsets and

Base Releases

[2118136.2]

Error Correction

Support Dates for

Oracle WebLogic Server

[950131.1]

Security Alert CVE-

2019-2725 Patch

Availability Document

for Oracle WebLogic

Server [2535708.1]

显示更多

of 7

5墨值下载

oracle 12c

关注

评论