Getting error while applying logs to MaxDB standby database.

[Former Member]

Hi,

I have created the MaxDB standby instance on other server and copied all the logs from Production instance to Standby instance. I tried applying logs from DBMCLI as below,

recover_start Log LOG 020

But getting error as below in dbm.prt

9 DBMKnl Statement: RESTORE LOG FROM '/oracle/SMD/sapdata5/saparch/BE1_LOG.020' FILE BLOCKSIZE 8 MEDIANAME 'Logu2019

10 DBMKnl Returncode: -7075

I also tried checking the log file using recover_check from DBMCLI getting error as below,

283 DBMSrv Command 'recover_check Log LOG 020' is being executed.

-24580 DBMSrv ERR_COMMAND_FAILED: Command 'recover_check' has ended and failed with return code -24933.

-24933 DBMSrv ERR_NOSERVICESESSION: No service session active

I also tried restoring the log files through DBMGUI but getting error as below in dbm.prt

283 DBMSrv Command 'backup_restore "Log" LOG 021' is being executed.

1 DBMKnl Sending an administrative statement to the database

8 DBMKnl Statement: RESTORE LOG FROM '/oracle/SMD/sapdata5/saparch/BE1_LOG.021' FILE BLOCKSIZE 8 MEDIANAME 'Logu2019

3 DBMKnl Received the result of an administrative statement from the database

9 DBMKnl Statement: RESTORE LOG FROM '/oracle/SMD/sapdata5/saparch/BE1_LOG.021' FILE BLOCKSIZE 8 MEDIANAME 'Logu2019

10 DBMKnl Returncode: -7075

5 DBMKnl Errortext: Current save skipped, next is ready to take on this tape

419 DBMSrv Command 'backup_restore' has ended with return code 0.

283 DBMSrv Command 'backup_restore "Log" LOG 022' is being executed.

1 DBMKnl Sending an administrative statement to the database

8 DBMKnl Statement: RESTORE LOG CANCEL

3 DBMKnl Received the result of an administrative statement from the database

9 DBMKnl Statement: RESTORE LOG CANCEL

10 DBMKnl Returncode: -104

5 DBMKnl Errortext: DBM command impossible at this time

1 DBMKnl Sending an administrative statement to the database

8 DBMKnl Statement: RESTORE LOG FROM '/oracle/SMD/sapdata5/saparch/BE1_LOG.022' FILE BLOCKSIZE 8 MEDIANAME 'L

1 DBMSrvCo A DBM Server client connection was established at 2011-01-31 21:24:47 (client process has process ID 5052 on computer testsap14

421 DBMSrv Command 'backup_restore_state' was executed since 2011-01-31 21:24:50.

-24580 DBMSrv ERR_COMMAND_FAILED: Command 'backup_restore_state' has ended and failed with return code -24994.

-24994 DBMSrv ERR_RTE: Runtime environment error

-24778 DBMSrv 1,wrong connection state:requested

-24580 DBMSrv ERR_COMMAND_FAILED: Command 'backup_restore' has ended and failed with return code -24994.

-24994 DBMSrv ERR_RTE: Runtime environment error

-24778 DBMSrv 4,connection broken server state 4

2 DBMSrvCo A DBM Server client connection was released (client process has process ID 5052 on computer testsap14

283 DBMSrv Command 'backup_restore_cancel' is being executed.

1 DBMKnl Sending an administrative statement to the database

8 DBMKnl Statement: RESTORE LOG CANCEL

-24580 DBMSrv ERR_COMMAND_FAILED: Command 'backup_restore_cancel' has ended and failed with return code -24994.

-24994 DBMSrv ERR_RTE: Runtime environment error

-24778 DBMSrv 1,wrong connection state:requested

And also can I do the switchover like oracle dataguard in MaxDB standby database. Before bringing the MaxDB standby as production instance I will do the log switch on production MaxDB database and apply that last log to standby MaxDB and then put the Production MaxDB database into Admin mode and then I will bring the MaxDB standby as production instance and shipped the log from standby to production and apply the logs as production is now standby is it possible to apply logs on production standby?

Thanks,

Narendra

[markus_doehr2 · Accepted solution]

You'll have to apply the last log -1 first. We use the following script to feed the standby database:

db_admin
db_connect
recover_start logsich log 21008
recover_replace logsich /archivelog/log 21009
recover_cancel

Markus

[Former Member]

Hi Narendra,

I am planning to configure max db in standby mode, so if any type of document that help me in configuration please share me.

and what is the step are perform  in switchover.

Thanks,

Moopnarain yadav

9911017247

email-moopnarain@gmail.com

[Former Member]

if you search on SAP help then you could you found the below link : SAP MaxDB Standby Database

http://help.sap.com/saphelp_nw70/helpdata/EN/17/4d574b3d3035438f73d022698620ef/content.htm

[former_member229109]

Hello Narendra,

1. After the u201CStatement: RESTORE LOG CANCELu201D , when you cancel the log recovery,

the database will be in offline status.

Please donu2019t bring the database online, if you are planning to continue the logbackups recovery after the new logbackups will be shipped.

2. Bring the database to admin < dbm command u2013 db_admin > before to run dbm command 'backup_restore_state'.

You will still be able to apply the next logbackup file by turning the database into admin status and executing the same commands.

To have the standby stay in cold, execute recover_cancel.

Please pay attention that if you will run recover_ignore instead of recover_cancel, the database will be online and will be NOT ably any more to apply the next shipped logbackup file.

3. u201CBefore bringing the MaxDB standby as production instance I will do the log switch on production MaxDB database and apply that last log to standby MaxDB and then put the Production MaxDB database into Admin mode and then I will bring the MaxDB standby as production instance and shipped the log from standby to production and apply the logs as production is now standby is it possible to apply logs on production standby?u201D

Before to give the answer on this question please let me what do you mean on u201CI will do the log switch on production MaxDB databaseu201D.

And let me know what disaster scenario you are planning to test? < It dependent on the status of the production database and issues occurred on the production database server. >

For example, if you planning the maintenance of the production database server, there are no problems on the production database server, then the scenario steps with small correction will work for you:

• bring the MaxDB database on the production server to admin, create the last logbackup to be shipped to the standby server.

• Recovery of all logbackups shipped from the production server

• Check u201Cdb_restartinfou201D and use u201Crecover_ignoreu201D after all shipped logbackups applied to the MaxDB standby database instance

• Create the complete databackup on the MaxDB standby database instance to be able to switch on the autolog

• And after the maintenance of the database production server will be done, you need to clear the log area, could be done in the admin status,

then you will be able to use the database on production server as standby database

Please review the SAP note, if you are SAP customer:

952783 FAQ: MaxDB high availability

MAXDB documentation:

a) http://maxdb.sap.com/doc/7_7/default.htm

-> Database Administration -> Replication and High Availability -> Standby Database

b) Review u201CSetting Up and Updating Standby Instancesu201D at

http://maxdb.sap.com/doc/7_7/44/e9bd51d0112952e10000000a11466f/content.htm

u201CCopying Log Volumes with Operating System Commands and Starting a Standby Instance as the Active Instanceu201D at

http://maxdb.sap.com/doc/7_7/44/eb62a1b6f0108ee10000000a11466f/content.htm

Thank you and best regards, Natalia Khlopina

[Former Member]

Hi Natalia,

I am trying to test disaster scenario as changing the mode of the MaxDB database. By switching MaxDB standby to Production and then Production to standby and then again production as Production i.e. controlled switchover like in Oracle DataGuard. I followed the same steps as above you mentioned (to use the database on production server as standby database), But when I try to apply logs from standby production to actual production which is now standby I am getting error as belowu2026

First I have run the db_restartinfo on standby production after switchover

./dbmcli on BE1>db_restartinfo

OK

Used LOG Page 24172

First LOG Page 0

Restartable 1

Id Restart Record ttcsolma:BE1_20110201_042430

Id LOG Info ttcsolma:BE1_20110201_042430

Consistent 1

Log Volume Enabled 1

Log Auto Overwrite 0

Master Servernode TTCSOLMA

Converter Version 71682

Oldest Compatible Version 7.7.03.07

Queue Count 8

Repository Imports NO

And then I have run the db_restartinfo on production which is now standby i.e. Admin mode after clearing the log area as you said.

./dbmcli on BE1>db_restartinfo

OK

Used LOG Page 1705101

First LOG Page 2147483647

Restartable 1

Id Restart Record ttcmaxdr:BE1_20110112_211956

Id LOG Info

Consistent 1

Log Volume Enabled 1

Log Auto Overwrite 0

Master Servernode

Converter Version 71666

Oldest Compatible Version 7.7.03.07

Queue Count 8

Repository Imports NO

Then I generated some logs on standby production after switchover and then tried to apply logs from standby production to actual production which is now standby I am getting error as belowu2026

./dbmcli on BE1>recover_start Log log 001

ERR

-24988,ERR_SQL: SQL error

-8003,Log and data must be compatible

20018,DBIdentifier of DataVolume (ttcmaxdr:BE1_20110112_211956) and LogArea (ttcsolma:BE1_20110201_042430) does not match

Log- and data area are not compatible because 'The DB Identifiers of data and logbackup are not equal'.

k38headmaster, Error code 3800 "incompatible_log"

RestoreLog, Error code 3800 "incompatible_log"

And dbm.prt log as below

283 DBMSrv Command 'recover_start Log log 001' is being executed.

1 DBMKnl Sending an administrative statement to the database

8 DBMKnl Statement: RESTORE log FROM '/software/saparch/BE1_LOG.001' FILE BLOCKSIZE 8 MEDIANAME 'Log'

3 DBMKnl Received the result of an administrative statement from the database

9 DBMKnl Statement: RESTORE log FROM '/software/saparch/BE1_LOG.001' FILE BLOCKSIZE 8 MEDIANAME 'Log'

10 DBMKnl Returncode: -8003

5 DBMKnl Errortext: Log and data must be compatible

-24580 DBMSrv ERR_COMMAND_FAILED: Command 'recover_start' has ended and failed with return code -24988.

-24988 DBMSrv ERR_SQL: SQL error

-24778 DBMSrv -8003,Log and data must be compatible

-24778 DBMSrv 20018,DBIdentifier of DataVolume (ttcmaxdr:BE1_20110112_211956) and LogArea (ttcsolma:BE1_20110201_042430) does not match

-24778 DBMSrv Log- and data area are not compatible because 'The DB Identifiers of data and logbackup are not equal'

-24778 DBMSrv k38headmaster, Error code 3800 "incompatible_log"

-24778 DBMSrv RestoreLog, Error code 3800 "incompatible_log"

Thanks,

Narendra

[former_member229109]

Hello,

1.

If you are SAP customer => I recommend to read the SAP notes:

869267 FAQ: SAP MaxDB LOG area

1377148 FAQ: SAP MaxDB backup/recovery

Did you check the documents u201CCopying Log Volumes with Operating System Commands and Starting a Standby Instance as the Active Instanceu201D at

http://maxdb.sap.com/doc/7_7/44/eb62a1b6f0108ee10000000a11466f/content.htm

And u201CSetting Up and Updating Standby Instancesu201D at

http://maxdb.sap.com/doc/7_7/44/e9bd51d0112952e10000000a11466f/content.htm

For example, if you planning the maintenance of the production database server,

there are no problems on the production database server, then the scenario steps with small correction will work for you:

TEST u2013 production database and TEST1 u2013 standby database

• bring the MaxDB database on the production server to admin, create the last logbackup to be shipped to the standby server.

• Recovery of all logbackups shipped from the production server

• Check u201Cdb_restartinfou201D and use u201Crecover_ignoreu201D after all shipped logbackups applied to the MaxDB standby database instance

• Create the complete databackup u2013 BACKUP2 - on the MaxDB standby database instance to be able to switch on the autolog

• And after the maintenance of the database production server will be done, you need to clear the log area, could be done in the admin status,

< in u201CSetting Up and Updating Standby Instancesu201D document you learned how to do it with dbm command u201Cdb_activate RECOVERu201D >,

Sintax: db_activate RECOVER <medium_name> DATA [ExternalBackupID <external_backup_ID>] [AUTOIGNORE]

You need the complete databackup u2013 BACKUP2 u2013 shipped to TEST server & create the medium pointed to this backup first >

• then you will be able to use the database on production server as standby database

2.

Please review the SAP note, if you are SAP customer:

952783 FAQ: MaxDB high availability

Are you SAP customer?

Regards, Natalia Khlopina

[Former Member]

Hi Natalia,

Thanks for the replyu2026Yes we are SAP Customer and I have gone through the SAP note as you mentioned.

But my question is on u201Cplanning the maintenance of the production database serveru201D And I have followed all the steps as you listed. I want to know without Complete Data Backup restore on new standby i.e. earlier was production database is not possible to restore only the logs which I have shipped. In other words I wonu2019t be able to use the database on production server as standby database unless I restore complete data backup?

But I have found in sap note: 952783 in that the document u201CStandby_MaxDB_pdf.zipu201D Says u201CAs of version 7.5.00 Build 17 the database can clear the log of the new standby instance and immediately restore log backups.u201D

Do this feature is not available in 7.7 or 7.8 MaxDB version or is only available in version 7.5.00 Build 17 ?

Thanks,

Narendra