Question 1: Is log_archive_dest_2 = service median lgwr, it must be prepared by the library standby_redo_log
Answer: No, even in maximum protection mode or maximum available, prepared by the library does not have to standby_redo_log
Question 2: if the maximum available modes, the preparation machine may be used archivelog
Answer: Yes, but the application of archivelog, the main library data can not be real-time access to the standby database, you can query the archive only after
Problem 3: the maximum available mode, after preparation machine downtime, restart preparation machine, whether DataGuard can continue, and the maximum available or mode
Answer: Yes, is the largest model available
If the host is started, there is no real-time application backup machine long standby redo log, there may be abnormal, you need to prepare the machine to perform alter database recover managed standby database disconnect from session; apply archived logs, and other archive log application after the completion of the implementation alter database recover managed standby database cancel, then execute alter database recover managed standby database using current logfile disconnect from session; application standby redo log
Question 4: maximum available mode, after the host is down, restart the host, dataguard whether to continue, and also the largest model available
Answer: Yes, is the largest model available
If the host is started, there is no real-time application backup machine long standby redo log, there may be abnormal, you need to prepare the machine to perform alter database recover managed standby database cancel, then execute alter database recover managed standby database disconnect from session log archiving applications, such as archiving log application after the completion of the implementation alter database recover managed standby database cancel, then execute alter database recover managed standby database using current logfile disconnect from session; use standby redo log
Question 5: maximum available mode, the host is down, unable to start, if the backup machine can be upgraded to host
Answer:
Preparation machine to perform the
1. alter database recover managed standby database finish force;
2. alter database commit to switchover to primary with session shutdown;
3. alter database open;
Question 6: Preparation library does not create standby redo log whether real-time application logs
Answer: No, the real-time application logs will complain ORA-38500: USING CURRENT LOGFILE option not available without standby redo logs
Question 7: maximum performance mode, real-time applications prepared by the library if you can log
Answer: If you use lgwr maximum performance mode to transfer logs, real-time applications can be prepared by the library logs
Question 8: maximum availability mode, whether prepared by the library can be used archive log
Answer:
Maximum availability mode, have been prepared by the library database master may receive data in real time, but may be selected archiving application
Question 9: The main library is not set up properly if db_unique_name can build datagard
Answer:
Because the main library is not set db_unique_name, the main library db_unique_name and service_name is the default db_name
Question 10: Can not be provided by the library db_unique_name
Answer: Yes, if the master library set db_unique_name, the standby database can not be provided, but this is not recommended settings
Usually the master library set is not db_unique_name, the parameter file prepared by the library provided db_unique_name, then the standby * .LOG_ARCHIVE_CONFIG = 'DG_CONFIG = (main db_name, standby db_unique_name)'
Tns and are arranged in the standby main service_name = db_name Lord, prepared db_unique_name prepared service_name =
Question 11: build a datagard, after testing LOG_AUTO_DELETE is true, to see the archive log will be automatically deleted
Exec DBMS_LOGSTDBY.APPLY_SET('LOG_AUTO_DELETE', 'TRUE');
Answer: No, I do not know can not in logic environment
Under physical standby, the standby database can not execute the statement, the statement is executed after the main library, the log after application, the main library and library equipment logs did not see deleted
Question 12: Do not copy the main library building when dataguard online journal, that when prepared by the library to begin creating online redo log
Answer: After the standby database execute alter database recover managed standby database disconnect from session or alter database recover managed standby database using current logfile disconnect from session, if the standby database * .log_file_name_convert parameters are set correctly, the standby database to start creating online redo log
As long as the standby database to perform a recover, prepared by the library began to create online redo log
Question 13: When using rman backup recovery set up dataguard, incremental backup package will be used when the standby database alter database recover managed standby database disconnect from session it
Since not, if you use the 0 level 1 backup and backup of level 2 backup, backup database restore using 0 backup, do not use Class 1 and Class 2 when backup package recover, but the use of self-0 backup: Answer All archived log
Question 14: datagard and build the operating system user name and password it matter
Answer: No, only the user name and password and the oracle, generally only about the user and sys
Question 15: datagard can automatically delete archived logs by rman command you
Answer:
The implementation of such a script on the line delete noprompt archivelog until time "sysdate-30";
Question 16: Before After configuring the standby database, the standby database archive log application has not yet begun, the standby database backup restore by 0, the main library will copy all archived logs will be backed up after 0, is not it alter database recover managed standby database disconnect from session requires that all logs are archived copy came to be
Answer: No, the copy will be part of
No, only the main library 0 backup archive logs are still there, and the first archive log transfer after 0 backup to the backup database, prepared by the library began to recover, during recover, the need to archive log from which main library archive logs to acquire which, if the main library archive log is lost, it will be incorrect report
Question 17: This error is not just because the password file disparity in primary and libraries
FAL[client, USER]: Error 1031 connecting to mierpdb2 for fetching gap sequence
ORA-01031: insufficient privileges
ORA-01031: insufficient privileges
Answer: Not necessarily, because sometimes recover standby database archive log, get less than the main library archive log also reported such a mistake.
Question 18: How to deal with the following issues
FAL[client]: Failed to request gap sequence
GAP - thread 1 sequence 168065-168066
DBID 3642507004 branch 645772988
FAL[client]: All defined FAL servers have been attempted.
Answer: the lack of explanation by the library to archive log sequence of 168 065 and 168 066
Solution 1: If the main library also lack the log and never found again, then datagard not continue, and needs to be redone if the main library have a backup of the log, then restore the backup archive logs, and to restore them in copy the archive logs to the standby database, and execute the following statement registered archive log, if V $ ARCHIVED_LOG have these logs, but no longer an MRP application may hang the process, alter database recover managed standby database cancel the execution after execution alter database recover managed standby database disconnect from session; - log application archive or library prepared alter database recover managed standby database using current logfile disconnect from session; - standby application library prepared redo logs
SQL>ALTER DATABASE REGISTER PHYSICAL LOGFILE '/mierp/arch/1_168065_836701255.dbf';
SQL>ALTER DATABASE REGISTER PHYSICAL LOGFILE '/mierp/arch/1_168066_836701255.dbf';
Solution 2: If the primary database is not the lack of this journal, which in turn how to understand it? In practice come across such a case, the archive log 168,065 before was a period of time taken from the main library, log shipping primary database and standby database are my disabled after this, then, when the main library to log points after 168,066 later, re-open the log shipping primary database and standby database, archive log 168065-168066 like this on the missing, there have been more than an error. After this problem out, to take the following measures: the database FAL_SERVER transmission path is disabled by the library archive log off after a period of time to re-start the path to the archive, the archive log and found that start automatically transmitted. In standby database can see the following logs, you can clearly see the standby database log shipping process RFS restart. Accordingly, understanding, RFS [10] and the RFS [11] These two standby database log shipping process and the lack of information is not acquired by the library some old archive log at startup, so it only after obtaining 168,164, by the main library the new generation of the log, while the standby database to restore the archive logs 168,065, only to find missing some old log, but RFS [10] and the RFS [11] will not be told this information, so they do not go get old log. And when the restart log shipping, RFS [12] and is currently still missing some of the old log RFS [13] These two new standby database log shipping process will be told, these two processes go to the main library attempts to obtain missing front after the log and found that, began the transfer of the old logs.
Question 19: datagard archive log backup library is how a generation process
Reply:
1. Main Library lgwr, prepared by the library has not applied to archive log process that is not yet executed alter database recover statements (ie, there is no real online redo log file backup library), prepared by the library does not establish standby, the main library to perform alter system archive log current archive will look to the standby database yet
1. will
Concluded that: as long as the equipment to mount the library open state (not even restore database), the main library will be transferred to the archive log backup libraries, the main library handover date, the archive will be transferred to the new library equipment
That did not restore database backup library do not have any data files and online redo log and standby redo log can also query the v $ log, query results are basically the same and the main library
Query v $ standby_log but there is no result because there is no established standby redo log
2. The main library lgwr, prepared by the library does not apply archived logs, prepared by the library to establish standby redo log, main library to perform alter system archive log current standby redo the sequence to see if changes
2. will change
3. The main library arch, prepared by the library does not use the archive log backup library does not establish standby, the main library to perform alter system archive log current archive will look to the standby database yet
3. will
Answer: All of the standby database archive log is prepared by RFS- library> ARCN generation, if the standby database has standby redo log, then the process is prepared by the library RFS-> standby redo log-> ARCN, and prepare their own online redo log library does not matter, because LOG_ARCHIVE_DEST_N start to finish are the main library decided to redo log store where the main library RFS station equipment library because the library does not even prepare online redo log information can also be found in the v $ log, and such information v $ log is actually a main library, even in the ARCN transmission mode, not the log file copied by ARCN from the main library to a position corresponding standby database, because the main library copy files to the standby database by the OS, username password need an OS verification. In fact we did not see when the log file generated by the library, the username and password is required of the OS.
Question 20: maximum availability mode, the standby database mount and restore database and create a standby redo log, main library archive log transfer to a standby database, but has not been applied before, if you can skip the first application of archive log applied directly to standby redo log
Answer:
Question 21: dataguard standby database select name from v $ datafile and select member from v $ logfile are normal, select tablespace_name from dba_tablespaces will appear ORA-03113: end-of-file on communication channel error, according to an alarm log information is ERROR: slave communication error with ASM; terminating process 7610, how to produce, how to solve
Answer: because * .db_file_name_convert only configuration data file, no temp files cause, you can reconfigure the pfile library equipment and restart solved, you can also perform select name from v $ tempfile find temporary file and then alter database tempfile '+ DATA / /temp.263.868896343 'drop including datafiles delete tempfile, you can then create a tempfile. alter tablespace temp add tempfile '/data01/ibm186/data01/temp.dbf' size 4096M;
Question 22: dataguard main library file is created by OMF, prepared by the library also created by OMF, whether prepared by the library can be configured only db_create_file_dest main library and library equipment can be in * .db_file_name_convert
Answer: not as the subdirectory name db_create_file_dest main library directory library equipment may be different, it will lead to not create file, create a file rule since OMF is db_create_file_dest \ instance name \ datafile or tempfile \ file name
Question 23: Maximum Availability mode, the primary database fails, the standby switching library Failover whether the following four steps must be performed
1. Preparation of libraries for performing ALTER DATABASE SET STANDBY DATABASE TO MAXIMIZE PERFORMANCE;
2. Preparation of libraries for performing alter database recover managed standby database finish force;
3. 备库执行alter database commit to switchover to primary with session shutdown;
4. Preparation of library to perform alter database open;
Answer: Yes, although you can skip the first step, but this is not recommended
According to Question 8, concluded
For maximum protection and maximum availability modes, stressed the fact is that, redo data must be transmitted in real time to the standby database (this sentence may be so understanding, redo the main library must be able to make use of real-time backup library, but not to say that change redo the main library involved, it must be written immediately to the data file backup library)