Shulou(Shulou.com)05/31 Report--

This article mainly explains "how to use duplicate to do DG matters needing attention", the article explains the content is simple and clear, easy to learn and understand, the following please follow the editor's ideas slowly in-depth, together to study and learn "how to use duplicate to do DG matters needing attention"!

prerequisite

1. The total amount of database is 5T

2. 1.1T of the catalogue of TMagneonline redo on the total daily online redo

3. The incremental backup is made at 23:00 every day, and the backup is completed at 2:00 the next morning. The backup script corresponds to the archive log PLUS ARCHIVELOG DELETE ALL INPUT.

4. Rman setting is CONFIGURE ARCHIVELOG DELETION POLICY TO APPLIED ON STANDBY for archivelog.

5. Use duplicate target database for standby from active database

6. The duplicate process takes about 16 hours.

The most important point of knowledge

After building the DG,duplicate using duplicate, the archive logs that need to be synchronized in the dg repository start at the point in time when the duplicate starts.

If the archive log is not deleted, each PLUS ARCHIVELOG will be backed up once, that is, an archive log can be backed up multiple times.

Question 1 to consider

Since the duplicate takes 16 hours, and the archive logs that need to be synchronized in the dg repository start at the time when the duplicate starts, you need to synchronize the archive logs 16 hours ago after the duplicate is completed. The archive logs that have been backed up will be backed up and cleared within 24 hours. The duplicate time 16 is less than the cleaning cycle 24. These archive logs can be obtained.

Question 2 to consider

If you start doing duplicate at 14:00 and perform backups at 23:00 in the evening, so that 14:00 to 23:00 is 9 hours, less than 16 hours, you can't get the archived logs from 16 hours ago, and DG can't synchronize, can't you do it? No, because the main library has CONFIGURE ARCHIVELOG DELETION POLICY TO APPLIED ON STANDBY set up.

Question 3 to consider

Continue to question 2 above, because the backup starts at 23:00 and ends at 2:00 the next day, that is, the retention time of the log is from 2:00 on the second day to 06:00 on the third day, that is, the retention time is more than 28 hours. This is not the retention limit of 24 hours, easy to burst the disk? Don't worry, you can perform the backup at 23:00 the next night, because the main database has set CONFIGURE ARCHIVELOG DELETION POLICY TO APPLIED ON STANDBY so that the archive logs will not be deleted. At this time, you can manually delete the archive logs before 14:00 and keep the ones after 14:00.

Draw a conclusion on the above three questions

As long as the length of duplicate for standby from active database is less than the time period in which logs must be cleaned, as above, the duplicate is 16 hours, and the log cleaning cycle is 24 hours.

Practical process

1. Duplicate starts at 14:00 on the first day, and now the archive log sequence number of the main database is 416005, and the archive log directory space is 500G (since the backup started at 23:00 the previous day and finished at 2:00 the next day, the data starts from 2:00 to 14:00 on the first day)

2. After backing up the main library at 2:00 the next morning, the archive log sequence number of the main database is 419005, and the archive log directory space is 900G (from 2:00 on the first day to 2:00 on the second day).

3. At 2:05 the next morning, the execution order is as follows. After execution, the archive log directory space is 400G (from 14:00 on the first day to 2:00 on the second day).

RMAN > list backup of archivelog sequence between 416000 and 416005

Make sure that these archived logs are in the backup, and if they are all in the backup, then execute the following

RMAN > delete force archivelog until SEQUENCE 416000

4. When the archive log directory space is 550G (from 14:00 on the first day to 6:00 on the second day), the archive log starts at 416001 a few minutes before 14:00 on the first day. DG is enabled to prepare the application log of the library, and the database starts to use the first archive log 416005.

5. On the morning of the next day, the 10glaze 00recoery DG was almost synchronized. At this time, the archive log directory of the main database was 700g (from 14:00 on the first day to 10:00 on the next day), and then executed in the following order. After execution, the archive log directory space was 300g (from 2:00 to 10:00 on the second day).

RMAN > list backup of archivelog sequence between 419000 and 419005

Make sure that these archived logs are in the backup, and if they are all in the backup, then execute the following

RMAN > delete force archivelog until SEQUENCE 419000

In fact, between steps 1 and 2 above, you can perform a separate backup of the archived logs, that is, a backup of the archived logs at 14:30, and then manually delete the archived logs before 13:30 after the backup ends at 15:30. Of course, the archived logs from 13:00 to 14:00 will be backed up again at 23: 00:00 in the evening, so it's okay to back up twice. As long as you make sure that the archive logs required by DG are not deleted after the completion of duplicate, the archive logs required by DG are those that start at 14:00 at the beginning of duplicate.

Thank you for your reading, the above is the content of "how to use duplicate to do DG notes", after the study of this article, I believe you have a deeper understanding of how to use duplicate to do DG matters needing attention, and the specific use needs to be verified in practice. Here is, the editor will push for you more related knowledge points of the article, welcome to follow!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.