Duplicate log sequence numbers when applying logs on standby.

Follow
Description


There are duplicate log sequence numbers when applying logs on standby.
Does this mean that the same archived log is attempted to apply to Archived Log?

Example

======================================================
Dbvisit Standby Database Technology (7.0.52.15306) (pid 26098)
dbvisit started on dbvlin823: Wed Feb 24 14:45:02 2016 ()
======================================================

>>> Log file(s) for testdb7 from dbvlin822 will be applied to dbvlin823
201602241445 - Log seq 138912 thread 1 applied to standby database testdb7.
201602241445 - Log seq 69453 thread 2 applied to standby database testdb7.
201602241445 - Log seq 69454 thread 2 applied to standby database testdb7.
201602241445 - Log seq 138912 thread 1 applied to standby database testdb7.
201602241445 - Log seq 138913 thread 1 applied to standby database testdb7.
201602241445 - Log seq 69454 thread 2 applied to standby database testdb7.

>>> Dbvisit Archive Management Module (AMM)

Config: number of archives to keep = 600
Config: number of days to keep archives = 1
Config: diskspace full threshold = 80%

Processing /archive...
Archive log dir: /archive
Total number of archive files : 1331
Number of archive logs deleted = 0
Current Disk percent full : 47%

======================================================
dbvisit ended on dbvlin823: Wed Feb 24 14:46:11 2016
======================================================

 Solution

When using RAC, multiple threads are used.
While one archive log from thread 1 for example 138912 are used during recovery,
it might require 2 logs from thread 2 (69453 and 69454).
So you will see the standby asking for the log from thread 1 multiple times.
Example some redo from log 69453 was needed, then some redo from 138912 then some redo from 69454 etc.

When using RAC it can jump between the logs during recovery.

Have more questions? Submit a request

Comments