The reason Oracle archive log exception to be deleted

Copyright attribution: Tiankai Technology
This link: https://blog.csdn.net/dbs_service/article/details/102761067

Description:
Due to business group needs, need to check the history of operating records in a table, when the table did not do the audit, so only use logmnr log mining functions to investigate, later found archived logs certain period of the previous day's disappearance , it should be the emergence of a middle distance, the problem is rather strange, because the backup job to clean up expired archive log is not at this time, and once verified and no one deletion.

Analysis:
The analysis investigation to determine the following:
This is due to automatic maintenance tasks calling process VKRM cleanup process recyclable space caused by the following archive deleted alert log:

Sat Nov 02 03:00:00 2013
Setting Resource Manager plan DEFAULT_MAINTENANCE_PLAN via parameter
Sat Nov 02 03:00:00 2013
Starting background process VKRM
Sat Nov 02 03:00:00 2013
VKRM started with pid=100, OS id=19638 
Sat Nov 02 03:05:05 2013
Thread 1 advanced to log sequence 42518 (LGWR switch)
  Current log# 1 seq# 42518 mem# 0: +REDOLOG1/erprac/onlinelog/group_1.256.788892321
  Current log# 1 seq# 42518 mem# 1: +REDOLOG2/erprac/onlinelog/group_1.256.788892321
Sat Nov 02 03:05:05 2013
Deleted Oracle managed file +ERPFRA/erprac/archivelog/2013_11_02/thread_2_seq_24595.7712.830393561
Deleted Oracle managed file +ERPFRA/erprac/archivelog/2013_11_02/thread_1_seq_42513.7717.830393709
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6618.7748.800372925
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6619.7749.800372929
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6621.7751.800372937
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_5871.2270.800369903
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6622.7753.800372941
Archived Log entry 67025 added for thread 1 sequence 42517 ID 0xda5588ea dest 1:
Sat Nov 02 04:09:13 2013
Thread 1 advanced to log sequence 42519 (LGWR switch)
  Current log# 2 seq# 42519 mem# 0: +REDOLOG1/erprac/onlinelog/group_2.257.788892333
  Current log# 2 seq# 42519 mem# 1: +REDOLOG2/erprac/onlinelog/group_2.257.788892333
Sat Nov 02 04:09:15 2013
Archived Log entry 67027 added for thread 1 sequence 42518 ID 0xda5588ea dest 1:
Sat Nov 02 04:25:06 2013
Thread 1 advanced to log sequence 42520 (LGWR switch)
  Current log# 3 seq# 42520 mem# 0: +REDOLOG1/erprac/onlinelog/group_3.258.788892339
  Current log# 3 seq# 42520 mem# 1: +REDOLOG2/erprac/onlinelog/group_3.258.788892339
Sat Nov 02 04:25:07 2013
Deleted Oracle managed file +ERPFRA/erprac/archivelog/2013_11_02/thread_2_seq_24596.7725.830394587
Deleted Oracle managed file +ERPFRA/erprac/archivelog/2013_11_02/thread_1_seq_42514.7737.830396251
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6630.7762.800372973
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6632.7765.800372981
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6633.7766.800372987
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6634.7767.800372991
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_6635.7768.800372995
Deleted Oracle managed file +ERPFRA/erprac/flashback/log_984.3809.803987449
Archived Log entry 67028 added for thread 1 sequence 42519 ID 0xda5588ea dest 1:

From time to remove the term, is carried out after a certain time interval, the deleted archived logs are deleted after successful backup after RMAN recovery, because at that time accounted for most of the entire archive log of capacity flash recovery area, and the whole fast recovery area is also close to 90%, so from resource considerations, will clean up the log recyclable, here mainly for the archive logs and Flashback logs can be released for deletion, note after No. 5, as data archiving tasks have been completed, and No. 4 that night I had to manually release a large number of space in this area, so the alert log after the No. 5 these automatic cleaning log information no longer appears.

In addition to this process of official notes:

RKRM:
VKRM manages the CPU scheduling for all managed Oracle processes. The process schedules managed processes in accordance with
an active resource plan.

This should be a new feature after 11G, in order to protect the flash recovery area, for a description of the function without too much detailed information, further studies are needed.

Guess you like

Origin blog.csdn.net/dbs_service/article/details/102761067