Oracle DataGuard之--手工解决日志GAP（日志间隙）
在Oracle DG 中当standby db 失联后，会照成主备库，日志不同步，照成日志的GAP；在配置DG参数（FAL_SERVER)，备库可以主动向主库request log，但有时候不能自动获取时，可以通过手工来解决日志GAP.
系统环境：
操作系统： RedHat EL55
Oracle： Oracle 11gR2
Primary DB: BJ
Standby DB: GZ
一、查看物理备库GAP 信息
要确定在你的物理备数据库上是否有归档中断，查询V$ARCHIVE_GAP 视图，如下面例子所示：
|
|
|
|
SQL> SELECT NAME FROM V$ARCHIVED_LOG WHERE THREAD
#=1 AND DEST_ID=1 AND SEQUENCE# BETWEEN X AND X；
|
二、Oracle DG 解决日志 gap（间隙）问题，手工在备库上注册archive log
1、在主库上将日志传送到备库
[oracle@bj admin]$ cd /dsk4/arch_bj/
[oracle@bj arch_bj]$ ls
arch_1_10_830282966.log arch_1_23_830282966.log arch_1_36_830282966.log
arch_1_11_830282966.log arch_1_24_830282966.log arch_1_37_830282966.log
arch_1_12_830282966.log arch_1_25_830282966.log arch_1_38_830282966.log
arch_1_13_830282966.log arch_1_26_830282966.log arch_1_39_830282966.log
arch_1_14_830282966.log arch_1_27_830282966.log arch_1_40_830282966.log
arch_1_15_830282966.log arch_1_28_830282966.log arch_1_41_830282966.log
arch_1_16_830282966.log arch_1_29_830282966.log arch_1_42_830282966.log
arch_1_17_830282966.log arch_1_30_830282966.log arch_1_43_830282966.log
arch_1_18_830282966.log arch_1_31_830282966.log arch_1_44_830282966.log
arch_1_19_830282966.log arch_1_32_830282966.log arch_1_45_830282966.log
arch_1_20_830282966.log arch_1_33_830282966.log arch_1_46_830282966.log
arch_1_21_830282966.log arch_1_34_830282966.log arch_1_47_830282966.log
arch_1_22_830282966.log arch_1_35_830282966.log arch_1_9_830282966.log
[oracle@bj arch_bj]$ scp *4* sh:/dsk4/arch_gz
arch_1_41_830282966.log 100% 4650KB 4.5MB/s 00:00
arch_1_42_830282966.log 100% 21KB 20.5KB/s 00:00
arch_1_43_830282966.log 100% 166KB 166.0KB/s 00:00
arch_1_44_830282966.log 100% 35KB 34.5KB/s 00:00
arch_1_45_830282966.log 100% 4096 4.0KB/s 00:00
arch_1_46_830282966.log 100% 33KB 33.0KB/s 00:00
arch_1_47_830282966.log 100% 34KB 33.5KB/s 00:00
[oracle@bj arch_bj]$
2、备库日志、根据日志信息，可以了解备库缺少sequence 41以后的日志
Media Recovery Waiting for thread 1 sequence 41
Mon Nov 04 15:55:01 2013
3、在备库上注册archive log
15:55:21 SYS@ gz>alter database register logfile '/dsk4/arch_gz/arch_1_41_830282966.log';
Database altered.
Elapsed: 00:00:00.03
15:55:33 SYS@ gz>alter database register logfile '/dsk4/arch_gz/arch_1_42_830282966.log';
Database altered.
Elapsed: 00:00:00.01
15:55:43 SYS@ gz>alter database register logfile '/dsk4/arch_gz/arch_1_43_830282966.log';
Database altered.
Elapsed: 00:00:00.02
15:56:01 SYS@ gz>
4、备库日志
-----备库会主动解决日志gap
alter database register logfile '/dsk4/arch_gz/arch_1_41_830282966.log'
There are 1 logfiles specified.
ALTER DATABASE REGISTER [PHYSICAL] LOGFILE
Completed: alter database register logfile '/dsk4/arch_gz/arch_1_41_830282966.log'
Mon Nov 04 15:55:35 2013
Media Recovery Log /dsk4/arch_gz/arch_1_41_830282966.log
Media Recovery Waiting for thread 1 sequence 42
Mon Nov 04 15:55:43 2013
alter database register logfile '/dsk4/arch_gz/arch_1_42_830282966.log'
There are 1 logfiles specified.
ALTER DATABASE REGISTER [PHYSICAL] LOGFILE
Resynchronizing thread 1 from sequence 41 to 42
Completed: alter database register logfile '/dsk4/arch_gz/arch_1_42_830282966.log'
Mon Nov 04 15:55:45 2013
Media Recovery Log /dsk4/arch_gz/arch_1_42_830282966.log
Media Recovery Waiting for thread 1 sequence 43
Mon Nov 04 15:56:01 2013
alter database register logfile '/dsk4/arch_gz/arch_1_43_830282966.log'
There are 1 logfiles specified.
ALTER DATABASE REGISTER [PHYSICAL] LOGFILE
Resynchronizing thread 1 from sequence 42 to 43
Completed: alter database register logfile '/dsk4/arch_gz/arch_1_43_830282966.log'
Mon Nov 04 15:56:06 2013
Media Recovery Log /dsk4/arch_gz/arch_1_43_830282966.log
Media Recovery Waiting for thread 1 sequence 44
Mon Nov 04 15:56:08 2013
Using STANDBY_ARCHIVE_DEST parameter default value as /dsk4/arch_gz
RFS[1]: Assigned to RFS process 3854
RFS[1]: Identified database type as 'physical standby': Client is ARCH pid 4244
RFS[1]: Opened log for thread 1 sequence 44 dbid 242121299 branch 830282966
Mon Nov 04 15:56:08 2013
RFS[2]: Assigned to RFS process 3856
RFS[2]: Identified database type as 'physical standby': Client is ARCH pid 4248
RFS[2]: Opened log for thread 1 sequence 45 dbid 242121299 branch 830282966
Archived Log entry 4 added for thread 1 sequence 45 rlc 830282966 ID 0xe741ed4 dest 2:
RFS[2]: Opened log for thread 1 sequence 46 dbid 242121299 branch 830282966
Archived Log entry 5 added for thread 1 sequence 44 rlc 830282966 ID 0xe741ed4 dest 2:
Archived Log entry 6 added for thread 1 sequence 46 rlc 830282966 ID 0xe741ed4 dest 2:
RFS[1]: Opened log for thread 1 sequence 47 dbid 242121299 branch 830282966
Archived Log entry 7 added for thread 1 sequence 47 rlc 830282966 ID 0xe741ed4 dest 2:
Media Recovery Log /dsk4/arch_gz/arch_1_44_830282966.log
Media Recovery Log /dsk4/arch_gz/arch_1_45_830282966.log
Media Recovery Log /dsk4/arch_gz/arch_1_46_830282966.log
Media Recovery Log /dsk4/arch_gz/arch_1_47_830282966.log
Media Recovery Waiting for thread 1 sequence 48
Mon Nov 04 15:56:14 2013
RFS[3]: Assigned to RFS process 3859
RFS[3]: Identified database type as 'physical standby': Client is LGWR ASYNC pid 4250
Primary database is in MAXIMUM PERFORMANCE mode
RFS[3]: Opened log for thread 1 sequence 48 dbid 242121299 branch 830282966
Archived Log entry 8 added for thread 1 sequence 48 rlc 830282966 ID 0xe741ed4 dest 2:
RFS[3]: Opened log for thread 1 sequence 49 dbid 242121299 branch 830282966
Mon Nov 04 15:56:21 2013
Media Recovery Log /dsk4/arch_gz/arch_1_48_830282966.log
Media Recovery Waiting for thread 1 sequence 49 (in transit)
5、问题解决
主库：
16:02:06 SYS@ prod>select max(sequence#) from v$archived_log;
MAX(SEQUENCE#)
--------------
48
Elapsed: 00:00:00.01
备库：
15:56:01 SYS@ gz>select max(sequence#) from v$archived_log;
MAX(SEQUENCE#)
--------------
48
