oracle 11g data guard 中RFS、MRP进程的说明

下面是主备库进程的一张关联图



RFS（remote file server）：运行在备库上的进程，用于在备库上进行主库的日志恢复。默认，这个进程用于接收从主库传送过来的归档日志。
当物理备库启用了 Redo Apply时，备库可以通过实时应用这个特性，直接使用这个进程从 standby redo log 中直接应用 redo 日志。


下面是在备库的 alert 日志中截取的一段关于 RFS 的信息


2851 RFS[1]: Selected log 7 for thread 1 sequence 40 dbid 3135732069 branch 895320808
2852 Sun Nov 15 01:00:48 2015
2853 Archived Log entry 17 added for thread 1 sequence 39 ID 0xbae9db91 dest 1:
2854 Mon Nov 16 22:01:29 2015
2855 RFS[1]: Selected log 6 for thread 1 sequence 41 dbid 3135732069 branch 895320808
2856 Mon Nov 16 22:01:30 2015
2857 Archived Log entry 18 added for thread 1 sequence 40 ID 0xbae9db91 dest 1:
2858 Tue Nov 17 14:15:33 2015
2859 Primary database is in MAXIMUM PERFORMANCE mode
2860 RFS[2]: Assigned to RFS process (PID:21867)
2861 RFS[2]: Selected log 7 for thread 1 sequence 42 dbid 3135732069 branch 895320808
2862 Tue Nov 17 14:15:33 2015
2863 RFS[3]: Assigned to RFS process (PID:21869)
2864 RFS[3]: Selected log 6 for thread 1 sequence 41 dbid 3135732069 branch 895320808
2865 Tue Nov 17 14:15:33 2015
2866 Archived Log entry 19 added for thread 1 sequence 41 ID 0xbae9db91 dest 1:
2867 Tue Nov 17 14:44:20 2015
2868 Primary database is in MAXIMUM PERFORMANCE mode
2869 RFS[4]: Assigned to RFS process (PID:23281)
2870 RFS[4]: Selected log 6 for thread 1 sequence 43 dbid 3135732069 branch 895320808
2871 Tue Nov 17 14:44:20 2015
2872 RFS[5]: Assigned to RFS process (PID:23285)
2873 RFS[5]: Selected log 7 for thread 1 sequence 42 dbid 3135732069 branch 895320808
2874 Tue Nov 17 14:44:22 2015
2875 Archived Log entry 20 added for thread 1 sequence 42 ID 0xbae9db91 dest 1:
2876 Tue Nov 17 16:10:46 2015
2877 Primary database is in MAXIMUM PERFORMANCE mode
2878 RFS[6]: Assigned to RFS process (PID:27628)
2879 RFS[6]: Selected log 7 for thread 1 sequence 44 dbid 3135732069 branch 895320808
2880 Tue Nov 17 16:10:47 2015
2881 RFS[7]: Assigned to RFS process (PID:27632)
2882 RFS[7]: Selected log 6 for thread 1 sequence 43 dbid 3135732069 branch 895320808

在主库的报错日志中，有如下一段，是主库尝试连接到备库的RFS进程


 4111 Error 12541 received logging on to the standby
 4112 Check whether the listener is up and running.
 4113 Tue Nov 10 14:57:49 2015
 4114 Error 1034 received logging on to the standby
 4115 Tue Nov 10 15:02:51 2015
 4116 ARC2: Attempting destination LOG_ARCHIVE_DEST_2 network reconnect (1094)
 4117 ARC2: Error 1094 attaching to RFS for reconnect
 4118 Tue Nov 10 15:03:01 2015
 4119 ARC2: Error 1094 attaching to RFS for reconnect
 4120 ARC2: Error 1094 attaching to RFS for reconnect
 4121 Tue Nov 10 15:03:11 2015
 4122 ARC2: Error 1094 attaching to RFS for reconnect
 4123 ARC2: Error 1094 attaching to RFS for reconnect
 4124 Tue Nov 10 15:03:21 2015
 4125 ARC2: Error 1094 attaching to RFS for reconnect
 4126 ARC2: Error 1094 attaching to RFS for reconnect
 4127 Tue Nov 10 15:03:31 2015
 4128 ARC2: Error 1094 attaching to RFS for reconnect
 4129 ARC2: Error 1094 attaching to RFS for reconnect
 4130 Tue Nov 10 15:03:41 2015
 4131 ARC2: Error 1094 attaching to RFS for reconnect
 4132 ARC2: Error 1094 attaching to RFS for reconnect
 4133 ARC2: Error 1094 attaching to RFS for reconnect

MRP：当物理备库启用了实时应用的时候，会创建这个进程，用于日志的实时应用。


下面是截取备库 alert 日志中的一段信息


2348 Starting background process MRP0
2349 Wed Nov 11 11:24:25 2015
2350 MRP0 started with pid=7, OS id=8220
2351 Wed Nov 11 11:24:26 2015
2352 MRP0: Background Managed Standby Recovery process started (newdb)


当数据库主备角色切换时，备库会杀掉这个进程，下面是 alert 日志中的一段


 4777 MRP0: Background Media Recovery process shutdown (newdb)
 4778 Wed Nov 11 11:22:51 2015
 4779 Role Change: Canceled MRP
 4780 Killing 2 processes (PIDS:27972,20487) (all RFS) in order to disallow current and future RFS connections. Requested by OS process 27978
 4781 Stopping Emon pool
 4782 All dispatchers and shared servers shutdown
 4783 CLOSE: killing server sessions.

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/26506993/viewspace-1846999/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/26506993/viewspace-1846999/