环境：win10   oracle 12.2.0.1单机

1.故障情况

2.查看告警日志

2020-11-17T20:27:09.226581+08:00

Completed: ALTER DATABASE   MOUNT

2020-11-17T20:27:09.274187+08:00

ALTER DATABASE OPEN

2020-11-17T20:27:09.593865+08:00

Ping without log force is disabled:

  instance mounted in exclusive mode.

Endian type of dictionary set to little

2020-11-17T20:27:09.724921+08:00

TT00: Gap Manager starting (PID:12528)

2020-11-17T20:27:09.966826+08:00

Errors in file D:\APP\XMC\VIRTUAL\diag\rdbms\xmc\xmc\trace\xmc_lgwr_2060.trc:

ORA-00313: ??????? 4 (???? 1) ???

ORA-00312: ???? 4 ?? 1: 'D:\1.LOG'

ORA-27041: ??????

OSD-04002: Ϟ·¨´򿫎ļ�-Error: (OS 2) ϵͳ֒²»µ½ָ¶¨µŎļ�020-11-17T20:27:09.973054+08:00

Errors in file D:\APP\XMC\VIRTUAL\diag\rdbms\xmc\xmc\trace\xmc_lgwr_2060.trc:

ORA-00313: ??????? 4 (???? 1) ???

ORA-00312: ???? 4 ?? 1: 'D:\1.LOG'

ORA-27041: ??????

OSD-04002: Ϟ·¨´򿫎ļ�-Error: (OS 2) ϵͳ֒²»µ½ָ¶¨µŎļ�020-11-17T20:27:09.989754+08:00

Errors in file D:\APP\XMC\VIRTUAL\diag\rdbms\xmc\xmc\trace\xmc_ora_15096.trc:

ORA-00313: 无法打开日志组 1 (用于线程 ) 的成员

ORA-00312: 联机日志 4 线程 1: 'D:\1.LOG'

2020-11-17T20:27:10.082878+08:00

USER (ospid: 15096): terminating the instance due to error 313

ERROR: Unable to normalize symbol name for the following short stack (at offset 125):

dbgeGetAsyncActivForED()+939<-dbkif_record_dde_actions()+421<-ksuitm_opt()+2482<-ksuitm()+14<-kcfopd()+10758<-adbdrv_options..0()+38247<-adbdrv()+161<-opiexe()+29912<-opiosq0()+6886<-kpooprx()+464<-kpoal8()+1173<-opiodr()+1378<-ttcpip()+1368<-opitsk()+2474<-opiino()+1265<-opiodr()+1378<-opidrv()+1833<-sou2o()+112<-opimai_real()+499<-opimai()+231<-OracleThreadStart()+745<-00007FF937D77C24<-00007FF938FECEA1

2020-11-17T20:27:10.212274+08:00

System state dump requested by (instance=1, osid=15096), summary=[abnormal instance termination].

System State dumped to trace file D:\APP\XMC\VIRTUAL\diag\rdbms\xmc\xmc\trace\xmc_diag_5220_20201117202710.trc

继续查看trc，搜索error

      32: TMON ospid 6352 sid 4 ser 55667, waiting for 'rdbms ipc message' 

      33: USER ospid 15096 sid 128 ser 25658, 

          Cmd: ALTER DATABASE

      34: TT01 ospid 14588 sid 251 ser 46152, waiting for 'Data Guard: Timer' 

      35: M000 ospid 10408 sid 374 ser 858, 

      36: M001 ospid 14208 sid 498 ser 26922, waiting for 'class slave wait' (DEAD)

System global information:

     processes: base 0x00007FF94A4BE638, size 640

     allocation: free sessions 0x00007FF94A7FB0E0, free calls 0x0000000000000000

     control alloc errors: 1092 (process), 1092 (session), 1092 (call), 0 (transaction)

这里看出是控制文件出问题了

3.手动创建控制文件

由于之前没有备份，这里直接手动创建控制文件，如下：

CREATE CONTROLFILE REUSE SET DATABASE "XMC" 

            MAXLOGFILES 16

            MAXDATAFILES 100

            MAXINSTANCES 2

            MAXLOGHISTORY 1

            DATAFILE

            'D:\app\xmc\virtual\oradata\xmc\SYSAUX01.DBF',

            'D:\app\xmc\virtual\oradata\xmc\SYSTEM01.DBF',

            'D:\app\xmc\virtual\oradata\xmc\UNDOTBS01.DBF',

            'D:\app\xmc\virtual\oradata\xmc\USERS01.DBF'

            LOGFILE

            GROUP 1 'D:\app\xmc\virtual\oradata\xmc\REDO01.LOG' SIZE 50M,

            GROUP 2 'D:\app\xmc\virtual\oradata\xmc\REDO02.LOG' SIZE 50M,

            GROUP 3 'D:\app\xmc\virtual\oradata\xmc\REDO03.LOG' SIZE 50M RESETLOGS;

控制文件已创建。

这里将DATAFILE防止后面一直出错，到前面就没问题了

然后继续

RMAN> recover database;

从位于 17-11月-20 的 recover 开始

使用目标数据库控制文件替代恢复目录

分配的通道: ORA_DISK_1

通道 ORA_DISK_1: SID=126 设备类型=DISK

正在开始介质的恢复

线程 1 序列 25 的归档日志已作为文件 D:\APP\XMC\VIRTUAL\ORADATA\XMC\REDO02.LOG 存在于磁盘上

归档日志文件名 = D:\APP\XMC\VIRTUAL\ORADATA\XMC\REDO02.LOG 线程 = 1 序列 = 25

介质恢复完成, 用时: 00:00:01

在 17-11月-20 完成了 recover

RMAN> alter database open resetlogs;

已处理语句

RMAN> select instance_name,status from gv$instance;

INSTANCE_NAME    STATUS

---------------- ------------

xmc              OPEN

至此，这个问题解决。