在解决由于一次误操作而引起的Oracle数据库大恢复的过程中， 由于shutdown abort实例有时起不到解决的作用，于是我们就采用杀Oracle进程，来删除其共享内存段的做法，将挂起Oracle数据库实力强制abort。

1、杀Oracle核心进程

引用

[ora10g@testbdump]$ps-ef|grepora_ 
ora10g1443110Feb21?00:01:32ora_pmon_streams 
ora10g1443310Feb21?00:00:46ora_psp0_streams 
ora10g1443510Feb21?00:00:47ora_mman_streams 
ora10g1443710Feb21?00:06:57ora_dbw0_streams 
ora10g1443910Feb21?00:06:24ora_lgwr_streams 
ora10g1444110Feb21?00:46:13ora_ckpt_streams 
ora10g1444310Feb21?00:01:02ora_smon_streams 
ora10g1444510Feb21?00:00:00ora_reco_streams 
ora10g1444710Feb21?00:05:53ora_cjq0_streams 
ora10g1444910Feb21?00:03:15ora_mmon_streams 
ora10g1445110Feb21?00:02:47ora_mmnl_streams 
ora10g1445310Feb21?00:00:01ora_d000_streams 
ora10g1445510Feb21?00:00:03ora_s000_streams 
ora10g1446010Feb21?00:00:05ora_qmnc_streams 
ora10g1446810Feb21?00:00:08ora_q000_streams 
ora10g1447010Feb21?00:00:02ora_q001_streams 
ora10g136221003:08?00:00:00ora_j000_streams 
ora10g1371012028003:13pts/500:00:00grepora_ 
[ora10g@testbdump]$kill-914431144331443514437
14439144411444314445144471444914451144531445514460144681447013622 

2、删除Oracle 共享内存段

引用

[ora10g@testbdump]$ipcs 
------SharedMemorySegments-------- 
keyshmidownerpermsbytesnattchstatus 
0xcc481b8c1441796ora10g64059978547200x40b3b558
2818054ora10g64010779361280 
------SemaphoreArrays-------- 
keysemidownerpermsnsems 
0x0d908ec4360448ora10g640154 
------MessageQueues-------- 
keymsqidownerpermsused-bytesmessages 
[root@test~]#ipcrm-m1441796 
[root@test~]#ipcrm-s360448 

在一次误操作引起的Oracle数据库大恢复的解决办法中再次尝试将实例打开，这里用到了10015事件。

引用

SQL>altersessionsetevents'
10015tracenameadjust_scnlevel1'; 
Sessionaltered. 
SQL>alterdatabaseopen; 
alterdatabaseopen 
* 
ERRORatline1: 
ORA-01113:file1needsmediarecovery 
ORA-01110:datafile1:
'/ora10g/app/oradata/ldbra/system01.dbf' 
SQL>recoverdatabase; 
Mediarecoverycomplete. 
SQL>alterdatabaseopen; 
Databasealtered. 

后续工作就是将tempfile添加到temp表空间中，终于恢复成功。

引用

SQL>altertablespacetempaddtempfile
'/ora10g/app/oradata/ldbra/temp01.dbf'size50mreuse; 
Tablespacealtered.