关闭oracle异常：ORA-01012: not logged on
今天是2014年的最后一天，此时此刻加班中，公司大厦停电，所有的服务器都要关闭。
在对一个生产库做shutdown immediate 的时候，可能是因为网络掉线了，导致正在shutdown的时候出现异常。当时看着一直卡在shutdown immediate，本以为是需要的时间长一点，但是十分钟后还是卡在那里，于是去看alert，结果alert也没有更新：
Private strand flush not complete
Current log# 7 seq# 92151 mem# 0: /u02/logfile/redo71.log
Current log# 7 seq# 92151 mem# 1: /u03/logfile/redo72.log
Thread 1 advanced to log sequence 92152 (LGWR switch)
Current log# 8 seq# 92152 mem# 0: /u04/logfile/redo81.log
Current log# 8 seq# 92152 mem# 1: /u02/logfile/redo82.log
Wed Dec 31 20:02:45 2014
Shutting down instance (immediate)
Stopping background process SMCO
Shutting down instance: further logons disabled
Wed Dec 31 20:03:05 2014
Stopping background process CJQ0
Wed Dec 31 20:03:05 2014
Stopping background process QMNC
Stopping background process MMNL
Stopping background process MMON
License high water mark = 153
All dispatchers and shared servers shutdown
ALTER DATABASE CLOSE NORMAL
然后想重新sqlplus 连进入 去再次shutdown。可以出现这样的情况：
[oracle@zhanglin ~]$ export ORACLE_SID=zhanglin
[oracle@zhanglin ~]$ hostname
zhanglin
[oracle@zhanglin ~]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.3.0 Production on Wed Dec 31 20:11:57 2014
Copyright (c) 1982, 2011, Oracle. All rights reserved.
Connected.
SQL> shutdown
ORA-01012: not logged on
SQL>
SQL> startup
ORA-01012: not logged on
SQL>
查看后台进程时oracle还在：
[oracle@zhanglin ~]$ ps -ef | grep oracle
root 14422 25243 0 18:06 ? 00:00:00 sshd: oracle [priv]
oracle 14424 14422 0 18:06 ? 00:00:00 sshd: oracle@pts/2
oracle 14425 14424 0 18:06 pts/2 00:00:00 -bash
oracle 16150 1 6 20:02 ? 00:00:36 oraclezhanglin (DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))
root 16165 25243 0 20:08 ? 00:00:00 sshd: oracle [priv]
root 16167 25243 0 20:08 ? 00:00:00 sshd: oracle [priv]
oracle 16169 16165 0 20:08 ? 00:00:00 sshd: oracle@pts/4
oracle 16170 16167 0 20:08 ? 00:00:00 sshd: oracle@pts/1
oracle 16171 16170 0 20:08 pts/1 00:00:00 -bash
oracle 16172 16169 0 20:08 pts/4 00:00:00 -bash
oracle 16230 16171 0 20:09 pts/1 00:00:00 tail -f alert_zhanglin.log
oracle 16280 16172 0 20:11 pts/4 00:00:00 sqlplus as sysdba
root 16284 25243 0 20:11 ? 00:00:00 sshd: oracle [priv]
oracle 16286 16284 0 20:11 ? 00:00:00 sshd: oracle@pts/3
oracle 16287 16286 0 20:11 pts/3 00:00:00 -bash
oracle 16316 16287 0 20:11 pts/3 00:00:00 tail -f alert_zhanglin.log
oracle 16318 16280 0 20:11 ? 00:00:00 oraclezhanglin (DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))
oracle 16320 16280 0 20:11 pts/4 00:00:00 /bin/bash
oracle 16346 16320 0 20:11 pts/4 00:00:00 sqlplus as sysdba
oracle 16348 16346 0 20:11 ? 00:00:00 oraclezhanglin (DESCRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))
oracle 16378 16346 0 20:12 pts/4 00:00:00 /bin/bash
oracle 16402 16378 0 20:12 pts/4 00:00:00 ps -ef
oracle 16403 16378 0 20:12 pts/4 00:00:00 grep oracle
oracle 25738 1 0 Dec25 ? 00:00:50 ora_pmon_zhanglin
oracle 25740 1 0 Dec25 ? 00:00:12 ora_psp0_zhanglin
oracle 25742 1 0 Dec25 ? 00:00:00 ora_vktm_zhanglin
oracle 25746 1 0 Dec25 ? 00:00:00 ora_gen0_zhanglin
oracle 25748 1 0 Dec25 ? 00:00:00 ora_diag_zhanglin
oracle 25750 1 0 Dec25 ? 00:00:00 ora_dbrm_zhanglin
oracle 25752 1 0 Dec25 ? 00:43:48 ora_dia0_zhanglin
oracle 25754 1 0 Dec25 ? 00:00:21 ora_mman_zhanglin
oracle 25756 1 0 Dec25 ? 00:09:17 ora_dbw0_zhanglin
oracle 25758 1 0 Dec25 ? 00:08:18 ora_dbw1_zhanglin
oracle 25760 1 0 Dec25 ? 00:08:17 ora_dbw2_zhanglin
oracle 25762 1 0 Dec25 ? 00:08:26 ora_dbw3_zhanglin
oracle 25764 1 0 Dec25 ? 00:08:01 ora_dbw4_zhanglin
oracle 25766 1 0 Dec25 ? 00:08:37 ora_dbw5_zhanglin
oracle 25768 1 0 Dec25 ? 00:07:47 ora_dbw6_zhanglin
oracle 25770 1 0 Dec25 ? 00:08:01 ora_dbw7_zhanglin
oracle 25772 1 0 Dec25 ? 00:08:24 ora_dbw8_zhanglin
oracle 25774 1 0 Dec25 ? 00:08:10 ora_dbw9_zhanglin
oracle 25776 1 0 Dec25 ? 00:08:29 ora_dbwa_zhanglin
oracle 25778 1 0 Dec25 ? 00:08:40 ora_dbwb_zhanglin
oracle 25780 1 0 Dec25 ? 00:08:26 ora_dbwc_zhanglin
oracle 25782 1 0 Dec25 ? 00:07:55 ora_dbwd_zhanglin
oracle 25784 1 0 Dec25 ? 00:08:15 ora_dbwe_zhanglin
oracle 25786 1 0 Dec25 ? 00:08:40 ora_dbwf_zhanglin
oracle 25788 1 0 Dec25 ? 01:01:25 ora_lgwr_zhanglin
oracle 25790 1 0 Dec25 ? 00:06:00 ora_ckpt_zhanglin
oracle 25792 1 0 Dec25 ? 00:01:41 ora_smon_zhanglin
oracle 25794 1 0 Dec25 ? 00:00:00 ora_reco_zhanglin
此时就没辙了。
没办法不得不采用强制手段了。
1、shutdown abort
2、kill -9 oracle 进程 方法：ps -ef|grep ora_dbw0_$Oracle_SID 找到这个进程 直接kill 掉。
[oracle@zhanglin dblog]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.3.0 Production on Wed Dec 31 20:22:58 2014
Copyright (c) 1982, 2011, Oracle. All rights reserved.
Connected.
SQL> shutdown abort
ORACLE instance shut down.
SQL>
后来startup了一次，可以正常起来，然后在shutdown immediate 正常关闭。谢天谢地！
2015年1月1号，元旦三天 加班中度过。。。祝所有的技术人员 新年快乐，事事顺利！
来自 “ ITPUB博客 ” ，链接：https://blog.itpub.net/30936525/viewspace-2016673/ ，如需转载，请注明出处，否则将追究法律责任。
