RAC下主机修改时区导致db无法open

11.2.0.4.rac 主机修改时区导致db无法open 

os:redhat linux

db:11.2.0.4 rac

desc:crs启动后，除db外都可以正常启动。

查看crs和db的log，文件最后修改时间跟主机时间一致，但log里的时间不一致，如下为crs和dblog里的日志

crs log

==========================================

2015-05-04 10:30:24.139:

[client(19430)]CRS-4743:File /u02/app/11.2.0/grid/oc4j/j2ee/home/OC4J_DBWLM_config/system-jazn-data.xml was updated from OCR(Size: 13365(New), 13378(Old) bytes)

2015-05-04 10:31:33.583:

[cssd(18287)]CRS-1601:CSSD Reconfiguration complete. Active nodes are gzsjdb1 gzsjdb2 .

2015-05-04 10:31:56.583:

[crsd(19096)]CRS-2772:Server 'gzsjdb2' has been assigned to pool 'Generic'.

2015-05-04 10:31:56.584:

[crsd(19096)]CRS-2772:Server 'gzsjdb2' has been assigned to pool 'ora.jzxn'. #出问题时日志

2015-05-04 16:31:02.619: #问题解决后的日志

[client(26295)]CRS-10051:CVU found following errors with Clusterware setup : PRVF-7573 : Sufficient swap size is not available on node "gzsjdb1" [Required = 16GB (1.6777216E7KB) ; Found = 10GB (1.0485752E7KB)]

PRVF-5415 : Check to see if NTP daemon or service is running failed

PRVG-1101 : SCAN name "scan-cl

db log

============================

Fri Jun 12 03:45:16 2015 #出问题时日志

Thread 1 advanced to log sequence 54 (LGWR switch)

  Current log# 3 seq# 54 mem# 0: /oradata/oradata/jzxn/redo03.log

Fri Jun 12 03:47:38 2015

Thread 1 cannot allocate new log, sequence 55

Checkpoint not complete

  Current log# 3 seq# 54 mem# 0: /oradata/oradata/jzxn/redo03.log

Fri Jun 12 03:47:43 2015

opidcl aborting process unknown ospid (29020) as a result of ORA-28

Fri Jun 12 05:13:24 2015

Thread 1 advanced to log sequence 55 (LGWR switch)

  Current log# 1 seq# 55 mem# 0: /oradata/oradata/jzxn/redo01.log

Fri Jun 12 14:31:04 2015  #问题解决后的日志

Starting ORACLE instance (normal)

************************ Large Pages Information *******************

Per process system memlock (soft) limit = UNLIMITED

Total Shared Global Region in Large Pages = 0 KB (0%)

解决过程：起初没有注意这个时间问题，对crs重启后，仍然是db的资源不正常，后发现日志时间不对，经询问项目组节点2由默认时间(Etc/GMT-1)改为东8区(Asia/Shanghai)的时区后，节点2就无法连接db，节点1修改时区后仍可连接。直至今天节点1也无法连接db.

出现db日志、crs日志等和主机时间不一致，是因为在rac新颁布中，grid也有属于自己的时区，在安装过程中，默认值与操作系统的TZ环境变量相同。

本例是因为时间不一致，导致db无法open，修改grid时区后，重启crs，db可以正常open.

1.查看系统时区

$cat /etc/sysconfig/clock

ZONE=Asia/Shanghai

2.查看crs里配置的时区

[root@gzsjdb1 install]# cat s_crsconfig_gzsjdb1_env.txt.bak

### This file can be used to modify the NLS_LANG environment variable, which determines the charset to be used for messages.

### For example, a new charset can be configured by setting NLS_LANG=JAPANESE_JAPAN.UTF8 

### Do not modify this file except to change NLS_LANG, or under the direction of Oracle Support Services

TZ=Etc/GMT-1 #仍然为默认的美国时区

NLS_LANG=AMERICAN_AMERICA.AL32UTF8

TNS_ADMIN=

ORACLE_BASE=

3.修改crs的时区为Asia/Shanghai

[root@gzsjdb1 install]# cat s_crsconfig_gzsjdb1_env.txt

### This file can be used to modify the NLS_LANG environment variable, which determines the charset to be used for messages.

### For example, a new charset can be configured by setting NLS_LANG=JAPANESE_JAPAN.UTF8 

### Do not modify this file except to change NLS_LANG, or under the direction of Oracle Support Services

TZ=Asia/Shanghai

NLS_LANG=AMERICAN_AMERICA.AL32UTF8

TNS_ADMIN=

ORACLE_BASE=

[root@gzsjdb1 install]# pwd

/u02/app/11.2.0/grid/crs/install

[root@gzsjdb1 install]#

4.重启crs

5.经验证crs日志，db日志等都与主机保持一致，通过plsqldev可以登录数据库。

[root@gzsjdb2 ~]# date

Fri Jun 12 15:12:58 CST 2015

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/26964624/viewspace-2565013/，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/26964624/viewspace-2565013/