启动VIP报CRS-1028/CRS-0223致使VIP状态为UNKNOWN故障分析与解决

CRS版本为10.2.0.4 1、VIP State为UNKNOWN [root@XXdb1 ~]# crs_stat -t Name Type Target State Host ------------------------------------------------------------ ora.dxp.db application ONLINE ONLINE XXdb1 ora....p1.inst application ONLINE ONLIN

CRS版本为10.2.0.4

1、VIP State为UNKNOWN

[root@XXdb1 ~]# crs_stat -t

Name Type Target State Host

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

ora.dxp.db application ONLINE ONLINE XXdb1

ora....p1.inst application ONLINE ONLINE XXdb1

ora....p2.inst application ONLINE ONLINE XXdb2

ora....SM1.asm application ONLINE ONLINE XXdb1

ora....B1.lsnr application OFFLINE OFFLINE

ora.XXdb1.gsd application ONLINE ONLINE XXdb1

ora.XXdb1.ons application ONLINE ONLINE XXdb1

ora.XXdb1.vip application ONLINE UNKNOWN XXdb1

ora....SM2.asm application ONLINE ONLINE XXdb2

ora....B2.lsnr application OFFLINE OFFLINE

ora.XXdb2.gsd application ONLINE ONLINE XXdb2

ora.XXdb2.ons application ONLINE ONLINE XXdb2

ora.XXdb2.vip application ONLINE UNKNOWN XXdb2

ora.zsXXdb.db application ONLINE ONLINE XXdb1

ora....b1.inst application ONLINE ONLINE XXdb1

ora....b2.inst application ONLINE ONLINE XXdb2

2、手动启动VIP时报CRS-1028/CRS-0223错误

[root@XXdb1 ~]# crs_start ora.XXdb1.vip

CRS-1028: Dependency analysis failed because of:

'Resource in UNKNOWN state: ora.XXdb1.vip'

CRS-0223: Resource 'ora.XXdb1.vip' has placement error.

3、分析crsd日志

2014-07-05 16:10:08.813: [ CRSAPP][1524836704]0StartResource error for ora.XXdb1.vip error code = -1

2014-07-05 16:10:08.826: [ CRSEVT][1524836704]0CAAMonitorHandler :: 0:Could not execute /home/oracle/product/10.2.0/crs/bin/racgwrap(stop) for ora.XXdb1.vip

category: 1234, operation: scls_canexec, loc: , OS error: 0, other: no exe permission, file /home/oracle/product/10.2.0/crs/bin/racgwrap

发现因为在操作系统上，因为对racgwrap文件没有执行权限

4、对racgwrap文件赋予O=x权限

＃chmod o=x /home/oracle/product/10.2.0/crs/bin/racgwrap

5、再次启动VIP，crsd中报出相关其它错误

2014-07-05 16:26:05.473: [ CRSAPP][1524836704]0StartResource error for ora.XXdb1.vip error code = 1

6、分析vip log

Oracle Database 10g CRS Release 10.2.0.4.0 Production Copyright 1996, 2008 Oracle. All rights reserved.

2014-07-05 17:25:03.565: [ RACG][2540928256] [26035][2540928256][ora.XXdb1.vip]: clsrcexecut: scls_process_spawn() error, status = 1, command = /home/oracle/product/10.2.0/crs/bin/racgeut

Category: 1234

Operation: scls_canexec

Location:

Other: no exe permission, file /home/oracle/product/10.2.0/crs/bin/racgeut

OS Error: Error 0

2014-07-05 17:25:03.565: [ RACG][2540928256] [26035][2540928256][ora.XXdb1.vip]: clsrcexecut: scls_process_spawn() error, status = 1, command = /home/oracle/product/10.2.0/crs/bin/racgeut

Category: 1234

Operation: scls_canexec

Location:

Other: no exe permission, file /home/oracle/product/10.2.0/crs/bin/racgeut

OS Error: Error 0

2014-07-05 17:25:03.566: [ RACG][2540928256] [26035][2540928256][ora.XXdb1.vip]: end for resource = ora.XXdb1.vip, action = start, status = 1, time = 0.030s

2014-07-05 17:25:03.634: [ RACG][2540928256] [26040][2540928256][ora.XXdb1.vip]: clsrcrundetach: scls_process_spawn() error, status = 1, command = /home/oracle/product/10.2.0/crs/bin/racgmain

Category: 1234

Operation: scls_canexec

Location:

Other: no exe permission, file /home/oracle/product/10.2.0/crs/bin/racgmain

OS Error: Error 0

从vip log中，又发现，最少存在有对racgeut、racgmain两个文件没有执行权限

7、赋予更多的权限解决

看来不止止是一两个文件不具有权限了，为了尽快解决问题，决定对/home/oracle/product/10.2.0/crs/bin/*都主赋予执行权限，对极对/home/oracle/product/10.2.0/crs/bin/racg*的拥有者改更改成root用户：

#chmod o=x /home/oracle/product/10.2.0/crs/bin/*

#chown root /home/oracle/product/10.2.0/crs/bin/racg*

到此再启动vip，已可以正常启动

本文作者：踩点，从事”系统架构、操作系统、存储设备、数据库、中间件、应用程序“六个层面系统性的性能优化工作

欢迎加入 系统性能优化专业群 ，共同探讨性能优化技术。群号：258187244