同事在安装REDHAT AS5操作系统RAC时出现如下故障现象:
有一个ASM节点始终启动不了,使用crs_stop -all后,然后使用crs_start -all。只有一个ASM节点能够启动,如下图情况:
[oracle@yxgisdb1 bin]$ crs_stat -t
Name Type Target State Host
------------------------------------------------------------
ora....SM1.asm application ONLINE OFFLINE
ora....B1.lsnr application ONLINE ONLINE yxgisdb1
ora....db1.gsd application ONLINE ONLINE yxgisdb1
ora....db1.ons application ONLINE ONLINE yxgisdb1
ora....db1.vip application ONLINE ONLINE yxgisdb1
ora....SM2.asm application ONLINE ONLINE yxgisdb2
ora....B2.lsnr application ONLINE ONLINE yxgisdb2
ora....db2.gsd application ONLINE ONLINE yxgisdb2
ora....db2.ons application ONLINE ONLINE yxgisdb2
ora....db2.vip application ONLINE ONLINE yxgisdb
操作系统版本:
[root@yxgisdb1 ~]# uname -a
Linux yxgisdb1 2.6.18-164.el5xen #1 SMP Tue Aug 18 15:59:52 EDT 2009 x86_64 x86_64 x86_64 GNU/Linux
数据库版本是10.2.0.1。
察看日志报错:
2011-06-04 08:13:29.826: [ RACG][2952254144] [5025][2952254144][ora.yxgisdb1.ASM1.asm]: clsrcexecut: rc = 99, time = 540.240s
2011-06-04 08:13:29.826: [ RACG][2952254144] [5025][2952254144][ora.yxgisdb1.ASM1.asm]: instance startup timeout, startup the instance in background
2011-06-04 08:13:29.826: [ RACG][2952254144] [5025][2952254144][ora.yxgisdb1.ASM1.asm]: clsrcasmapp: failed to initiate ASM startup [99]
2011-06-04 08:19:58.786: [ RACG][9933504] [11557][9933504][ora.yxgisdb1.ASM1.asm]:
SQL*Plus: Release 10.2.0.4.0 - Production on Sat Jun 4 08:19:52 2011
Copyright (c) 1982, 2007, Oracle. All Rights Reserved.
Enter user-name: Connected.
SQL> ASM instance shutdown
SQL> Disconnected
为什么会初始化失败?
察看一下网卡配置:
cd /oracle/product/10.2.0/crs_1/bin
[oracle@yxgisdb2 bin]$ ./oifcfg getif
usb0 169.254.95.0 global public
eth3 192.168.0.0 global cluster_interconnect
virbr0 192.168.122.0 global cluster_interconnect
eth2 10.168.174.0 global public
显然是XEN惹的祸:
[oracle@yxgisdb2 bdump]$ cd /oracle/product/10.2.0/crs_1/bin
[oracle@yxgisdb2 bin]$ ./oifcfg getif
usb0 169.254.95.0 global public
eth3 192.168.0.0 global cluster_interconnect
virbr0 192.168.122.0 global cluster_interconnect
eth2 10.168.174.0 global public
[oracle@yxgisdb2 bin]$ ./oifcfg delif -global virbr0/192.168.122.0
[oracle@yxgisdb2 bin]$ ./oifcfg getif
usb0 169.254.95.0 global public
eth3 192.168.0.0 global cluster_interconnect
eth2 10.168.174.0 global public
[oracle@yxgisdb2 bin]$ ./oifcfg delif -global usb0/169.254.95.0
[oracle@yxgisdb2 bin]$ ./oifcfg getif
eth3 192.168.0.0 global cluster_interconnect
eth2 10.168.174.0 global public
问题解决。
重新启动CRS,如下情况:
[oracle@yxgisdb1 bin]$ crs_stat -t
Name Type Target State Host
------------------------------------------------------------
ora....SM1.asm application ONLINE ONLINE yxgisdb1
ora....B1.lsnr application ONLINE ONLINE yxgisdb1
ora....db1.gsd application ONLINE ONLINE yxgisdb1
ora....db1.ons application ONLINE ONLINE yxgisdb1
ora....db1.vip application ONLINE ONLINE yxgisdb1
ora....SM2.asm application ONLINE ONLINE yxgisdb2
ora....B2.lsnr application ONLINE ONLINE yxgisdb2
ora....db2.gsd application ONLINE ONLINE yxgisdb2
ora....db2.ons application ONLINE ONLINE yxgisdb2
ora....db2.vip application ONLINE ONLINE yxgisdb2
问题解决。总结一下,由于XEN,虚拟化出一个
virbr0 192.168.122.0 global cluster_interconnect
和
usb0 169.254.95.0 global public
导致。删除后,就正常了,可以继续安装数据库。
来自 “ ITPUB博客 ” ,链接:http://blog.itpub.net/8334342/viewspace-697791/,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/8334342/viewspace-697791/