安装ORACLE_RAC遇到的问题与解决方法

while running:

/u01/app/oracle/product/10.2.0/db_1/root.sh

Checking to see if Oracle CRS stack is already configured
/etc/oracle does not exist. Creating it now.
Setting the permissions on OCR backup directory
Setting up NS directories
PROT-1:Failedtoinitializeocrconfig
Failedto upgrade Oracle Cluster Registry configuration

        dd if=/dev/zero f=/dev/rdsk/V1064_vote_01_20m.dbf bs=8192 count=2560
        dd if=/dev/zero f=/dev/rdsk/ocrV1064_100m.ora bs=8192 count=12800

解决方法1
Failedto upgrade Oracle Cluster Registry configuration
在安装CRS时,在第二个节点执行./root.sh时,出现如下提示,我在第一个节点执行正常.请大虾指点一些,不胜感激!谢谢!
[root@RACtest2 crs]# ./root.sh
WARNING: directory '/app/oracle/product/10.2.0' is not owned by root
WARNING: directory '/app/oracle/product' is not owned by root
WARNING: directory '/app/oracle' is not owned by root
WARNING: directory '/app' is not owned by root
Checking to see if Oracle CRS stack is already configured
Setting the permissions on OCR backup directory
Setting up NS directories
PROT-1:Failedtoinitializeocrconfig
Failedto upgrade Oracle Cluster Registry configuration
错误原因:
是因为安装crs的设备权限有问题,例如我的设备用raw来放置ocr和vote,此时要设置好这些硬件设备以及连接的文件的权限,下面是我的环境:
[root@rac2 oracrs]#
lrwxrwxrwx 1 root root 13 Jan 27 12:49 ocr.crs -> /dev/raw/raw1
lrwxrwxrwx 1 root root 13 Jan 26 13:31 vote.crs -> /dev/raw/raw2
chown root:oinstall /dev/raw/raw1
chown root:oinstall /dev/raw/raw2
chmod 660 /dev/raw/raw1
chmod 660 /dev/raw/raw2
其中/dev/sdb1放置ocr,/dev/sdb2放置vote.
[root@rac2 oracrs]# service rawdevices reload
Assigning devices:
           /dev/raw/raw1 -->   /dev/sdb1
/dev/raw/raw1: bound to major 8, minor 17
           /dev/raw/raw2 -->   /dev/sdb2
/dev/raw/raw2: bound to major 8, minor 18
Done
然后再次执行就ok了.
[root@rac2 oracrs]# /oracle/app/oracle/product/crs/root.sh
WARNING: directory '/oracle/app/oracle/product' is not owned by root
WARNING: directory '/oracle/app/oracle' is not owned by root
Checking to see if Oracle CRS stack is already configured
Setting the permissions on OCR backup directory
Setting up NS directories
Oracle Cluster Registry configuration upgraded successfully
WARNING: directory '/oracle/app/oracle/product' is not owned by root
WARNING: directory '/oracle/app/oracle' is not owned by root
clscfg: EXISTING configuration version 3 detected.
clscfg: version 3 is 10G Release 2.
assigning default hostname rac1 for node 1.
assigning default hostname rac2 for node 2.
Successfully accumulated necessary OCR keys.
Using ports: CSS=49895 CRS=49896 EVMC=49898 and EVMR=49897.
node <nodenumber>: <nodename> <private interconnect name> <hostname>
node 1: rac1 priv1 rac1
node 2: rac2 priv2 rac2
clscfg: Arguments check out successfully.

 

Oracle gave me a patch that allowed us to format the OCR and voting disk. Now the problem may just be I need to run root.sh on node1 or wipe both clean and start fresh.
Devices formatted, now we need to get the CRS daemon up and running on both nodes..
[root@dr2db2 ~]# dd if=/dev/zero f=/dev/raw/raw2 bs=1048576 count=1000
1000+0 records in
1000+0 records out
[root@dr2db2 ~]# /orahome/app/oracle/product/10.1.2.0.2/CRS/root.sh
WARNING: directory '/orahome/app/oracle/product/10.1.2.0.2' is not owned by root
WARNING: directory '/orahome/app/oracle/product' is not owned by root
WARNING: directory '/orahome/app/oracle' is not owned by root
WARNING: directory '/orahome/app' is not owned by root
Checking to see if Oracle CRS stack is already configured
Setting the permissions on OCR backup directory
Setting up NS directories
Oracle Cluster Registry configuration upgraded successfully
WARNING: directory '/orahome/app/oracle/product/10.1.2.0.2' is not owned by root
WARNING: directory '/orahome/app/oracle/product' is not owned by root
WARNING: directory '/orahome/app/oracle' is not owned by root
WARNING: directory '/orahome/app' is not owned by root
Successfully accumulated necessary OCR keys.
Using ports: CSS=49895 CRS=49896 EVMC=49898 and EVMR=49897.
node <nodenumber>: <nodename> <private interconnect name> <hostname>
node 1: dr2db2 dr2db2-eth2 dr2db2
node 2: dr2db1 dr2db1-eth2 dr2db1
Creating OCR keys for user 'root', privgrp 'root'..
Operation successful.
Now formatting voting device: /dev/raw/raw6
Format of 1 voting devices complete.
Startup will be queued to init within 90 seconds.
Adding daemons to inittab
Expecting the CRS daemons to be up within 600 seconds.
CSS is active on these nodes.
dr2db2
CSS is inactive on these nodes.
dr2db1
Local node checking complete.
Run root.sh on remaining nodes to start CRS daemons.
[root@dr2db2 ~]# ps -ef | grep ora
root 6572 2526 0 13:39 ? 00:00:00 sshd: oracle [priv]
oracle 6574 6572 0 13:39 ? 00:00:00 sshd: oracle@pts/1
oracle 6575 6574 0 13:39 pts/1 00:00:00 -bash
root 14172 2526 0 17:54 ? 00:00:00 sshd: oracle [priv]
oracle 14176 14172 0 17:55 ? 00:00:00 sshd: oracle@pts/2
oracle 14177 14176 0 17:55 pts/2 00:00:00 -bash
root 14682 1 0 17:57 ? 00:00:00 /bin/su -l oracle -c sh -c 'ulimit -c unlimited; cd /orahome/app/oracle/product/10.1.2.0.2/CRS/log/dr2db2/evmd; exec /orahome/app/oracle/product/10.1.2.0.2/CRS/bin/evmd '
root 14686 1 0 17:57 ? 00:00:00 /orahome/app/oracle/product/10.1.2.0.2/CRS/bin/crsd.bin reboot
oracle 14959 14682 0 17:58 ? 00:00:00 /orahome/app/oracle/product/10.1.2.0.2/CRS/bin/evmd.bin
root 15018 14942 0 17:58 ? 00:00:00 /bin/su -l oracle -c /bin/sh -c 'ulimit -c unlimited; cd /orahome/app/oracle/product/10.1.2.0.2/CRS/log/dr2db2/cssd; /orahome/app/oracle/product/10.1.2.0.2/CRS/bin/ocssd || exit $?'
oracle 15021 15018 0 17:58 ? 00:00:00 /bin/sh -c ulimit -c unlimited; cd /orahome/app/oracle/product/10.1.2.0.2/CRS/log/dr2db2/cssd; /orahome/app/oracle/product/10.1.2.0.2/CRS/bin/ocssd || exit $?
oracle 15057 15021 0 17:58 ? 00:00:00 /orahome/app/oracle/product/10.1.2.0.2/CRS/bin/ocssd.bin
root 15996 14206 0 17:59 pts/2 00:00:00 grep ora
crs_setperm crs_setperm.bin crs_start crs_start.bin crs_stat crs_stat.bin crs_stop crs_stop.bin
[root@dr2db2 ~]# /orahome/app/oracle/product/10.1.2.0.2/CRS/bin/crs_stat -t
CRS-0184: Cannot communicate with the CRS daemon.
on next node
[root@dr2db1 ~]# /orahome/app/oracle/product/10.1.2.0.2/CRS/root.sh
WARNING: directory '/orahome/app/oracle/product/10.1.2.0.2' is not owned by root
WARNING: directory '/orahome/app/oracle/product' is not owned by root
WARNING: directory '/orahome/app/oracle' is not owned by root
WARNING: directory '/orahome/app' is not owned by root
Checking to see if Oracle CRS stack is already configured
/etc/oracle does not exist. Creating it now.
Setting the permissions on OCR backup directory
Setting up NS directories
Oracle Cluster Registry configuration upgraded successfully
WARNING: directory '/orahome/app/oracle/product/10.1.2.0.2' is not owned by root
WARNING: directory '/orahome/app/oracle/product' is not owned by root
WARNING: directory '/orahome/app/oracle' is not owned by root
WARNING: directory '/orahome/app' is not owned by root
/orahome/app/oracle/product/10.1.2.0.2/CRS/bin/crsctl.bin: error while loading shared libraries: libstdc++.so.5: cannot open shared object file: No such file or directory
Failure initializing entries in /etc/oracle/scls_scr/dr2db1.
I think whats needed is a clean install.

http://www.puschitz.com/InstallingOracle10gRAC.shtml#CreatingPartitionsForRawDevices

在 node1 上执行：/opt/ora10g/product/10.2.0/crs_1/root.sh; 在 node2 上执行：/opt/ora10g/product/10.2.0/crs_1/root.sh; 

通常在最后一个节点执行root.sh时会遇到错误，就我们的情况而言当然就是node2~~

提示：一般常见的错误有如下三种：

A).如果你碰到了这个错误：
/opt/ora10g/product/10.2.0/crs_1/jdk/jre//bin/java: error while loading shared libraries:  libpthread.so.0: cannot open shared object file: No such file or directory
可以按照如下方式解决：
===============================
修改vipca文件
[root@node2 opt]# vi /opt/ora10g/product/10.2.0/crs_1/bin/vipca
找到如下内容： 
       Remove this workaround when the bug 3937317 is fixed 
       arch=`uname -m` 
       if [ "$arch" = "i686" -o "$arch" = "ia64" ] 
       then 
            LD_ASSUME_KERNEL=2.4.19 
            export LD_ASSUME_KERNEL 
       fi 
       #End workaround
在fi后新添加一行：
unset LD_ASSUME_KERNEL
以及srvctl文件
[root@node2 opt]# vi /opt/ora10g/product/10.2.0/crs_1/bin/srvctl
找到如下内容：
LD_ASSUME_KERNEL=2.4.19
export LD_ASSUME_KERNEL
同样在其后新增加一行：
unset LD_ASSUME_KERNEL
保存退出，然后在node2重新执行root.sh
当然，既然我们已经知道了有这个问题，建议最好在node2执行root.sh之前，首先修改vipca。
其实同时需要你改的还有$ORACLE_HOME/bin/srvctl文件，不然等装完数据库之后，srvctl命令也是会报这个错误地。要知道srvctl这么常用，如果它执行老报错，那可是相当致命啊。不过呢你现在才安装到crs，离create db还远着呢，大可以等到创建完数据库，待到需要管理时再修改该文件。

B).如果你碰到了这个错误：
The given interface(s), "eth0" is not public. Public interfaces should  be used to configure virtual IPs.
解决方式如下：
==============================
图形界面上运行$CRS_HOME/bin/vipca，手工重新配置rac1-vip和rac2-vip。
[root@node2 opt]# xhost +
[root@node2 opt]# /opt/ora10g/product/10.2.0/crs_1/bin/vipca
按照提示点击下一步
点击finish即可vipca开始自动配置
全部配置完成之后，点击exit退出操作窗口。

C).如果你碰到了这个错误：
Error 0(Native: listNetInterfaces:[3]) 
 [Error 0(Native: listNetInterfaces:[3])]
解决方式如下：
===============================
[root@node2 bin]# ./oifcfg iflist
eth1  10.10.17.0
virbr0  192.168.122.0
eth0  192.168.100.0
[root@node2 bin]# ./oifcfg setif -global eth0/192.168.100.0:public[root@node2 bin]# ./oifcfg setif -global eth1/10.10.10.0:cluster_ interconnect
[root@node2 bin]# ./oifcfg getif
eth0  192.168.100.0  global  public
eth1  10.10.10.0  global  cluster_interconnect
然后在视窗界面重新执行vipca即可，如上b例中所示。