安装Oracle 11g RAC时，我们需要配置ntp服务。在使用虚拟机的情况下对于时钟同步方式的配置有很多种方式，可以使用vmware自带的时钟同步功能，也可以直接将本地的一个节点用作时间服务器。本文介绍直接配置ntp方式的时钟服务器。

1、查看两节点的hosts配置
 [root@node1 ~]# cat /etc/hosts
 # Do not remove the following line, or various programs
 # that require network functionality will fail.
 #127.0.0.1              localhost.localdomain localhost
 #::1            localhost6.localdomain6 localhost6

 127.0.0.1       localhost.szdb.com   localhost
 # Public eth0
 192.168.7.71   node1.szdb.com        node1
 192.168.7.72   node2.szdb.com        node2

 #Private eth1
 10.10.7.71   node1-priv.szdb.com   node1-priv
 10.10.7.72   node2-priv.szdb.com   node2-priv

 #Virtual
 192.168.7.81   node1-vip.szdb.com    node1-vip
 192.168.7.82   node2-vip.szdb.com    node2-vip

2、确认各节点的ntp包已经安装
 [oracle@node1 ~]$ rpm -qa | grep ntp
 ntp-4.2.2p1-9.el5_4.1
 chkfontpath-1.10.1-1.1      #这个是和字体有关，非ntp包
 [oracle@node1 ~]$ ssh node2 rpm -qa | grep ntp
 ntp-4.2.2p1-9.el5_4.1
 chkfontpath-1.10.1-1.1      #这个是和字体有关，非ntp包

3、编辑两节点的ntp.conf文件
 [oracle@node1 ~]$ su - root
 Password:
 [root@node1 ~]#  vi /etc/ntp.conf

 #New ntp server added by Robinson
 server  127.127.1.0 prefer  # 添加首选的时钟服务器
 restrict 192.168.7.0  mask 255.255.255.255 nomodify notrap #只允许192.168.7.*网段的客户机进行时间同步
 broadcastdelay 0.008

 [root@node2 ~]# vi /etc/ntp.conf

 #New ntp server added by Robinson
 server 192.168.7.71 prefer
 broadcastdelay 0.008

4、编辑两节点的ntpd参数
 [root@node1 ~]# vi /etc/sysconfig/ntpd
 #The following item added by Robinson
 #Set to 'yes' to sycn hw clock after successful ntpdate
 SYNC_HWCLOCK=yes      #此选项用于自动校准系统时钟与硬件时钟
 OPTIONS="-x -u ntp:ntp -p /var/run/ntpd.pid"

 #注意理解Linux的时钟类型。在Linux系统中分为系统时钟和硬件时钟.
 #系统时钟指当前Linux kernel中的时钟，而硬件时钟指的是BIOS时钟，由主板电池供电的那个时钟
 #当Linux启动时，硬件时钟会读取系统时钟的设置，之后系统时钟就独立于硬件时钟运作

 [root@node2 ~]# vi /etc/sysconfig/ntpd
 The following item added by Robinson
 SYNC_HWCLOCK=yes
 OPTIONS="-x -u ntp:ntp -p /var/run/ntpd.pid"

5、配置ntp自启动服务
 [root@node1 ~]# chkconfig ntpd on
 [root@node2 ~]# chkconfig ntpd on


6、在两节点启动ntp服务
 [root@node1 ~]# service ntpd stop
 Shutting down ntpd: [FAILED]
 [root@node1 ~]# service ntpd start
 ntpd: Synchronizing with time server: [FAILED]
 Starting ntpd: [  OK  ]

 [root@node2 ~]# service ntpd restart
 Shutting down ntpd: [  OK  ]
 ntpd: Synchronizing with time server: [  OK  ]
 Syncing hardware clock to system time [  OK  ]
 Starting ntpd: [  OK  ]

7、查看ntp状态
 [root@node1 ~]# ntpq -p
      remote           refid      st t when poll reach   delay   offset  jitter
 ==============================================================================
  LOCAL(0)        .LOCL.          10 l   40   64    1    0.000    0.000   0.001

 [root@node2 ~]# ntpq -p
      remote           refid      st t when poll reach   delay   offset  jitter
 ==============================================================================
  node1.szdb.com  .INIT.          16 u   60   64    0    0.000    0.000   0.000
  LOCAL(0)        .LOCL.          10 l   59   64    1    0.000    0.000   0.001

 #Author : Robinson
 #Blog: http://blog.csdn.net/robinson_0612
 也可以使用watch ntpq -p方式查看实时状态

8、ntp的相关日志

[root@bigboy tmp]# cat /var/log/messages | grep ntpd
转：http://blog.csdn.net/leshami/article/details/8268079