广东rac库一个节点crs无法启动,远程过去如下检查。
[root@GDDBA ~]# fdisk -l
盘阵可以认到

[moe@GDDBA ~]$ crsctl check css
Failure 1 contacting CSS daemon
Cannot communicate with CRS
Cannot communicate with EVM

检查crs进程
[moe@GDDBA ~]$ ps -ef|grep css
root      6929     1  0 19:56 ?        00:00:00 /bin/sh /etc/init.d/init.cssd fatal
root      6960  6928  0 19:56 ?        00:00:00 /bin/sh /etc/init.d/init.cssd startcheck
root      6963  6929  0 19:56 ?        00:00:00 /bin/sh /etc/init.d/init.cssd startcheck
root      7064  6935  0 19:56 ?        00:00:00 /bin/sh /etc/init.d/init.cssd startcheck

从上可以看出init.cssd停在startcheck中,并没运行ocssd.bin daemon


检查crsd.log和ocssd.log,没发现有用信息。
在$ORA_CRS_HOME/log/gddba/client中如下:
[moe@GDDBA client]$ more css308.log
Oracle Database 10g CRS Release 10.2.0.4.0 Production Copyright 1996, 2008 Oracle.  All rights reserved.
2012-05-23 14:58:17.482: [ CSSCLNT][2401135200]clsssInitNative: connect failed, rc 9

[moe@GDDBA client]$ more clsc3.log
Oracle Database 10g CRS Release 10.2.0.4.0 Production Copyright 1996, 2008 Oracle.  All rights reserved.
2012-05-23 14:58:03.907: [ COMMCRS][1664445024]clsc_connect: (0x3ab9640) no listener at (ADDRESS=(PROTOCOL=IPC)(KEY=CRSD_UI_SOCKET))

2012-05-23 14:58:05.554: [ COMMCRS][1664445024]clsc_connect: (0x3ab4590) no listener at (ADDRESS=(PROTOCOL=ipc)(KEY=SYSTEM.evm.acceptor.auth))

[moe@GDDBA client]$

检查/var/log/message
May 23 15:33:46 GDDBA logger: Cluster Ready Services waiting on dependencies. Diagnostics in /tmp/crsctl.7002.
May 23 15:33:46 GDDBA logger: Cluster Ready Services waiting on dependencies. Diagnostics in /tmp/crsctl.7181.
May 23 15:34:48 GDDBA logger: Cluster Ready Services waiting on dependencies. Diagnostics in /tmp/crsctl.7203.
May 23 15:34:48 GDDBA logger: Cluster Ready Services waiting on dependencies. Diagnostics in /tmp/crsctl.7293.
May 23 15:34:48 GDDBA logger: Cluster Ready Services waiting on dependencies. Diagnostics in /tmp/crsctl.7348.

[root@GDDBA ~]# more /tmp/crsctl.7002
Failed 3 to bind listening endpoint: (ADDRESS=(PROTOCOL=tcp)(HOST=gddba-priv))

难道网络配置有问题,检查网络配置
/etc/hosts内容和权限没问题
监听配置没问题
ifconfig时发现有问题节点的私有地址没配置(可能是人为删除)
[root@GDDBA network-scripts]# more ifcfg-eth2
# Broadcom Corporation NetXtreme II BCM5709S Gigabit Ethernet
DEVICE=eth2
BOOTPROTO=dhcp
HWADDR=B8:AC:6F:D2:9B:04
ONBOOT=yes
HOTPLUG=no
DHCP_HOSTNAME=oracledb1

配置私有地址并注释DHCP_HOSTNAME=oracledb1

重启crs,问题解决。