11gR2解决CRS-4530: Communications failure contacting Cluster Synchronization Services daemon

作者 阿九【转载时请务必以超链接形式标明文章原始出处和作者信息】

RHEL6.3上的Oracle11gR2 版本为：11.2.0.1 在重启服务器后不能启动CSS服务，在以前服务器重启后没有出现这个情况，检查css服务器状态，报错：

CRS-4530: Communications failure contacting Cluster Synchronization Services daemon

根据网友的一篇文章解决了这个问题，文章作者及地址：

文章作者：tomszrp

地址：http://tomszrp.itpub.net/post/11835/493005

[grid@vm11gr2] /home/grid> sqlplus "/as sysasm" SQL*Plus: Release 11.2.0.1.0 Production on Sun Oct 25 10:16:21 2009 Copyright (c) 1982, 2009, Oracle. All rights reserved. Connected to an idle instance. SQL> startup ORA-01078: failure in processing system parameters ORA-29701: unable to connect to Cluster Synchronization Service SQL> 无法连接到CSS服务上.到操作系统上检查一下看看 [grid@vm11gr2] /home/grid> crsctl check css CRS-4530: Communications failure contacting Cluster Synchronization Services daemon [grid@vm11gr2] /home/grid> [grid@vm11gr2] /home/grid> ps -ef|grep cssd 果然没有CSS的服务daemon进程,再看一下HAS(High Availability Service)的状态 [grid@vm11gr2] /home/grid> crsctl check has CRS-4638: Oracle High Availability Services is online [grid@vm11gr2] /home/grid> ps -ef|grep d.bin grid 5886 1 0 10:06 ? 00:00:01 /u01/app/grid/product/11.2/grid/bin/ohasd.bin reboot [grid@vm11gr2] /home/grid> 发现HAS的服务确实启动了的,而ora.cssd和ora.diskmon这2个服务是依赖于HAS维护的. 进一步查看各资源的状态 [grid@vm11gr2] /home/grid> crs_stat -t Name Type Target State Host -------------------------------------------------------------- ora.FLASH_DATA.dg ora.diskgroup.type OFFLINE OFFLINE vm11gr2 ora.SYS_DATA.dg ora.diskgroup.type OFFLINE OFFLINE vm11gr2 ora.asm ora.asm.type OFFLINE OFFLINE vm11gr2 ora.cssd ora.cssd.type OFFLINE OFFLINE vm11gr2 ora.diskmon ora.diskmon.type OFFLINE OFFLINE vm11gr2 [grid@vm11gr2] /home/grid> [grid@vm11gr2] /home/grid> crsctl status resource -t -------------------------------------------------------------------------------- NAME TARGET STATE SERVER STATE_ DETAILS -------------------------------------------------------------------------------- Local Resources -------------------------------------------------------------------------------- ora.FLASH_DATA.dg OFFLINE OFFLINE vm11gr2 ora.SYS_DATA.dg OFFLINE OFFLINE vm11gr2 ora.asm OFFLINE OFFLINE vm11gr2 -------------------------------------------------------------------------------- Cluster Resources -------------------------------------------------------------------------------- ora.cssd 1 OFFLINE OFFLINE ora.diskmon 1 OFFLINE OFFLINE 再看一下ora.cssd和ora.diskmon的属性 [grid@vm11gr2] /home/grid> crs_stat -p ora.cssd NAME=ora.cssd TYPE=ora.cssd.type ACTION_SCRIPT= ACTIVE_PLACEMENT=0 AUTO_START=never CHECK_INTERVAL=30 DESCRIPTION="Resource type for CSSD" FAILOVER_DELAY=0 FAILURE_INTERVAL=3 FAILURE_THRESHOLD=5 HOSTING_MEMBERS= PLACEMENT=balanced RESTART_ATTEMPTS=5 SCRIPT_TIMEOUT=600 START_TIMEOUT=600 STOP_TIMEOUT=900 UPTIME_THRESHOLD=1m [grid@vm11gr2] /home/grid> crs_stat -p ora.diskmon NAME=ora.diskmon TYPE=ora.diskmon.type ACTION_SCRIPT= ACTIVE_PLACEMENT=0 AUTO_START=never CHECK_INTERVAL=20 DESCRIPTION="Resource type for Diskmon" FAILOVER_DELAY=0 FAILURE_INTERVAL=3 FAILURE_THRESHOLD=5 HOSTING_MEMBERS= PLACEMENT=balanced RESTART_ATTEMPTS=10 SCRIPT_TIMEOUT=60 START_TIMEOUT=60 STOP_TIMEOUT=60 UPTIME_THRESHOLD=5s [grid@vm11gr2] /home/grid> 到这里基本就找到了原因了,可以看到这两个资源的AUTO_START属性默认都设置为never,也就是说他们不会随着HAS服务的启动而自动启动的, 尽管默认情况下HAS服务是开机自动启动的.好了，那我们就手动启动一下吧: [grid@vm11gr2] /home/grid> crsctl start resource ora.cssd CRS-2672: Attempting to start 'ora.cssd' on 'vm11gr2' CRS-2679: Attempting to clean 'ora.diskmon' on 'vm11gr2' CRS-2681: Clean of 'ora.diskmon' on 'vm11gr2' succeeded CRS-2672: Attempting to start 'ora.diskmon' on 'vm11gr2' CRS-2676: Start of 'ora.diskmon' on 'vm11gr2' succeeded CRS-2676: Start of 'ora.cssd' on 'vm11gr2' succeeded [grid@vm11gr2] /home/grid> 注:ora.cssd和ora.diskmon这两个服务是有依赖关系的，启动哪个都会把两个都起来. [grid@vm11gr2] /home/grid> crs_stat -t Name Type Target State Host -------------------------------------------------------------- ora.FLASH_DATA.dg ora.diskgroup.type OFFLINE OFFLINE vm11gr2 ora.SYS_DATA.dg ora.diskgroup.type OFFLINE OFFLINE vm11gr2 ora.asm ora.asm.type OFFLINE OFFLINE vm11gr2 ora.cssd ora.cssd.type ONLINE ONLINE vm11gr2 ora.diskmon ora.diskmon.type ONLINE ONLINE vm11gr2 [grid@vm11gr2] /home/grid> CSS服务起来了，重启动asm instance [grid@vm11gr2] /home/grid> sqlplus "/as sysasm" SQL*Plus: Release 11.2.0.1.0 Production on Sun Oct 25 10:30:03 2009 Copyright (c) 1982, 2009, Oracle. All rights reserved. Connected to an idle instance. SQL> startup ASM instance started Total System Global Area 284565504 bytes Fixed Size 1336036 bytes Variable Size 258063644 bytes ASM Cache 25165824 bytes ASM diskgroups mounted SQL> exit Disconnected from Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production With the Automatic Storage Management option [grid@vm11gr2] /home/grid> crs_stat -t Name Type Target State Host -------------------------------------------------------------- ora.FLASH_DATA.dg ora.diskgroup.type ONLINE ONLINE vm11gr2 ora.SYS_DATA.dg ora.diskgroup.type ONLINE ONLINE vm11gr2 ora.asm ora.asm.type ONLINE ONLINE vm11gr2 ora.cssd ora.cssd.type ONLINE ONLINE vm11gr2 ora.diskmon ora.diskmon.type ONLINE ONLINE vm11gr2 [grid@vm11gr2] /home/grid> tips 1)默认情况下HAS(High Availability Service)是自动启动的.通过如下命令可以取消和启用自动启动 crsctl disable has crsctl enable has 2)HAS手动启动和停止 crsctl start has crsctl stop has 3)查看HAS的状态 crsctl check has 4)如果想让ora.css和ora.diskmon服务随着HAS的启动而自动启动，那么你可以这两个服务的AUTO_START属性 crsctl modify resource "ora.cssd" -attr "AUTO_START=1" or crsctl modify resource "ora.diskmon" -attr "AUTO_START=1" 5)如果想取消ora.css和ora.diskmon的Auto start crsctl modify resource "ora.cssd" -attr "AUTO_START=never" crsctl modify resource "ora.diskmon" -attr "AUTO_START=never"

但是我发现在RHEL6.3上启动has服务并不能只使用crsctl start has 这个命令来启动，还必须以root用户去执行

/etc/init.d/init.ohasd run 这命令，才能启动has服务，不然crsctl start has 会一直挂着且也不能启动。

[grid@rh6-ora11g ~]$ crsctl stop has CRS-2791: Starting shutdown of Oracle High Availability Services-managed resources on 'rh6-ora11g' CRS-2673: Attempting to stop 'ora.ora11g.db' on 'rh6-ora11g' CRS-2673: Attempting to stop 'ora.LISTENER.lsnr' on 'rh6-ora11g' CRS-2677: Stop of 'ora.LISTENER.lsnr' on 'rh6-ora11g' succeeded CRS-2677: Stop of 'ora.ora11g.db' on 'rh6-ora11g' succeeded CRS-2673: Attempting to stop 'ora.DATA.dg' on 'rh6-ora11g' CRS-2677: Stop of 'ora.DATA.dg' on 'rh6-ora11g' succeeded CRS-2673: Attempting to stop 'ora.asm' on 'rh6-ora11g' CRS-2677: Stop of 'ora.asm' on 'rh6-ora11g' succeeded CRS-2673: Attempting to stop 'ora.cssd' on 'rh6-ora11g' CRS-2677: Stop of 'ora.cssd' on 'rh6-ora11g' succeeded CRS-2673: Attempting to stop 'ora.diskmon' on 'rh6-ora11g' CRS-2677: Stop of 'ora.diskmon' on 'rh6-ora11g' succeeded CRS-2793: Shutdown of Oracle High Availability Services-managed resources on 'rh6-ora11g' has completed CRS-4133: Oracle High Availability Services 已停止。 [grid@rh6-ora11g ~]$  [grid@rh6-ora11g ~]$ crsctl start has

[root@rh6-ora11g ~]# /etc/init.d/init.ohasd run mkfifo: 无法创建先进先出文件"/var/tmp/.oracle/npohasd": 文件已存在 ^C [root@rh6-ora11g ~]#

[grid@rh6-ora11g ~]$ crsctl start has CRS-4123: Oracle High Availability Services 已启动。 [grid@rh6-ora11g ~]$ crs_stat -t 名称           类型           目标      状态      主机         ------------------------------------------------------------ ora.DATA.dg    ora....up.type ONLINE    ONLINE    rh6-ora11g   ora....ER.lsnr ora....er.type ONLINE    ONLINE    rh6-ora11g   ora.asm        ora.asm.type   ONLINE    ONLINE    rh6-ora11g   ora.cssd       ora.cssd.type  ONLINE    ONLINE    rh6-ora11g   ora.diskmon    ora....on.type ONLINE    ONLINE    rh6-ora11g   ora.ora11g.db  ora....se.type ONLINE    ONLINE    rh6-ora11g   [grid@rh6-ora11g ~]$

这个问题其实在装软件的时候就碰到了，在RHEL6.3上装11gR2 在执行root.sh脚本的时候会报错：

Adding daemon to inittab CRS-4124: Oracle High Availability Servicesstartup failed. CRS-4000: Command Start failed, orcompleted with errors. ohasd failed to start: 对设备不适当的 ioctl操作 ohasdfailed to start: 对设备不适当的 ioctl 操作 at /grid/base/grid/crs/install/roothas.pl line296.

在RHEL6.3上解决这个错误的方法是：

官网的解决办法（亲测~没有用，但还是写出来）

1、/usr/local/bin目录下没有perl这个文件

解决办法：

ln -s/usr/bin/perl /usr/local/bin/perl

 

2、存在 /tmp/.oracle or /var/tmp/.oracle or/usr/tmp/.oracle.这几个文件

解决办法：

rm -rf /tmp/.oracle/*/usr/tmp/.oracle/* /var/tmp/.oracle/*

非官网的解决办法（解决办法来自于博客“语虎店”的文章 ，连接：http://yuhushop.com/?p=453）

在执行root.sh的时候，会生成/var/tmp/.oracle/npohasd文件，解决的办法就是在执行root.sh的时候，当生成了 /var/tmp/.oracle/npohasd 后，立即执行 ：

dd if=/var/tmp/.oracle/npohasd of=/dev/null bs=1024 count=1

 然后再执行root.sh命令。重新执行root.sh命令之前要先执行一个deconfig操作

在GRID_HOME/crs/install目录下

./roothas.pl-deconfig -force -verbose