Attempting to start CRS stack Failure at scls_scr_create with code 1

ALLSTARS_ORACLE

Attempting to start CRS stack  Failure at scls_scr_create with code 1

Attempting to start CRS stack
Failure at scls_scr_create with code 1
Internal Error Information:
  Category: 1234
  Operation: scls_scr_create
  Location: mkdir
  Other: Unable to make user dir
  Dep: 2

问题分析
localconfig命令是用于单实例的RDBMS来配置cssd进程，来支持ASM的。
在RAC系统中，cssd是CRS的其中一个部分，在RAC数据库中，CRS已经包含和支持ASM功能了。
绝对没有必要在RAC系统中运行localconfig，并且永远不应该在RAC系统中使用，因为这将毁坏CRS的配置。
以上异常是因为在RAC环境运行localconfig所导致

建议和意见
根据How to Restore CRS after accidentally run localconfig on RAC system [ID 747415.1]，其中介绍了
localconfig reset和localconfig delete两种情况下如何进行恢复。
在这种情况下，尝试运行localconfig reset之后，然后参考localconfig reset这种情况的恢复
On the node(s) where "localconfig reset" has been run:
1. /etc/oracle/ocr.loc is like:
ocrconfig_loc=<$ORACLE_HOME>/cdata/localhost/local.ocr
local_only=TRUE

2. in /etc/inittab, following lines exist:
h1:35:respawn:/etc/init.d/init.evmd run >/dev/null 2>&1 </dev/null
h3:35:respawn:/etc/init.d/init.crsd run >/dev/null 2>&1 </dev/null
h1:35:respawn:/etc/init.d/init.cssd run >/dev/null 2>&1 </dev/null <<< this one

CRS can not start on the affected node. Reboot will not help.

Caution! proceed with followings only when /etc/oracle/ocr.loc has not been manually modified to point back to real OCR. Eg, its content should still be:

ocrconfig_loc=<$ORACLE_HOME>/cdata/localhost/local.ocr
local_only=TRUE

If you have modified ocr.loc to point to real OCR device, please restore the content back to above before continue with the followings. Otherwise the following commands will zero out the real OCR device.


as root user:
1. run $ORA_CRS_HOME/install/rootdelete.sh, it will remove the init* scripts and place back the blank inittab
$ORA_CRS_HOME/install/rootdelete.sh
2. run  $ORA_CRS_HOME/install/rootdeinstall.sh, it will blank out the $ORACLE_HOME/cdata/localhost/local.ocr and remove the ocr.loc
$ORA_CRS_HOME/install/rootdeinstall.sh
3. run $ORA_CRS_HOME/root.sh, CRS should start automatically after this.
$ORA_CRS_HOME/root.sh



在执行$ORA_CRS_HOME/root.sh命令后，发现操作挂起在以下操作上
Startup will be queued to init within 30 seconds.
Adding daemons to inittab
WARNING: directory '/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' 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>
…
Creating OCR keys for user 'root', privgrp 'system'..
Operation successful.
…
Format of 3 voting devices complete.
Startup will be queued to init within 30 seconds.
Adding daemons to inittab
操作一直挂起在上述操作

检查后发现oracle用户不能进入OCR和Voting Disk的盘所在的目录
修改权限，oracle用户能够进入OCR和Voting Disk的盘所在的目录
此后oot.sh成功执行完毕

但是root.sh脚本执行完毕后，crs一些进程不能正常启动，其中crsd.log产生大量的类似以下的报警信息

  clssgsGroupJoin: CSS has not reached fatal mode.Registration is not yet safe. Retrying
这是因为在RAC环境中，第一个节点运行root.sh后，在剩余节点上没有运行root.sh所导致的。
因为此时2号节点已经被操作系统回收，无法在第二个节点上运行root.sh

lunar

这个很好，提请初玩RAC的童鞋们注意，做事情要按照readme和官方文档，不能想当然和没章法