rac 新增节点报错 以加情况介绍

alicejapan

在安装crs软件最后运行rootaddnode.sh脚本的时候报错

/u01/app/crs/bin/srvctl add nodeapps -n rac1 -A rac1-vip/255.255.255.0/eth0 -o /u01/app/crs
CRS-0210: Could not find resource 'ora.rac1.vip'.


无法找到rac1（新增节点）的vip资源

毫无头绪，各位有没有什么入手点？

DB:10.2.0.1
OS：redhat 5.5


具体情况是，一个双节点的rac环境删除rac1这个节点后再添加一个节点也叫rac1，通过OUI 下面的addnode这个脚本执行图形化安装，在最后跑rootaddnode.sh这个脚本的时候报以上的错误，通过crs_stat -t这个命令没有发现vip.rac1的这个组件，其余两个rac1组件正常安装，导出ocr磁盘进行观察发现里面有vip.rac1的信息，怀疑是当时rac1的删除有问题，删的时候vip.rac1还遗留在ocr中，不知道是否能这样理解，谢谢。

[ 本帖最后由 alicejapan 于 2012-6-11 09:38 编辑 ]

Maclean Liu(刘相兵

一个好的问题 的基本要素：一个好的标题、　完整的版本信息、 完整的日志信息、 报错信息

提问至少说明版本！！

确认系统中存在vip 资源 ， crs_stat -t |grep vip  ， 若没有使用vipca添加vip资源

alicejapan

当初发现问题的时候时间比较急，现在加入情况说明，谢谢

Maclean Liu(刘相兵

ODM FINDING:

Hdr: 13843692 10.2.0.5 RDBMS 10.2.0.5 RAC PRODID-5 PORTID-226
Abstract: DELETING THE NODE FAILED  TO REMOVE ALL ENTRIES OF THE DELETED NODE FROM OCR


PROBLEM:
--------
Deleting the node left few entries for vip in the OCR.  As a result, adding
the node with the same vip and hostname failed when rootaddnode.sh is run.
It failed with
CRS-0210: Could not find resource 'ora.xxxxxx.vip'.
where xxxxxx is the node name.

The OCR dump showed incomplete entries for ora.xxxxxx.vip.
Due to this inconsistent, the OCR had to be rebuilt.

Hdr: 13781591 10.2.0.5 PCW 10.2.0.5 PRODID-5 PORTID-226
Abstract: CRS-0210: COULD NOT FIND RESOURCE VIP - ADDING A NEW NODE

*** 02/25/12 02:37 pm ***

PROBLEM:
--------
Three node RAC environment, customer had problem with node-2 few months ago.
They removed/deleted node-2 from the cluster. Bought a new hardware and now
trying to adde node-2 back into the cluster.

There seems to be inconsistencies in the clusterware from the previous
cleanup that prevents from adding the new hardware node-2 back into the
cluster.

DIAGNOSTIC ANALYSIS:
--------------------
# ./rootaddnode.sh
clscfg: EXISTING configuration version 3 detected.
clscfg: version 3 is 10G Release 2.
Attempting to add 1 new nodes to the configuration
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 'root'..
Operation successful.
/ora-ds1/oracrs/app/oracle/product/10.2.0/crs/bin/srvctl add nodeapps -n
mplttap
/ora-ds1/oracrs/app/oracle/pr
oduct/10.2.0/crs
CRS-0210: Could not find resource 'ora.mplttapjs402.vip'.

$ crs_stat -t
Name           Type            Target    State      Host
------------------------------------------------------------

CRSP> srvctl config nodeapps -n mplttapjs401 -a -g -s -l
GSD exists.
ONS daemon exists.
Listener exists.
[oracr] mplttapjs401:/cr-ds1/ora01/app/oracle/oraInventory/logs:
GSD does not exist.
ONS daemon does not exist.
Listener does not exist.
[oracr] mplttapjs401:/cr-ds1/ora01/app/oracle/oraInventory/logs:
GSD exists.
ONS daemon exists.
Listener exists.
[oracr] mplttapjs401:/cr-ds1/ora01/app/oracle/oraInventory/logs:
CRSP>

PRKH-1010 : Unable to communicate with CRS services.
[Communications Error(Native: prsr_initCLSS:[3])]



WORKAROUND:
-----------
No work around currently to add node-2 back in the cluster. Node-1 and Node-3
are up and supporting production environment currently.




确实是一个BUG ， DELETE NODE 在OCR中会残余 vip资源信息，导致rootaddnode.sh 加回这个node时 报 CRS-0210


advice

1. 尝试忽略这个bug

2. 重新添加节点 不要再使用RAC1这个nodename

alicejapan

thx  果然是个BUG