暂无图片
暂无图片
暂无图片
暂无图片
暂无图片

ocr盘模拟故障、重建恢复

ctrl2hy 2025-11-26
169

[grid@rac01 diag]$ asmcmd lsdg

State    Type    Rebal  Sector  Block       AU  Total_MB  Free_MB  Req_mir_free_MB  Usable_file_MB  Offline_disks  Voting_files  Name

MOUNTED  EXTERN  N         512   4096  1048576     10240     8532                0            8532              0             N  DATA/

MOUNTED  EXTERN  N         512   4096  1048576     20480    15017                0           15017              0             N  FRA/

MOUNTED  NORMAL  N         512   4096  1048576      3072     2146             1024             561              0             Y  OCR/

 

检查ocr备份情况

 

手动备份一个

[root@rac01 rules.d]# ocrconfig -export /u01/app/11204/cdata/rac-cluster/backup_lastest.ocr

[root@rac01 rules.d]# ll /u01/app/11204/cdata/rac-cluster/backup_lastest.ocr

-rw------- 1 root root 122239 Nov 17 07:20 /u01/app/11204/cdata/rac-cluster/backup_lastest.ocr

 

清空ocr磁盘前10M模拟损坏故障

dd if=/dev/zero of=/dev/asm-ocr-01 bs=1M count=10

dd if=/dev/zero of=/dev/asm-ocr-02 bs=1M count=10

 

重启crs 观察日志---坏了两块ocr盘

[cssd(13948)]CRS-1637:Unable to locate configured voting file with ID 8d933740-80eb4ff2-bfe131fa-31544751; details at (:CSSNM00020:) in /u01/app/11204/log/rac01/cssd/ocssd.log

2025-11-17 07:49:17.617:

[cssd(13948)]CRS-1637:Unable to locate configured voting file with ID 446f8215-4e4f4f78-bf3ac759-6b3d5f9f; details at (:CSSNM00020:) in /u01/app/11204/log/rac01/cssd/ocssd.log

2025-11-17 07:49:17.617:

[cssd(13948)]CRS-1705:Found 1 configured voting files but 2 voting files are required, terminating to ensure data integrity; details at (:CSSNM00021:) in /u01/app/11204/log/rac01/cssd/ocssd.log

2025-11-17 07:49:17.617:

[cssd(13948)]CRS-1656:The CSS daemon is terminating due to a fatal error; Details at (:CSSSC00012:) in /u01/app/11204/log/rac01/cssd/ocssd.log

 

 

停掉集群资源:

[root@rac01 rules.d]# crsctl stop crs -f

CRS-2791: Starting shutdown of Oracle High Availability Services-managed resources on 'rac01'

 

以独占模式启动

[root@rac01 rules.d]# crsctl start crs -excl -nocrs

CRS-4123: Oracle High Availability Services has been started.

 

记录集群资源状态

 

此时asmcmd工具并没有显示相关磁盘组信息,是因为asm实例的参数文件存储在ocr磁盘组上的,现在已经一起损毁了,需要重建pfile。

[grid@rac01 ~]$ asmcmd

ASMCMD> lsdg

 

重建asm参数文件

vi /home/grid/asmpfile.ora

+ASM2.asm_diskgroups='DATA','FRA'

+ASM1.asm_diskgroups='OCR','DATA','FRA'

*.asm_diskstring='/dev/*'

*.asm_power_limit=1

*.diagnostic_dest='/u01/app/grid'

*.instance_type='asm'

*.large_pool_size=12M

*.remote_login_passwordfile='EXCLUSIVE'

 

参数文件启动asm实例

startup nomount pfile='/home/grid/asmpfile.ora';

ASM instance started

 

Total System Global Area 1135747072 bytes

Fixed Size                               2260728 bytes

Variable Size                      1108320520 bytes

ASM Cache                           25165824 bytes

此时无法自动mount OCR磁盘组,因为Ocr磁盘组超过2/3的磁盘已损坏,应该彻底将ocr盘移除后新建磁盘组

 

准备好三块磁盘新建ocr磁盘组

set line 200

col path for a50;

select path, header_status, os_mb, free_mb

from v$asm_disk

where header_status != 'member'

or header_status != 'former';

 

select name, state, type, total_mb, free_mb from v$asm_diskgroup;

 

创建磁盘组

create diskgroup OCR normal redundancy

disk '/dev/asm-ocr-new-01','/dev/asm-ocr-new-02','/dev/asm-ocr-new-03';

 

 

安装集群软件时使用了asmca图形化工具创建OCR盘,compatible.asm默认设置就已经为11.2,如果使用命令行创建则默认设置为10.1,所以需要手动修改磁盘组的版本

create spfile='+OCR' from pfile='/home/grid/asmpfile.ora';

 

 

 alter diskgroup OCR set attribute 'COMPATIBLE.ASM'='11.2';

 

启动asm实例,磁盘组正常挂载

SQL> startup pfile='/home/grid/asmpfile.ora';

ASM instance started

 

Total System Global Area 1135747072 bytes

Fixed Size                               2260728 bytes

Variable Size                      1108320520 bytes

ASM Cache                           25165824 bytes

ASM diskgroups mounted

ASM diskgroups volume enabled

 

 

 

 

开始还原ocr文件、恢复voting disk

 

[root@rac01 oracle]# ocrconfig -restore /u01/app/11204/cdata/rac-cluster/backup_lastest.ocr

PROT-40: The specified file '/u01/app/11204/cdata/rac-cluster/backup_lastest.ocr' has an invalid format

 

手动备份的ocr文件有问题,换成自动备份的文件还原

[root@rac01 bin]# ./ocrconfig -restore /u01/app/11204/cdata/rac-cluster/backup00.ocr

 

恢复voting disk

[root@rac01 bin]# crsctl replace votedisk +OCR

Successful addition of voting disk 0352babe760d4fe9bf877ff52a4f4b30.

Successful addition of voting disk 28fff47157624f7bbfb40a51a0181841.

Successful addition of voting disk 02791d7cfce64fadbf600c868227b3a5.

Successfully replaced voting disk group with +OCR.

CRS-4266: Voting file(s) successfully replaced

[root@rac01 bin]# ocrcheck

Status of Oracle Cluster Registry is as follows :

               Version                  :          3

               Total space (kbytes)     :     262120

               Used space (kbytes)      :       3100

               Available space (kbytes) :     259020

               ID                       : 1643724806

               Device/File Name         :       +OCR

                                    Device/File integrity check succeeded

 

                                    Device/File not configured

 

                                    Device/File not configured

 

                                    Device/File not configured

 

                                    Device/File not configured

 

               Cluster registry integrity check succeeded

 

               Logical corruption check succeeded

 

关闭crs,重新正常启动两个节点的crs,观察日志

集群资源正常

 

 

「喜欢这篇文章,您的关注和赞赏是给作者最好的鼓励」
关注作者
【版权声明】本文为墨天轮用户原创内容,转载时必须标注文章的来源(墨天轮),文章链接,文章作者等基本信息,否则作者和墨天轮有权追究责任。如果您发现墨天轮中有涉嫌抄袭或者侵权的内容,欢迎发送邮件至:contact@modb.pro进行举报,并提供相关证据,一经查实,墨天轮将立刻删除相关内容。

评论