《课程笔记 | Oracle DBA武林秘籍》2

《课程笔记 | Oracle DBA武林秘籍》2-Oracle RAC OCR磁盘组故障导致宕机

一.前言

记得有次扩容asm磁盘组的时候，意外格式化了FRA磁盘组对应的磁盘，导致集群异常。 原因是当时依照盘符来操作了，没有看uuid，两边盘符一致，但是对应的盘不是同一个，导致出现了问题。所以说，无论是绑定udev还是扩容等操作，都需要来依靠uuid或者其他唯一编码来确认。

二.笔记

学习专栏课程《Oracle DBA武林秘籍》，如何恢复OCR磁盘，这关键时刻绝对顶。
当OCR磁盘被格式化或者异常导致集群宕机，怎么恢复OCR磁盘。
课程链接：https://www.modb.pro/course/article/116

三.步骤

0.背景

CRS磁盘组被破坏，数据丢失。集群异常。
ocr自动备份成功。

ocrcheck
ocrconfig -showbackup
crsctl query css votedisk
select name,state from v$asm_diskgroup;
select a.group_number, a.name, b.path from v$asm_diskgroup a, v$asm_disk b where a.group_number = b.group_number order by 1, 3;

1.强制停止所有节点CRS

./crsctl stop crs -f

2.某个节点以独占模式启动CRS

-excl：独占模式
-nocrs：不检查crs及votedisk
./crsctl start crs -excl -nocrs

3.进入ASM实例，创建新的磁盘组CRS

[oracle@host01 ~]$ export ORACLE_BASE=/u01/app/grid
[oracle@host01 ~]$ export ORACLE_HOME=/u01/app/11.2.0/grid
[oracle@host01 ~]$ export PATH=$ORACLE_HOME/bin:$PATH
[oracle@host01 ~]$ export ORACLE_SID=+ASM1
[oracle@host01 ~]$ sqlplus / as sysasm
SQL> select name,state from v$asm_diskgroup;
NAME                           STATE
------------------------------ -----------
FRA                            DISMOUNTED
DATA                           DISMOUNTED

[oracle@host01 ~]$ sqlplus / as sysasm
SQL> create diskgroup CRS normal redundancy disk 'ORCL:ASMDISK13','ORCL:ASMDISK14','ORCL:ASMDISK15' attribute 'COMPATIBLE.ASM' = '11.2.0';
Diskgroup created.

SQL> select name,state from v$asm_diskgroup;
NAME                           STATE
------------------------------ -----------
FRA                            DISMOUNTED
DATA                           DISMOUNTED
CRS                            MOUNTED

4.恢复OCR

[root@host02 bin]# ./ocrconfig -showbackup
PROT-26: Oracle Cluster Registry backup locations were retrieved from a local copy

host01     2017/11/05 21:42:56     /u01/app/11.2.0/grid/cdata/cluster/backup00.ocr
host01     2017/11/05 21:42:56     /u01/app/11.2.0/grid/cdata/cluster/day.ocr
host01     2017/11/05 21:42:56     /u01/app/11.2.0/grid/cdata/cluster/week.ocr
PROT-25: Manual backups for the Oracle Cluster Registry are not available
You have new mail in /var/spool/mail/root

[root@host01 ~]# cd /u01/app/11.2.0/grid/bin/
[root@host01 bin]# ./ocrconfig -restore /u01/app/11.2.0/grid/cdata/cluster/backup00.ocr

[root@host01 bin]# ./ocrcheck
Status of Oracle Cluster Registry is as follows :
         Version                  :          3
         Total space (kbytes)     :     262120
         Used space (kbytes)      :       3036
         Available space (kbytes) :     259084
         ID                       : 1940676675
         Device/File Name         :       +CRS
                                    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

5.恢复votedisk

[root@host01 bin]# ./crsctl query css votedisk
Located 0 voting disk(s).

[root@host01 bin]# ./crsctl replace votedisk +CRS
Successful addition of voting disk 35d4a3b952f84fe6bf0a1260c5f147ed.
Successful addition of voting disk 21fc8a3162ba4fddbf7478ca6c9e8a32.
Successful addition of voting disk 3abb687e95d04f32bf732f359fca48c1.
Successfully replaced voting disk group with +CRS.
CRS-4266: Voting file(s) successfully replaced

[root@host01 bin]# ./crsctl query css votedisk
##STATE    File Universal Id                File Name Disk group
--  -----    -----------------                --------- ---------
 1. ONLINE   35d4a3b952f84fe6bf0a1260c5f147ed (ORCL:ASMDISK13) [CRS]
 2. ONLINE   21fc8a3162ba4fddbf7478ca6c9e8a32 (ORCL:ASMDISK14) [CRS]
 3. ONLINE   3abb687e95d04f32bf732f359fca48c1 (ORCL:ASMDISK15) [CRS]
Located 3 voting disk(s).

6.所有节点重启CRS

./crsctl stop crs -f
./crsctl start crs

7.所有节点重启Oracle实例