作者: Maclean Liu, post on January 30th, 2012 虽然我并不推荐你使用ASMLIB绑定设备名， 详见这篇文章《Why ASMLIB and why not?》 。 但大概因为介绍ASMLIB的文章远多于UDEV的缘故， 导致有大量对RAC安装配置不太熟悉的朋友仍执意采用ASMLIB，又因为ASMLIB的配置不算太简单所以在实际安装RAC之前的ASMLIB实施过程中有不少人遇到了问题， 其次在ASMLIB的使用过程中麻烦也不少。
这里总结了一下ASMLIB的诊断思路， 如下脚本：
cat /etc/sysconfig/oracleasm 1) uname -a 2) rpm -qa | grep ^oracleasm 3) rpm -V oracleasmlib 4) multipath -ll 1) output of command line # rpm -V oracleasm-support # /etc/init.d/oracleasm scandisks # /etc/init.d/oracleasm listdisks # ls -l -R /dev/oracleasm/ # ls -l /etc/sysconfig/oracleasm . . . → Read More: 如何诊断ASMLIB故障
作者: Maclean Liu, post on December 8th, 2011 X$KFFXP是ASM(Automatic Storage Management)自动存储管理特性的重要内部视图，该视图反应了File Extent Map映射关系，ASM会将文件split成多个多个piece分片，这些分片被称为Extents。 在Disk上存放这些Extent的位置，就是我们常说的”Allocation Unit”。
KFF意为Kernel File，X$KFFXP即Kernel File Extent Maps， 该内部视图的一条记录代表一个Extent。
其字段含义如下：
GROUP_KFFXP diskgroup number (1 – 63) ASM disk group number. Join with v$asm_disk and v$asm_diskgroup NUMBER_KFFXP file number for the extent ASM file number. Join with v$asm_file and v$asm_alias COMPOUND_KFFXP (group_kffxp << 24) + file # File identifier. . . . → Read More: Oracle内部视图X$KFFXP
作者: Maclean Liu, post on September 17th, 2011 什么是” ASM file metadata operation “等待事件?
这是一个内部的undocumented等待事件：
“ASM file metadata operation” event is associated with database (instance) communication with ASM (instance). If the database has files in ASM disk group(s) it needs to access disk group(s), get extent maps for files that it already has, get updated extent info (e.g. after a rebalance), create new . . . → Read More: ASM file metadata operation等待事件
作者: Maclean Liu, post on August 4th, 2011 客户的一套11.2.0.1 RAC系统采用ASM diskgroup 存放ocr和votedisk，该REG diskgroup中的某个LUN disk由于硬件的原因损坏了，导致冗余的votedisk表决磁盘有一个处于OFFLINE状态，客户希望能删除该OFFLINE的votedisk并新增一个可用的。
在删除该votedisk文件时出现了CRS-4258的错误，错误如下:
crsctl query css votedisk ## STATE File Universal Id File Name Disk group — —– —————– ——— ——— 1. OFFLINE 5b3380d6367e4f94bf19e9db5f2f684e () [] 2. ONLINE 6802e6d139354fb3bf95725dd01a02fd (/dev/ocr2) [REG] 3. ONLINE a433d51ebd2d4facbfc8e95b017f5393 (/dev/asm-disk1) [REG] 4. ONLINE 3784d344bffa4f6ebff21c4dd3c873bd (/dev/asm-disk2) [REG] Located 4 voting disk(s). crsctl delete css votedisk 5b3380d6367e4f94bf19e9db5f2f684e CRS-4258: . . . → Read More: CRS-4258: Addition and deletion of voting files are not allowed because the voting files are on ASM
作者: Maclean Liu, post on July 10th, 2011 在<Why ASMLIB and why not?>我们介绍了使用ASMLIB作为一种专门为Oracle Automatic Storage Management特性设计的内核支持库(kernel support library)的优缺点，同时建议使用成熟的UDEV方案来替代ASMLIB。
这里我们就给出配置UDEV的具体步骤，还是比较简单的:
1.确认在所有RAC节点上已经安装了必要的UDEV包 [root@rh2 ~]# rpm -qa|grep udev udev-095-14.21.el5 2.通过scsi_id获取设备的块设备的唯一标识名，假设系统上已有LUN sdc-sdp for i in c d e f g h i j k l m n o p ; do echo “sd$i” “`scsi_id -g -u -s /block/sd$i` “; done sdc 1IET_00010001 sdd 1IET_00010002 sde 1IET_00010003 sdf . . . → Read More: 利用UDEV服务解决RAC ASM存储设备名
作者: Maclean Liu, post on June 28th, 2011 今天在给ASM扩磁盘组的时候发现11.2.0.2上asmcmd中lsdg命令所显示的磁盘组数不正确，现象如下:
Node A lsdg正常: ASMCMD> lsdg State Type Rebal Sector Block AU Total_MB Free_MB Req_mir_free_MB Usable_file_MB Offline_disks Voting_files Name MOUNTED EXTERN N 512 4096 1048576 142976 50596 0 50596 0 Y DATA/ MOUNTED EXTERN N 512 4096 1048576 51200 50742 0 50742 0 N FRA/ Node B lsdg仅显示DATA磁盘组 ASMCMD> lsdg State Type Rebal Sector Block . . . → Read More: 11.2.0.2 asmcmd lsdg show incorrect diskgroup number
作者: Maclean Liu, post on June 11th, 2011 kfk: async disk IO等待事件是ASM下异步的System I/O等待事件，kfk内核层面在disk_asynch_io=true时被激活。当rbal或其他ASM相关后台进程在维护ASM磁盘组时可能进入kfk: async disk IO等待。
SQL> col name for a20 SQL> col PARAMETER1 for a10 SQL> col PARAMETER2 for a10 SQL> col PARAMETER3 for a10 SQL> col WAIT_CLASS for a15 SQL> select name,parameter1,parameter2,parameter3,wait_class from v$event_name where name=’kfk: async disk IO’; NAME PARAMETER1 PARAMETER2 PARAMETER3 WAIT_CLASS ——————– ———- ———- ———- ————— kfk: async . . . → Read More: Oracle等待事件kfk:async disk IO
作者: Maclean Liu, post on May 30th, 2011 收集ASM诊断信息最佳的工具仍是RDA，对于不能使用RDA的环境可以采用如下脚本:
spool asm_diag1.txt set pagesize 1000 set lines 500 col “Group Name” form a25 col “Disk Name” form a30 col “State” form a15 col “Type” form a7 col “Free GB” form 9,999 alter session set nls_date_format=’DD-MON-YYYY HH24:MI:SS’; select sysdate “Date and Time” from dual; select * from v$asm_diskgroup order by 1; select * from . . . → Read More: Script:收集ASM诊断信息
|
|
最新评论