after OS reboot ,11g RAC asm not automatic startup ORA-27102
[ohasd(4507)]CRS-2807:Resource ‘ora.asm’ failed to start automatically.
SQL> startup nomount
ORA-27102: out of memory
OEL5.8 Install RAC Grid 11203 PRVF-5184 : Check of following Udev attributes failed
PRVF-5184 : Check of following Udev attributes of “rac2:/dev/asm-diskc” failed: “[Owner: Found=’root’ Expected=’grid’, Group: Found=’root’ Expected=’asmadmin’, Permissions: Found=’0600′ Expected=’0660′]”
11gr2 install database加大AMM内存时INS-35171
Oracle Error :: INS-35171
Target database memory (stringMB) exceeds at least one of the selected nodes available shared memory (string MB).
rac 11203 grid 安装PRVF-5636 : The DNS response time for an unreachable node exceeded “15000” ms
OUI默认是带域名的名称如znode1.anbob.com/znode1-vip.anbob.com,去掉改为了znode1/znode1-vip 新增加的znode2/znode2-vip，再下一步到检查的时，检查通过。
Never bind peeking or Wrong execution plan?
注意当用veriable变量绑定时sqlplus autotrace有可能显示的执行计划不是实际的，用dbms_xplan.display_cursor查看
RHEL/OEL 5启用xmanager 远程桌面x
For XDMCP connection to Red Hat 5 EL or CentOS 5 or OEL5
1. Change runlevel to 5
Open /etc/inittab and set the initial runlevel to 5 as following:
id:5:initdefault:
11gr2安装Grid 时 node2 root.sh失败Start of resource “ora.asm -init” failed
Start of resource “ora.asm -init” failed
Failed to start ASM
Failed to start Oracle Clusterware stack
11gr2安装Grid 时 ins-40719 ins-40912
ins-40719 ipaddress configured for single client access name(scan):
rac-scan.anbob.com is already assigned to another system
ORA-00600: internal error code, arguments: [Skgmfail] 案例
数据库在运行一段时间后crash,遭遇ORA-00600: internal error code, arguments: [SKGMFAIL], [2], [4], [4], [1]，重启后又可以正常运行。
Why not use index oracle? 为什么不用索引？ 检查列表
sys_op_descend – Returns value as it would be used in a descending index. This is essentially
reverse() function with output in hex.
RAC那些事，概念
rac 全名Real Application Clusters，是ORACLE DATABASE 上的一个组件， 用共享存储的结构可以把1个数据库运行在多个分别计算的节点instance中 ，来提高了数据库的可用性，可以安装在企业版或标准版中，rac 出生于2001年的9I release 1, 以它的前身是oracle parallel server,rac 引入cache fusion机制提高实例间的资源同步。
Utilities-rlwarp, how to clear history command? do it
RLWRAP_HOME:
directory in which the history and completion files are kept.
RedHat LINUX el 安装UEK内核并安装ORACLE Validated RPM
Oracle为Linux推出了坚不可摧的企业级内核Unbreakable Enterprise Kernel， UEK是一个高速的、新式的、可靠的Linux内核，它们自己的Exadata Database Machine就是基于该内核，也是为oracle自己的软、硬个件优化过的内核
BITMAP CONVERSION FROM/TO ROWIDS
sql就是根据条件一走btree索引把rowid转换为BITMAP,条件二也是走另一个btree索引把rowid转换为另一个BITMAP,然后两个BITMAP 再做and 合并，合并后的结果再转换为ROWID，表再根据rowid取会行记录过滤没有在索引上的其它条件
oracle Memory Management 改进
Automatic Memory Management (AMM) 在ORACLE 11g中被引入，这是从OS级别内存管理的重大改变，回顾ORACLE的版本更新历程，内存管理也在逐步的更新
ORA-02049 ‘timeout: distributed transaction waiting for lock’ 解决
ORA-02049是一个分布式事务等待超时的异常，当一个session 持有一个行级锁，另一个分布式事务比如通过dblink也想修改相同行的数据，就会产生 tx队列等待，当等待时间超过了系统参数distributed_lock_timeout的值时，就产生了这个错误，任何通过通过DBLINK的操作都是分布式事物。
install grid control 10201 on linux ,oc4j CA OUI-25031 error
vi 打开，增加-Djava.net.preferIPv4Stack=true，后如下
$self->logAndExec(“$self->{‘OH’}/jdk/bin/java -Xmx512M -DemLocOverride=$self->
{‘OH’} -Djava.net.preferIPv4Stack=true
10g stop/disable isqlplus/EM
平时装完数据库都习惯停掉EM,ISQLPLUS,两个服务是独立的，利用J2EEE为用户提供的WEB操作或监控的应用,10G EM可能还不太完美，这之后的版本应该是最好监控ORACLE的工具软件了，如GRID,12cloud
Control the order in which the tables accessed join using sql hint leading and ordered
The LEADING hint causes Oracle to use the specified table as the first table in the join order.
If you specify two or more LEADING hints on different tables, then all of them are ignored. If you specify the ORDERED hint, then it overrides all LEADING hints.
