2008-07-25 23:10
总是这么凑巧，前段日子看到beta版，想什么时候会发布呢？
今早一看，以于2008-7-24发布了！ |
2008-07-24 09:32
最常用的命令为：
crs_stat -t
crsctl start/stop/check crs
ps -ef|grep d.bin
srvctl start/stop nodeapps -n nodename
crs_relocate ora.nodename.vip (VIP挂在其它节点，死活不回来的时候用)
对于由于private网络故障等引起的节点cssd进程僵死，而又不想重起server，采用如下方法：
As root:
# crsctl stop crs;//也可以用/etc/init.crs stop （可能会stop不了，那就只有kill了）
-- make sure all crs process are down. Check with command: |
2008-07-24 09:21
Oracle Clusterware 在10204patch后，不再需要CRS 10203 bundle#2 |
2008-07-24 09:18
#-------------------------------------------------------------------------
# Interim Patch for Base Bugs: 5613796
#-------------------------------------------------------------------------
#
# DATE: Thu Jan 18 11:55:01 2007
# -------------------------------
# Platform Patch for : AIX Based Systems 64-bit for 5L
# Product Version # : 9.2.0.8
# Product Patched : RDBMS
#
# Bugs Fixed by this patch:
# ---------------- |
2008-07-17 15:47
For exp: 1)direct path 2)conventional path -- set parameter buffer; export only needed data with parameter query; seperate tables to parallel export operation if you are about to export some tables
For imp: set the parameter buffer properly; commit=y/n depends on different situations; use indexfile parameter to spool the index creation script and do not import the indexes at first(with index=n); do a backup and disable archivelog. almost there...
|
2008-07-16 20:05
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03009: failure of backup command on ORA_DISK_1 channel at 07/16/2008 17:16:16
ORA-19566: exceeded limit of 0 corrupt blocks for file /dev/rxxx
$ dbv file=/dev/rxxx BLOCKSIZE=8192
DBVERIFY: Release 9.2.0.8.0 - Production on Wed Jul 16 17:42: |
2008-07-15 13:29
you can downgrade db using startup migrate since 10g.
you can upgrade db using startup upgrade since 10g.
you can upgrade/downgrade db using startup migrate till 9i.
在9i，无论升级/降级 数据库都是startup migrate
10g后增加了upgrade参数，升级可直接用startup upgrade，降级仍是startup migrate.
http://forums.oracle.com/forums/thread.jspa?threadID=433852 |
2008-07-15 12:59
2008-07-14 16:23
这个模型很简单，primary/standby 均安装10g 二进制文件，然后在primary开始migrate，standby以10g的instance启动，继续apply redo files，完成！ |
2008-07-14 16:10
9i迁移到10g的数据库，如果表空间为空的话，构成表空间的数据文件将不会被改变！
migrate到底会做些什么呢？除了数据字典、数据库组件。 |
|
| 
