-
编辑 删除
SKGXP_KEEPALIVE_RESET messages in trace files
昨天下午，网络的同事调整接入层设备导致public 网络有短暂的中断，在RAC双节点的trace目录中各产生了如下的信息：/oracle/app/admin/pcendb/bdump/*****_lck0_1634484.trcOracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit ProductionWith the Partitioning, Real Application Clusters, Oracle Label Security, OLAP,Data Mining and Real Application Testing optionsORACLE_HOME = /oracle/app/product/10.2.0System name: AIXNode name: *****Release: 1Version: 6Machine: 00C23A554C00Instance name: *****Redo
-
编辑 删除
ksukia: Starting kill, force = 0 Message in trace files
/oracle/app/admin/pcendb/udump/*****_ora_3055724.trcOracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit ProductionWith the Partitioning, Real Application Clusters, Oracle Label Security, OLAP,Data Mining and Real Application Testing optionsORACLE_HOME = /oracle/app/product/10.2.0System name: AIXNode name: *****Release: 1Version: 6Machine: 00C238C54C00Instance name: *****Redo thread mounted by this instance: 2Oracle process number: 61Unix process pid: 3055724, image: oracle@***** (T
-
编辑 删除
SKGXPLOSTACK message in trace files
因为操作系统的一个bug, 我们对一个两节点的RAC做了一次切换以升级操作系统，导致有些应用在两个节点上都有连接，随后的几天经常在两个节点上的 trace 文件中发现如下的信息： /oracle/app/admin/pcendb/bdump/******_m000_1982704.trcOracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit ProductionWith the Partitioning, Real Application Clusters, Oracle Label Security, OLAP,Data Mining and Real Application Testing optionsORACLE_HOME = /oracle/app/product/10.2.0System name: AIXNode name: *****Release:
-
编辑 删除
WARNING:Could not lower the asynch I/O limit to 160 for SQL direct I/O. It is set to -1
最近一个运行了三年的系统bdump和udump目录突然生成大量类似下面关于异步io的警告。/oracle/app/admin/pcendb/bdump/*****_m001_1065178.trcOracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bit ProductionWith the Partitioning, Real Application Clusters, Oracle Label Security, OLAP,Data Mining and Real Application Testing optionsORACLE_HOME = /oracle/app/product/10.2.0System name: AIXNode name: *****Release: 1Version: 6Machine: 00C23A554C00Instance name: *****Redo thread mounted by this instan
-
编辑 删除
Oracle 中手工创建数据库的语法
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Oracle 8i：(还不支持 9i 的 AUM 特性，需要 DBA 手工管理 Undo 表空间大小和回滚段数量)CREATE DATABASE "test"maxdatafiles 254maxinstances 8maxlogfiles 32character set US7ASCIInational character set UTF8DATAFILE '/opt/oracle/oradata/test/system01.dbf' SIZE 260M AUTOEXTEND ON NEXT 10240Klogfile '/opt/oracle/oradata/test/redo01.log' SIZE 4M, '/opt/oracle/oradata/test/redo02.log' SIZE 4M, '/opt/oracle/oradata/test/redo03.lo
-
编辑 删除
Install Oracle 10gR2 on SuSE Linux 10 SP3 on IBM PPC 64
同事的朋友包了个项目，用了小机但却没买 AIX 操作系统，于是有幸体验了一次 Oracle 在 PPC Linux上的安装。~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1.OS Installation SUSE Linux Enterprise Server 10 SP3 for PPC (kernel：2.6.16.60-0.54.5-ppc64)note:为了安装数据库便利和系统监控，可以把orarun 和 sysstat 包也装上。操作系统安装完成后可以确认以下 RPM 包已经在系统上了，不过版本要比列出的高：glibc:glibc-devel-2.4-31.2 (ppc)glibc-devel-64bit-2.4-31.2 (ppc)Asynchronous IO su
