ITPub博客

发现alert log中有创建tablespace的语句,把db中所有表空间都重建一次,当然,由于db中已经存在同名表空间,语句都出错了, 显然不是执行自定义脚本形成的语句Fri Jan 29 18:41:28 2010CREATE TABLESPACE "SYSAUX" BLOCKSIZE 32768 DATAFILE '+DATA1/isddw/datafile/sysaux.1757.705671987' SIZE 5294M AUTOEXTEND ON NEXT 10485760 MAXSIZE 131071M EXTENT MANAGEMENT LOCAL AUTOALLOCATE ONLINE PERMANENT SEGMENT SPACE MANAGEMENT AUTOORA-1543 signalled during: CREATE TABLESPACE "SYSAUX" BLOCKSIZE 32768 DATAFIL

注意,dwpri/dwstb/dwisd/dwied/bmdmk/dwbi/bistb对应的ip要放到hosts中脚本发两个包去探测，基本上可以准确知道机器是否可达.#more ping.sh#!/bin/shcase "$1" indwpri|dwstb|dwisd|dwied|bmdmk|dwbi|bistb)#send 2 packets to pingif [ $1 = "dwpri" -o $1 = "dwstb" ];then nodeNUM=24elif [ $1 = "dwisd" -o $1 = "dwied" -o $1 = "bmdmk" ];then nodeNUM=8elif [ $1 = "dwbi" -o $1 = "bistb" ];then nodeNUM=4fipkgCNT=2while [ $nodeNUM

用于check repair DISKGROUP用[@more@]#!/bin/shexport ORACLE_BASE=/oracle/appexport ORA_ASM_HOME=${ORACLE_BASE}/11gR1/dbexport ORACLE_SID=+ASM24function repair_dg(){sqlstr="alter diskgroup $2 check DISK "x=`expr $1 * 8`i=`expr $x - 8`if [ $1 -eq 9 ];then j=70else j=`expr $x - 1`fiwhile ((i<=j))do if [ $i -eq $j ];then flag=`printf '%04d' $i` sqlstr1=$sqlstr" DATA_"$flag ((i=i+1)) continue fi flag=`printf '%04d' $i` sqlstr=$sqlstr" $2_"$

ronon在执行入库任务时遇到这个事件,它是一个idle event.按通常的理解，idle event一般说明不了问题,但具体问题还要具体分析. 这样可以查到所有空闲事件grep -i "STATS$IDLE_EVENT" /oracle/app/11gR1/db/rdbms/admin/spctab.sql相关链接.[@more@]

DW由于下面kxfplsig_serial错误导致所有节点全部重启Thu Nov 19 17:55:08 2009Errors in file /oracle/app/admin/diag/rdbms/start/start12/trace/start12_lck0_2507.trc:ORA-00600: internal error code, arguments: [kxfplsig_serial], [], [], [], [], [], [], [], [], [], [], []LCK0 (ospid: 2507): terminating the instance due to error 480System state dump is made for local instanceSystem State dumped to trace file /oracle/app/admin/diag/rdbms/start/start12/trace/start12_diag_1944.trcInstance terminated by LCK0,

Bug 5402988: DATAPUMP EXPORT FAILS ORA-31655[@more@]SQL> select count(*) from database_block_corruption_bak2; COUNT(*)---------- 396540create directory dmpdir as '/home/oracle/kyte';expdp "sys/xxx@dwdbetl as sysdba" DUMPFILE=database_block_corruption_bak2.dmp DIRECTORY=dmpdir TABLES=(sys.database_block_corruption_bak2)日志如下:Export: Release 11.1.0.7.0 - 64bit Production on Thursday, 19 November, 2009 15:17:23Copyright (c) 2003, 2007, Oracle. All rights reserved.Connected to:

在遇到下述错时，如果是index,直接drop掉index重建就ok,如果是table一般有两种解决方法 (关于判断块所在对像的类型,见另一个贴子)[@more@]Wed Nov 18 15:16:17 2009Errors in file /oracle/app/admin/diag/rdbms/start/start10/trace/start10_p004_13490.trc (incident=2152710):ORA-01578: ORACLE 鏁版嵁鍧楁崯鍧?(鏂囦欢鍙?5782, 鍧楀彿 85299)ORA-01110: 鏁版嵁鏂囦欢 5782: '+NL02/start/datafile/isdh03m200907090'ORA-26040: 鏁版嵁鍧楁槸浣跨敤 NOLOGGING 閫夐」鍔犺浇鐨?Incident details in: /oracle/app/admin/diag/rdbms/start

在执行hash group by 查询时,经常报not free slot伴随ORA-32695错误, 这是由于11.1.0.7中存在bug6471770 ,是由于大表hash join group by时超出了自动内存管理器分配的内存的限制.[@more@]在执行hash group by 查询时,经常报not free slot伴随ORA-32695错误, 这是由于11.1.0.7中存在bug6471770 https://support.oracle.com/CSP/ui/flash.html#tab=KBHome(page=KBHome&id=()),(page=KBNavigator&id=(bmDocTitle=Bug%206471770%20-%20ora-32690/OERI%20%5B32695%5D%20%5Bhash%20aggregation%20can't%20be%20done%5D%20from%20Hash%20