rac 环境 阻塞的一点研究.一: 单实例数据库[php]>time1 session 1usera@oradb> select * from v$mystat where rownumusera@oradb> update tab_2 set object_id=1 where object_id=59100;

alert.logErrors in file /app/oracle/admin/racdb/udump/racdb2_ora_479398.trc:ORA-00600: internal error code, arguments: [4080], [1], [131], [], [], [], [], []Mon Jul 28 17:04:04 2008Following on-commit snapshots not refreshed :user_a.tab_mvError 600 trapped in 2PC on transaction 30.14.44852. Cleaning up.Error stack returned to user:ORA-00600: internal error code, arguments: [4080], [1], [131], [], [], [], [], []Mon Jul 28 17:04:04 2008Errors in file /app/oracle/admin/rac

DB: 10.2.0.3RACOS: AIX 5.3[php]SQL> SELECT l.sid,l.type,l.id1,l.id2,l.lmode,l.request,l.block FROM gv$lock l WHERE l.sid=471 AND l.inst_id=2;SID TYPEID1ID2LMODEREQUESTBLOCK---------- ---- ---------- ---------- ---------- ---------- ----------471 TX78647095738602SQL> SELECT w.sid ,w.event , w.p1raw, w.p2,w.p3 FROM gv$session_wait w WHERE w.sid=471 AND w.inst_id=2;SID

os: linux as4 u5 64bitdb: oracle 10.2.0.3 racnode1:SQL> SELECT COUNT(0) FROM test.tab_1;COUNT(0)----------1596576Elapsed: 00:00:02.57SQL>node2 :SQL> SELECT COUNT(0) FROM test.tab_1;同样的sql 就是查询不出结果.这时 查看 v$session_wait : 等待事件:gc cr multi block request如果在node1上 flush buffer_cache , node2 马上就出来结果了.原帖子: http://www.itpub.net/thread-998912-1-1.html

os : redhat linux u5db : oracle 10.2.0.3 rac[php]rac1:SQL> set time on;09:35:58 SQL> SELECT COUNT(0) FROM tab_test;COUNT(0)----------106289609:36:03 SQL>rac2:09:36:52 SQL> select * from v$mystat where rownumSELECT COUNT(0) FROM tab_test;SELECT COUNT(0) FROM tab_test*ERROR at line 1:ORA-01013: user requested cancel of current operationSQL> SELECT n.SID,n.EVENT,n.P1TEXT,n.P1,n.P2TEXT,n.P2,n.WAIT_CLASS,n.WAIT_TIME FROM

我在node1 上利用 lvcreate 创建逻辑卷 .怎么能看 node2 看到呢?看hp-unix 可以用vgexportmyfile datavg 让后在vgimport 到node2 上.那么在linux 上怎么办呢?我参考了yangtingkun 的安装文档上没有具体说明.http://yangtingkun.itpub.net/post/468/213417

今天早上发现数据库已经关闭.startup时发现错误.原来是硬盘有损坏.简单做了一下记录.DB :ORACLE 9204OS :CentOS AS4-- startup 数据库发现如下错误SQL> startupORACLE instance started.Total System Global Area 420549952 bytesFixed Size 451904 bytesVariable Size 201326592 bytesDatabase Buffers 218103808 bytesRedo Buffers 667648 bytesDatabase mounted.ORA-16038: log 2 sequence# 326 cannot be archivedORA-00333: redo log read error block countORA-00312: online log 2 thread 1:

[php]create table TAB_A(IDVARCHAR2(20) not null,R_DATE VARCHAR2(20) not null);alter table TAB_Aadd constraint PK_TAB_A_ID primary key (ID);create index IND_TAB_B_R_DATE on TAB_A (SUBSTR(R_DATE,1,10));create table TAB_B(IDVARCHAR2(30) not null,P_DATE VARCHAR2(20) not null);alter table TAB_Badd constraint PK_TAB_B_ID primary key (ID);create index IND_TAB_B_P_DATE on TAB_B (SUBSTR(P_DATE,1,10));test@test.us.oracle.com

os:LINUX AS4DB :9204Wed Dec 19 09:52:41 2007KCF: write/open error block=0x589 online=1file=3 /opt/oracle/oradata/oadb/oa_data.dbferror=27072 txt: 'Linux Error: 5: Input/output errorAdditional information: 1416'Automatic datafile offline due to write error onfile 3: /opt/oracle/oradata/oadb/oa_data.dbf什么导致datafile 的状态为RECOVER .alter database datafile 3 offline;ALTER DATABASE RECOVERdatafile 3;alter database datafile 3 online;这样data

oralce ： 9201os: win2000 sp4ora-08104 The index is being created or rebuild or waited for recovering from the online (re)build.alter index index_name rebulid online ;kill掉session后。drop index index_name ;还是提示：ora-08104 ；drop table table_name ;-- 索引所在的表还是提示：ora-08104 ；drop table SYS_JOURNAL_[objid];truncate table table_name; -- 要做好备份。drop index index_name ;就是提示：ora-08104 ; （看到ifree的帖子说他truncte后，就可以drop inde