I was observing high DFS lock handle wait during Siebel CRM performance testing and i found below SQL very useful to understand the root cause of Watis.
The following environment variables are set as:
ORACLE_OWNER= oracle
ORACLE_HOME= /u01/app/product/grid/11.2.0
Enter the full pathname of the local bin directory: [/usr/local/bin]:
The contents of "dbhome" have not changed. No need to overwrite.
The contents of "oraenv" have not changed. No need to overwrite.
The contents of "coraenv" have not changed. No need to overwrite.
Entries will be added to the /etc/oratab file as needed by
Database Configuration Assistant when a database is created
Finished running generic part of root script.
Now product-specific root actions will be performed.
Using configuration parameter file: /u01/app/product/grid/11.2.0/crs/install/crsconfig_params CRS-4046: Invalid Oracle Clusterware configuration. CRS-4000: Command Create failed, or completed with errors. Failure initializing entries in /var/opt/oracle/scls_scr/host1 /u01/app/product/grid/11.2.0/perl/bin/perl -I/u01/app/product/grid/11.2.0/perl/lib -I/u01/app/product/grid/11.2.0/crs/install /u01/app/product/grid/11.2.0/crs/install/rootcrs.pl execution failed
DBNEWID: Release 11.2.0.2.0 - Production on Sun Mar 20 22:17:55 2011
Keyword Description (Default)
----------------------------------------------------
TARGET Username/Password (NONE)
DBNAME New database name (NONE)
LOGFILE Output Log (NONE)
REVERT Revert failed change NO
SETNAME Set a new database name only NO
APPEND Append to output log NO
HELP Displays these messages NO
To Change Database ID/NAME :
1) Database must be mounted exclusively, if run command in other status you will get the following error
NID-00120: Database should be mounted exclusively
If you running RAC , you need to stop all the nodes and set CLUSTER_DATABASE=FALSE ,before restarting the database instance in mount state.
Oracle 11g has many feature which allow DBA's manage there database more esily and quickly. One of the feature which made DBA's life much easy is build standby database without backup i.e. building from Primary/Active Database.
This feature comes with following benifits.
* No Need to backup
* NO need to sheep backup to Standby server,
* No hessle of Mounting backup to Standby server as nfs or anything like that.
Building standby database is done using RMAN command DUPLICATE which we used before to duplicate database.
Step by Step.
1. Create pfile from spfile on the primay database.
2. scp pfile to new server.
3. edit pfile
4. Create spfile
5. startup nomount;
6. Create listener and Tnsnames.
7. Start standby database using RMAN duplicate database $ rman
Recovery Manager: Release 11.2.0.2.0 - Production on Thu Feb 10 22:18:23 2011
Copyright (c) 1982, 2009, Oracle and/or its affiliates. All rights reserved.
RMAN> CONNECT TARGET SYS/password@DB_DG_SITEB
connected to target database: DBPRD (DBID=39856714)
Deploying Database Vault Administrator to the Database Console OC4J Container
To manually deploy Database Vault Administrator to the Database Console OC4J container:
Stop Oracle Database Console.
UNIX: Go to the $ORACLE_HOME/bin directory and run the following command:
./emctl stop dbconsole
Windows: In the Administrative Services, select the Services utility, and then right-click the OracleDBConsolesid service. Select Stop from the menu.
Create a backup copy and then open the $ORACLE_HOME/oc4j/j2ee/OC4J_DBConsole_service_name/config/server.xml file.
Add the following line before the element:
On Windows systems, replace $ORACLE_HOME with the absolute path to your Oracle Database home.
Create a backup copy and then open the $ORACLE_HOME/oc4j/j2ee/OC4J_DBConsole_service_name/config/http-web-site.xml file.
Add the following line before the element:
Restart Oracle Database Console.
UNIX: Go to the $ORACLE_HOME/bin directory and run the following command:
./emctl start dbconsole
Windows: In the Administrative Services, select the Services utility, and then right-click the OracleDBConsolesid service. Select Start from the menu.
After you complete these steps, you can start Oracle Database Vault Administrator by using the following URL:
https://hostname:port/dva
For example:
https://myserver:1158/dva
If you are unsure of the port number, open the ORACLE_HOME/host_sid/sysman/config/emd.properties file and search for REPOSITORY_URL. In most cases, the host name and port number are the same as Oracle Enterprise Manager Database Control.
Oracle Cluster Verfication utility may fail with followin error on HP-UX .
--------cluvfy log --------------------------
File "/etc/resolv.conf" is consistent across nodes
Time zone consistency check passed
Checking settings of device file "/dev/async" Check for settings of device file "/dev/async" failed. ERROR: PRVF-2415 : The minor number of device file "/dev/async" is incorrect on the nodes: Check failed on nodes:
server3,server1
Starting check for The SSH LoginGraceTime setting ...
Problem: By Default device ID is incorrect for HP-UX 11.31, and need to be created as per oracle minor and major device.
Oracle Database 11g : Database Capture and Database Replay Feature.
A.1 Capture the workload from production database.
BEGIN
DBMS_WORKLOAD_CAPTURE.START_CAPTURE (name => 'TEST_CAPTURE01',
dir => 'DUMPDIR',
duration => 60);
END;
/
B.1 Process the Workload. You must process the captured workload before replay. Its recommended to perform this activity on Test system to avoid any adverse impact on production db performance.
BEGIN
DBMS_WORKLOAD_REPLAY.PROCESS_CAPTURE (capture_dir => 'DUMPDIR');
END;
/
Following steps would be require to replay a captured load on Test System.
1) Initialize the Captured workload you like to replay
Execute DBMS_WORKLOAD_REPLAY.INITIALIZE_REPLAY (replay_name => 'REPLAY001', replay_dir => 'DUMPDIR');
2) Query DBA_WORKLOAD_CONNECTION_MAP and ensure you happy with Captured and Replay mapping. update the Remap Connection if necessary using below procedure.
