How to configure YUM on Oracle Linux

How to configure YUM on Oracle Enterprise Linux?

This is an example for Enterprise Linux 6:

go to the directory /etc/yum.repos.d and edit the file public-yum-el5.repo

# cd /etc/yum.repos.d
# wget http://public-yum.oracle.com/public-yum-el5.repo

Open the yum configuration file in a text editor:
vi public-yum-el5.repo
Change enabled=0 to enabled=1 from the repository you need.

If you have other Linux installtions, then go to http://public-yum.oracle.com/ for instructions on how to configure the yum repository for other Linux versions.

Now we can try it!
yum list
yum install firefox

ORA-04021: timeout occurred while waiting to lock object error while executing catbundle.sql script

ORA-04021: timeout occurred while waiting to lock object error while executing the catbundle.sql script

During the patching process, we need to apply the patch sqls to the database. To apply the patches to the database, we execute

cd $ORACLE_HOME/rdbms/admin
sqlplus / as sysdba
@catbundle.sql apply PSU

the above script hangs due to locked object and the error is written to the log ORA-04021: timeout occurred while waiting to lock object

This was due to the locked object dbms_advisor (in my case).

Solution:
Restarting the database cleared the locks and the catbundle apply completed in few minutes.

Things to consider before upgrading to 11.2.0.4 from 11.2.0.3 - Part 1

Things to consider before upgrading to 11.2.0.4 from 11.2.0.3 

Oracle recommends out-of place upgrade from 11.2.0.3 to 11.2.0.4

At the planning stage of the database and grid Infrastructure upgrade, Download and run the ORAchk utility from Oracle to verify and fix any issues before upgrading the clusterware, ASM and the database.

ORAchk Health Checks For The Oracle Stack

ORAchk tool is the advanced version of RACcheck tool and Oracle recommends to download the ORAchk tool to verify All the Oracle stack.

To download and more details about the script is in Oracle support

ORAchk - Health Checks for the Oracle Stack (Doc ID 1268927.2)

$ ./orachk -u -o pre
Enter upgrade target version (valid versions are 11.2.0.3.0, 11.2.0.4.0, 12.1.0.1.0, 12.1.0.2.0):- 11.2.0.4.0

CRS stack is running and CRS_HOME is not set. Do you want to set CRS_HOME to /u01/app/grid?[y/n][y]y

Checking ssh user equivalency settings on all nodes in cluster

Node dev02 is configured for ssh user equivalency for oracle user


Searching for running databases . . . . .

. .
List of running databases registered in OCR
1. rac
2. None of above

Select databases from list for checking best practices. For multiple databases, select 1 for All or comma separated number like 1,2 etc [1-2][1].1
. .


Checking Status of Oracle Software Stack - Clusterware, ASM, RDBMS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
-------------------------------------------------------------------------------------------------------
                                                 Oracle Stack Status
-------------------------------------------------------------------------------------------------------
Host Name  CRS Installed  RDBMS Installed  CRS UP    ASM UP    RDBMS UP  DB Instance Name
-------------------------------------------------------------------------------------------------------
dev01    Yes             Yes             Yes        Yes      Yes      rac01
dev02    Yes             Yes             Yes        Yes      Yes      rac02
-------------------------------------------------------------------------------------------------------


---------------------------------------------------------------------------------------------------------------------------------
                                       Installed components summary
---------------------------------------------------------------------------------------------------------------------------------
GI_HOME                                  ORACLE_HOME                                                  Database Names
---------------------------------------------------------------------------------------------------------------------------------
/u01/app/grid - 11.2.0.3.0               /u01/app/oracle/product/11.2.0/db - 11.2.0.3.0               rac
---------------------------------------------------------------------------------------------------------------------------------


Copying plug-ins

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .


Checking Grid Infrastructure and RDBMS upgrade readiness to version 112040



Collections and audit checks log file is
/oracle/software/11204_DB/orachk_dev01_rac_110514_175251/log/orachk.log



Checking for prompts in /oracle/.bash_profile on dev01 for oracle user...




Checking for prompts in /oracle/.bash_profile on dev02 for oracle user...



Starting to run orachk in background on dev02

=============================================================
                    Node name - dev01
=============================================================

Collecting - ASM Diskgroup Attributes
Collecting - ASM initialization parameters
Collecting - Database Parameters for rac database
Collecting - Database Undocumented Parameters for rac database
Collecting - Pre-upgrade File Locations for rac database
Collecting - /proc/cmdline
Collecting - /proc/modules
Collecting - CPU Information
Collecting - DiskFree Information
Collecting - DiskMount Information
Collecting - Kernel parameters
Collecting - Maximum number of semaphore sets on system
Collecting - Maximum number of semaphores on system
Collecting - Maximum number of semaphores per semaphore set
Collecting - Memory Information
Collecting - NUMA Configuration
Collecting - Network Interface Configuration
Collecting - Network Performance
Collecting - Network Service Switch
Collecting - OS Packages
Collecting - OS version
Collecting - Patches for Grid Infrastructure
Collecting - Patches for RDBMS Home
Collecting - number of semaphore operations per semop system call


Data collections completed. Checking best practices on dev01.
--------------------------------------------------------------------------------------


 INFO =>    Real World Network Bandwidth Guidelines
 INFO =>    Important Automatic Storage Management (ASM) Notes and Technical White Papers
 INFO =>    Unattended Installation and Configuration Options
 INFO =>    Be aware of provided optional integrity checks
 INFO =>    Patching Notes by Release
 INFO =>    Prior to Upgrade Verify NO Materialized Views Being Refreshed (Automatically or Manually) for rac
 WARNING => ORACLE_HOME environment variable is set
 FAIL =>    Opatch version is lower than recommended in RDBMS_HOME for /u01/app/oracle/product/11.2.0/db
 FAIL =>    Opatch version is lower than recommended in GRID_HOME
 INFO =>    Special Considerations for SCAN and HAIP of 11.2.0.2 and above  Grid Infrastructure
 INFO =>    Minimizing Planned Downtime for Upgrades and Patching
 INFO =>    Upgrade Related References
 INFO =>    Lifetime Support Policy and Product Certification INFO
 INFO =>    Capture performance baseline, backup important configuration files and batch, cron, DBMS_JOBS and DBMS_SCHEDULER jobs
 INFO =>    Known Issues Integrating pre-11gR2 Databases with  Grid Infrastructure
 INFO =>    Be Aware of New Instant Client for Simplified Deployment
 INFO =>    Oracle E-Business Suite interoperability and migration resources
 INFO =>    Oracle Software Download Sites
 WARNING => One or More Object Names in ALL_OBJECTS table are Reserved Words for rac
 WARNING => One or More Column Names in ALL_TAB_COLUMNS table are Reserved Words for rac
 WARNING => Some Users Needing Network ACLs for Oracle Utility Packages Found for rac
 INFO =>    Steps to execute rootupgrade.sh
 INFO =>    Save "Oracle Enterprise Manager Database Control" files and data with the emdwgrd Utility before upgrading  database.
 INFO =>    Information about ASM process parameter when its not set to default value
 INFO =>    Information about Grid Infrastructure software installation directory
 WARNING => OS parameter vm.swappiness is NOT set to the recommended value



Copying results from dev02 and generating report. This might take a while. Be patient.

=============================================================
                    Node name - dev02
=============================================================

Collecting - /proc/cmdline
Collecting - /proc/modules
Collecting - CPU Information
Collecting - DiskFree Information
Collecting - DiskMount Information
Collecting - Kernel parameters
Collecting - Maximum number of semaphore sets on system
Collecting - Maximum number of semaphores on system
Collecting - Maximum number of semaphores per semaphore set
Collecting - Memory Information
Collecting - NUMA Configuration
Collecting - Network Interface Configuration
Collecting - Network Performance
Collecting - Network Service Switch
Collecting - OS Packages
Collecting - OS version
Collecting - Patches for Grid Infrastructure
Collecting - Patches for RDBMS Home
Collecting - number of semaphore operations per semop system call


Data collections completed. Checking best practices on dev02.
--------------------------------------------------------------------------------------



 WARNING => ORACLE_HOME environment variable is set
 FAIL =>    Opatch version is lower than recommended in RDBMS_HOME for /u01/app/oracle/product/11.2.0/db
 FAIL =>    Opatch version is lower than recommended in GRID_HOME
 INFO =>    Information about ASM process parameter when its not set to default value



---------------------------------------------------------------------------------
                      CLUSTERWIDE CHECKS
---------------------------------------------------------------------------------
---------------------------------------------------------------------------------

Detailed report (html) - /oracle/software/orachk_dev01_rac_101014_175251/orachk_dev01_rac_101014_175251.html


UPLOAD(if required) - /oracle/software/orachk_dev01_rac_101014_175251.zip


$

ORAchk generated a detailed report in html which provides the fixes for the failure and warning.

The sample HTML report is as below for the opatch version fix

Oracle RAC Upgrade Readiness Report

System Health Score is 92 out of 100 (detail)

CRS Opatch version

Recommendation
 Most recent opatch version is 11.2.0.3.6 (or higher). Download latest opatch from following MOS note and install.

Note :- report data may be empty if its role separated environment where oracle does not have permission to call opatch in GRID_HOME. Please run $CRS_HOME/OPatch/opatch version manually and verify that version is equal or higher than 11.2.03.6

Links
Note: 1373255.1 - 11.2.0.1/11.2.0.2 to 11.2.0.3 Database Upgrade on Exadata Database Machine
Note: 274526.1 - How To Download And Install The Latest OPatch Version
Needs attention on dev03, dev04
Passed on -

Status on dev03:
FAIL => Opatch version is lower than recommended in GRID_HOME


DATA FROM dev03 - CRS OPATCH VERSION



OPatch Version: 11.2.0.3.4

OPatch succeeded.
Status on dev04:
FAIL => Opatch version is lower than recommended in GRID_HOME


DATA FROM dev04 - CRS OPATCH VERSION



OPatch Version: 11.2.0.3.4


OPatch succeeded.


Great ORAchk tool from Oracle.

Part 2 coming soon...

ORA_01017: Invalid username/password error while creating a standby database

Issue: ORA_01017: Invalid username/password error while creating a standby database

$ rman TARGET sys/change_on_install@TEST

Recovery Manager: Release 11.2.0.4.0 - Production on Fri Oct 24 15:24:44 2014

Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.

RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-00554: initialization of internal recovery manager package failed
RMAN-04005: error from target database:
ORA-01017: invalid username/password; logon denied

Though I changed the passwords and recreated the password file in primary and copied back to standby server, still i received the invalid username and password on the standy

Solution:


SQL> show parameter REMOTE_LOGIN_PASSWORD

NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
remote_login_passwordfile            string      NONE

SQL> show parameter REMOTE_OS_AUTHENT

NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
remote_os_authent                    boolean     FALSE
SQL>


The parameter remote_login_password file should be set to exclusive. Changed the value to EXCLUSIVE in primary and restarted the database

SQL> alter system set remote_login_passwordfile='EXCLUSIVE' scope=spfile;

System altered.


SQL> shutdown immediate;
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> startup
ORACLE instance started.

Total System Global Area 2405122048 bytes
Fixed Size                  2255592 bytes
Variable Size             503317784 bytes
Database Buffers         1895825408 bytes
Redo Buffers                3723264 bytes
Database mounted.
Database opened.
SQL>

Now the rman command duplicate target database for standby works fine.

Errors: ORA-00301, ORA-17502, ORA-15005 name is already used by an existing alias

Errors: ORA-00301, ORA-17502, ORA-15005

SQL> ALTER DATABASE ADD LOGFILE THREAD 2
 GROUP 5 ('+TEST_REDO/TEST_redo_t2g1m1',
  2    3           '+TEST_REDO/TEST_redo_t2g1m2') size 20M,
  4   GROUP 6 ('+TEST_REDO/TEST_redo_t2g2m1',
  5           '+TEST_REDO/TEST_redo_t2g2m2') size 20M,
  6   GROUP 7 ('+TEST_REDO/TEST_redo_t2g3m1',
  7           '+TEST_REDO/TEST_redo_t2g3m2') size 20M,
  8   GROUP 8 ('+TEST_REDO/TEST_redo_t2g4m1',
  9           '+TEST_REDO/TEST_redo_t2g4m2') size 20M;
ALTER DATABASE ADD LOGFILE THREAD 2
*
ERROR at line 1:
ORA-00301: error in adding log file '+TEST_REDO/TEST_redo_t2g1m1' - file cannot be created
ORA-17502: ksfdcre:4 Failed to create file +TEST_REDO/TEST_redo_t2g1m1
ORA-15005: name "TEST_redo_t2g1m1" is already used by an existing alias

The problem is because the logfile exists in the ASM disks as below

servTEST501:grid $ asmcmd
ASMCMD> cd TEST_REDO
ls
ASMCMD> TEST/
TEST_def_temp_01
TEST_redo_t1g1m1
TEST_redo_t1g1m2
TEST_redo_t1g2m1
TEST_redo_t1g2m2
TEST_redo_t1g3m1
TEST_redo_t1g3m2
TEST_redo_t1g3m3
TEST_redo_t1g3m4
TEST_redo_t1g4m1
TEST_redo_t1g4m2
TEST_redo_t2g1m1
TEST_redo_t2g1m2
TEST_redo_t2g2m1
TEST_redo_t2g2m2
TEST_redo_t2g3m1
TEST_redo_t2g3m2
TEST_redo_t2g4m1
TEST_redo_t2g4m2
TEST_temp_01
ASMCMD> rm TEST_redo_t2g1m1
ASMCMD> rm TEST_redo_t2g1m2
;
:
ASMCMD>rm TEST_redo_t2g4m2

remove the logfiles created for thread 2 in the asm and recreate the logfiles

SQL> ALTER DATABASE ADD LOGFILE THREAD 2
 GROUP 5 ('+TEST_REDO/TEST_redo_t2g1m1',
         '+TEST_REDO/TEST_redo_t2g1m2') size 20M,
  2    3    4   GROUP 6 ('+TEST_REDO/TEST_redo_t2g2m1',
  5           '+TEST_REDO/TEST_redo_t2g2m2') size 20M,
  6   GROUP 7 ('+TEST_REDO/TEST_redo_t2g3m1',
  7           '+TEST_REDO/TEST_redo_t2g3m2') size 20M,
  8   GROUP 8 ('+TEST_REDO/TEST_redo_t2g4m1',
  9           '+TEST_REDO/TEST_redo_t2g4m2') size 20M;

Database altered.

Explanation: Though we drop the logs using the sql command it is registered as dropped in the controlfile but the physical file exists which has to be removed manually.

How to change the hostname on Oracle Enterprise Linux 6

How to change the hostname on Oracle Enterprise Linux 6?

If you want to change the hostname of the server, then edit the below two files

/etc/sysconfig/network
/etc/hosts

Lets take the example that we want to change the hostname "myserver001" to "hostserver001"
#vi /etc/sysconfig/network

HOSTNAME=myserver001
NETWORKING=yes
GATEWAY=172.21.110.54
NETWORKING_IPV6=no
IPV6INIT=no
NOZEROCONF=yes

After change

HOSTNAME=hostserver001
NETWORKING=yes
GATEWAY=172.21.110.54
NETWORKING_IPV6=no
IPV6INIT=no
NOZEROCONF=yes

Then update the /etc/hosts file
172.21.110.160          myserver001

After change


172.21.110.160           hostserver001


Now reboot the server..Your new name is assigned to the host.

How to check the status of Oracle VM manager through perl / shell script?

How to check the status of Oracle VM manager through perl / shell script?

Scripting helps to automate a task. If you are looking for a script to check the status of the OVM then the below perl script can be used to check the status of the Oracle VM manager.

Save the file as check_vm_status.pl

#!/usr/bin/perl -w
my $status = `service ovmm status`;
chomp($status);
my $statusok = "Oracle VM Manager is running...";
if ( $status ne $statusok ) {
        print "oracle vm is not running";
}
else
{
        print "$status";
}

To run the script,

$ perl check_vm_status.pl

The output of the above script is as below, If the OVM is running then,

Oracle VM Manager is running...

If the OVM is not running then

oracle vm is not running

The same code can be used as Korn shell script as below

Save the file as check_vm_status.ksh

#!/usr/bin/ksh -w
$status = `service ovmm status`;

$statusok = "Oracle VM Manager is running...";
if ( $status ne $statusok ) {
        print "oracle vm is not running";
}
else
{
        print "oracle vm is running";
}

To run the script,

$ perl check_vm_status.ksh

Hope this helps!

Using dd to copy multiple files or disks to a remote system

Using dd to copy multiple files / raw disks to a remote system

Let us say we have raw disks at server A and we want to copy the disks to server B

here are the steps

login to the server A and go to the directory

serverA001 # cd /dev/disks/
serverA001 # ls | while read i
> do
> dd if=$i ibs=4096 conv=notrunc,noerror | (ssh 172.02.129.198 dd of=/dev/disks/$i obs=4096)
> echo $i
> done


You need to enter the user password of the remote host to copy the files.

Oracle Application Server Status

How to check Oracle Application Server Status?

On Windows services, check the status of OracleASControl and OracleProcessManager.

And for Oracle Process Monitor(OPMN) on command prompt execute opmnctl status command as shown below

ORACLE_HOME\opmn\bin>opmnctl status

Processes in Instance: OAS.XYZ.com
-------------------+--------------------+---------+---------
ias-component | process-type | pid | status
-------------------+--------------------+---------+---------
LogLoader | logloaderd | N/A | Down
dcm-daemon | dcm-daemon | 7688 | Alive
HTTP_Server | HTTP_Server | 22196 | Alive
WebCache | WebCache | 2332 | Alive
WebCache | WebCacheAdmin | 2340 | Alive
OC4J | home | 2372 | Alive
OC4J | OC4J_BI_Forms | 2364 | Alive
OC4J | OC4J_BI_Forms | 2380 | Alive
OC4J | OC4J_BI_Forms | 2396 | Alive
OC4J | OC4J_BI_Forms | 2388 | Alive
DSA | DSA | N/A | Down



The dcm, HTTP Server, WebCache, OC4J status should be alive.

11g DIAGNOSTIC_DEST Initialization parameter

Checking the background_dump_dest for alert log files and user_dump_dest for trace files and errors are going to be a past.

I was excited to know about the new oracle11g initialization parameter DIAGNOSTIC_DEST which replaced BACKGROUND_DUMP_DEST, USER_DUMP_DEST and CORE_DUMP_DEST parameters.

All the trace informations are stored in a single location specified using DIAGNOSTIC_DEST. This helps the Oracle 11g database advisors to have a single point of contact for all the information required to auto tune the SQL or database.

If we do not specify a location for this initialization parameter then it will default to the $ORACLE_BASE/diag

Error 45 initializing SQL*PLUS Internal Error

During pre-11g check the sql script execution fails with the "Error 45 Initializing SQL*PLUS" on the sqlplus as below

SQL> spool utlu1112i.log

SQL> @utlu112i.sql

Error 45 initializing SQL*Plus

Internal error

The reason is the path where the 11g script utlu112i.sql script copied to on the 10g database server.

Solution:

Copy the utlu112i.sql script to the $ORACLE_HOME/rdbms/admin directory and execute the script.

SQL> spool utlu1112i.log

SQL> @utlu112i.sql

Enterprise Manager 11g fails database instance unavailable

In Oracle 11g RAC the enterprise manager dbconsole doesn't start and gives an OC4J not available error.
If we try to deconfig using emca as below we receive an error database instance unavailable. If you face the same issue then here is one of the possible solution

export ORACLE_SID to the instance id not db id for example if your 2 node rac db is orcl then instance will be orcl1 & orcl2.

Set the ORACLE_SID to orcl2 and export the oracle sid

Go to oracle_home/bin directory and check the emca and emctl files and update if required for the ORACLE_HOME & ORACLE_SID.

And then execute the below command to remove the EM repository


[oracle@home bin]$ emca -deconfig dbcontrol db -repos drop

STARTED EMCA at May 2, 2012 11:17:49 AM
EM Configuration Assistant, Version 11.2.0.0.2 Production
Copyright (c) 2003, 2005, Oracle.  All rights reserved.

Enter the following information:
Database SID: orcl1
Listener port number: 1521
Password for SYS user:
Password for SYSMAN user:

Do you wish to continue? [yes(Y)/no(N)]: Y
May 2, 2012 11:18:03 AM oracle.sysman.emcp.EMConfig perform
INFO: This operation is being logged at /opt/oracle/cfgtoollogs/emca/racdb/emca_2012_05_02_11_17_49.log.
May 2, 2012 11:18:03 AM oracle.sysman.emcp.EMDBPreConfig performDeconfiguration
WARNING: EM is not configured for this database. No EM-specific actions can be performed.
May 2, 2012 11:18:06 AM oracle.sysman.emcp.EMReposConfig invoke
INFO: Dropping the EM repository (this may take a while) ...

And after dropping recreate the repository.

How to write a blob pdf data from the Oracle database to a pdf file

How to write a blob pdf data from the Oracle database to a pdf file

create directory orcl_dir as '/mnt/oracle/orcl_dir'

--In my example I am passing two variables
CREATE OR REPLACE PROCEDURE BLOBToFILESample (typeidin in number,varnamein IN VARCHAR2) IS
type_id number(11);
var_name varchar2(40);
v_blob         BLOB;
  blob_length    INTEGER;
  out_file       UTL_FILE.FILE_TYPE;
  v_buffer       RAW(32767);
  chunk_size     BINARY_INTEGER := 32767;
  blob_position  INTEGER := 1;

BEGIN

  -- Retrieve the BLOB for reading
  SELECT myblobcolumn INTO v_blob FROM mytablename WHERE mycol1=typeidin and mycol2 = varnamein;

  -- Retrieve the SIZE of the BLOB
  blob_length:=DBMS_LOB.GETLENGTH(v_blob);

  --ORCL_DIR is the directory which i have created on my database.
  out_file := UTL_FILE.FOPEN ('ORCL_DIR', varnamein, 'wb', chunk_size);

  -- Write the BLOB to file in chunks
  WHILE blob_position <= blob_length LOOP
    IF blob_position + chunk_size - 1 > blob_length THEN
      chunk_size := blob_length - blob_position + 1;
    END IF;
    DBMS_LOB.READ(v_blob, chunk_size, blob_position, v_buffer);
    UTL_FILE.PUT_RAW(out_file, v_buffer, TRUE);
    blob_position := blob_position + chunk_size;
  END LOOP;

  -- Close the file handle
  UTL_FILE.FCLOSE (out_file);
END;
/

The above code will save the file in the directory orcl_dir.

Some of the error I have encountered is when passing the parameters to the procedure do not give the same name as the column name. If you specify then you may encounter ora-01422 as below

ORA-01422: exact fetch returns more than requested number of rows

Voting Disk and OCR backups in Oracle 11g R2

In Oracle 11g R2, the OCR and voting disks are backed up automatically into a single file. To check the backup of the OCR and voting disks

#  $GRID_HOME/bin/ocrconfig -showbackup

Where GRID_HOME is your Grid installation home directory. In my example is


#  /u01/app/11.2.0/grid/bin/ocrconfig -showbackup

How to move ocr and voting disks to a new directory in OCFS

How to move ocr and voting disks to a new directory in OCFS?

Moving OCR (Oracle Cluster Registry)

1.       Login as root user and check for a recent backup of the ocr file.

            #ocrconfig –showbackup

2.       If the backup does not exists then take a backup

             #ocrconfig –export –s online

3.       Check whether the crs and its processes are running

             #crsctl check crs

4.       Check the location of the crs files

             #ocrcheck

5.       To move the ocr file to a new location - /u01/oradata/racdb

             #ocrconfig –replace ocr /u01/oradata/racdb/OCRFile

6.       Check whether the crs is pointed to new location

             #ocrcheck

             #cat /etc/oracle/ocr.loc

             #Device/file /u02/oradata/racdb/OCRFile getting replaced by device /u01/oradata/racdb

             ocrconfig_loc=/u01/oradata/racdb/OCRFile

             ocrmirrorconfig_loc=/u02/oradata/racdb/OCRFile_mirror

7.       Check the privilege and permissions of OCR file. It should be owned by root and group dba/oinstall. If not change as below

            #chown root:oinstall /u01/oradata/racdb/OCRFile

Moving voting disks to a new location

1.       Login as root

2.       Stop the clusterware

#crsctl stop crs

3.       Check the voting disks

#crsctl query crs votedisk

4.       Take a backup of the voting disk

#dd if=voting_disk_name of=backup_file_name   

          

5.       Add the voting disk to a new location

#crsctl add css votedisk /u01/oradata/racdb/CSSFile –force

6.       Delete the old voting disk

#crsctl delete css votedisk /u02/oradata/racdb/CSSFile –force

7.       Check the voting disk file permission

The css file should be owned by oracle with oinstall / dba group depends on your primary group

8.       Start the cluster

#crsctl start crs

9.       Check the voting disk new location

#crsctl query crs votedisk

How to Stop a cluster in Oracle RAC

How to Stop a cluster in Oracle RAC

To stop a cluster in Oracle RAC, As a root user, execute the command

# /bin/crsctl stop crs

If your crs home is /opt/oracle/crs then

#/opt/oracle/crs/bin/crsctl stop crs

How to find the large log files older than one year to delete on unix

How to find the large log files older than one year to delete on unix?

To find the one year old large log files, execute the below command

find . -name '*.log' -size +200k -mtime +360 -print

where find is the command
. is for the current directory ( if you want to check on /var/log then specify the directory here)
'*.log' is to check for the log files
-size  is to check for the size of the log files. I gave 200k
-mtime is to check for the time +360 for one year old files
-print is to show the files on the screen.

Flashback Recovery Area Full - Unable to connect RMAN

Yesterday, There was an issue with the Flashback Recovery Area. The FRA disk is full and when we try to connect as "sys as sysdba" or "rman target", the database hangs.


If you are looking for a solution to similar scenario then read on...

Check the disk space for total, available and used space using df command in linux.

move to the archivelog folder in the flash recovery area for the database and check for the old logs.

If you have set the retention window to 7 days for example or have a backup taken 2-3 days before then you can remove the old logs to reclaim the disk space.

Now try to connect using the RMAN

rman target nocatalog

execute the crosscheck command to verify the backups and delete expired backups and logs.

Note, By removing the logs at OS level will not update the data dictionary hence the crosscheck command needs to be executed to cross verify the logs and files.

How to create flashback restore point on database

How to create flashback restore point on database?

If the database is up, shutdown the database and restart the database in mount mode.

Check whether the flashback is on, If the flashback is not on then enable the flashback. Note: If you are using RAC then before enabling flashback make sure all other instances are shutdown

If you executed the flashback on command without shutting down the other instances, you will receive the below error

ERROR at line 1:
ORA-38777: database must not be started in any other instance.

Solution is to make sure all other instances are down.

After enabling the flashback, check the status to make sure the flashback is enabled on the database.

And check the DB_FLASHBACK_RETENTION_TARGET. By default it is 1440 which is one day. If you want for “n” no of days then the formula is 1440*n=value. For example if you want to retain for 3 days then 1440*3=4320

Alter the database to set the retention to 3 days.

ALTER SYSTEM SET DB_FLASHBACK_RETENTION_TARGET=4320;

And then create the restore point as below

CREATE RESTORE POINT before_my_change;

PXE-E32: TFTP Open Timeout - VMWARE - REDHAT LINUX Installtion

PXE-E32: TFTP Open Timeout

When Installing Redhat Linux on the newly created vmware vm machine, I received the error TFTP open timeout error.

Casuse:  TFTP open request was not acknowledged.

Solution:

Verify that the TFTP service is running. If you are not using the TFTP then simply wait for few minutes, the boot prompt will come back automatically after sometime with operating system not found message. Now select the operating system iso image.