ON CALL DBA SUPPORT

— Database blog

Archive for the ‘RAC’ Category

CLUVFY HEALTHCHECK

Posted by ssgottik on 24/07/2015

From oracle 11g R2 we have a new feature in oracle cluvfy , It will generate the HEALTHCHECK report in html format FOR RAC .

$mkdir -p /tmp/clu_tmp

$cluvfy comp healthcheck -collect cluster -html -save -savedir /tmp/clu_tmp

 

Posted in RAC | Leave a Comment »

SRVCTL COMMAND LIST

Posted by ssgottik on 13/11/2013

HELP

$srvctl ACTION PROCESS – h –> -h Stands for help, this command will give you the details about the command which we can use:

example:

$srvctl stop instance -h

$srvctl -V

srvctl version: 11.0.0.0.0

INSTANCE :

$srvctl stop instance -i orcl1 -d orcl -o immediate

$srvctl start instance -i orcl1 -d orcl

$srvctl status instance -i orcl1 -d orcl

Few other options
—————–
-f                  Include disabled applications
-v                Verbose output
-S                 <level> Additional information for EM Console

ASM INSTANCE :

$srvctl status asm -n vmlinux1

$srvctl start asm -n vmlinux1 -i +ASM1 -o open

-o open, mount, or nomount

$srvctl stop asm -n vmlinux1 -i +ASM1 -o normal

-o normal, transactional, immediate, or abort

SERVICE :

srvctl status service -d orcl

srvctl add service -d orcl -s orcldb -r orcl1 -a orcl2

–Relocating the serivce from one node to another ( In this example ORCL2 to ORCL1)

srvctl relocate service -d ngprod -s orcldb -i orcl2 -t orcl1

SCAN LISTENER :

$srvctl config scan

$srvctl status scan_listener

$srvctl status scan_listener -i 1

-i 1 or 2 or 3 as we can have up to 3 scan listeners

$ srvctl relocate scan_LISTENER -i 2 -n orcl1

$srvctl start scan_listener -i 1 -n vmlinux1

$srvctl stop scan_listener -i 1 -n vmlinux1

Posted in RAC | Leave a Comment »

ORA-00443 / ORA-00445 or Process LMD0 died

Posted by ssgottik on 12/11/2013

kjmdmi: pmon timed out in attaching.

When you try to start the one of the instance in RAC and you may encounter below details in the alter log and the instance statup gets terminated :

Thu Oct 10 10:38:29 2011
Process LMD0 died, see its trace file
USER (ospid: 45968): terminating the instance due to error 443
Instance terminated by USER, pid = 45968

If you check the LMDn trace file you will get the actula erro:

$ vi orcl1_lms0_45968.trc

kjmdmi: pmon timed out in attaching.

CAUSE :

There is a problem with the LDAP configuration such that the LDAP names resolution is failing, but there is no other naming resolution method defined in sqlnet.ora.

This was logged as Bug 7599912 INSTANCE STARTUP OPERATION FAILED; however this was closed as Not a Bug. This is a configuration issue.

SOLUTION :

Check whether there is any ldap.ora found in ORACLE_HOME.

$ORACLE_HOME/ldap/admin/ldap.ora
$ORACLE_HOME/network/admin/ldap.ora

If any backup and remove them to reconfigure the LDAP.

or

Add another names resolution method, such as TNSNAMES:

NAMES.DIRECTORY_PATH=(LDAP, TNSNAMES, …)

Reference Doc :

LDAP: ORA-00445: background process “LMD0” did not start after 120 seconds (Doc ID 752758.1)

 

Posted in RAC | Leave a Comment »