olsnodes command in oracle rac
Olsnodes,oifcfg,cluvfy,
Introduction
olsnodes command provides the list of nodes and other information of all the nodes particapating in the cluster.
crsctl tool
will verify health check of cluster services
from one node, can verify of all nodes.
#cd /u01/app/11.2.0/grid/bin
#./crsctl check cluster -all
crs
css
evm - upper stack
ohasd - lower stack -
Only specific node - service status of that particular node.
./crsctl check cluster
Note: On server boot, the cluster services get started on its own.
----------------------------------------------------
To stop the clusterware services
#./crsctl stop cluster -all ( includes all the nodes)
But make sure to stop lower stack manually of every node.
./crsctl stop crs (execute on every node)
-------------------------------------------
To start .. the service if manually then
just start crs
both lower and upper get started.
./crsctl start crs
Monitor ./crsctl check cluster -all
------------------------------------------
Verify Unix logs and clusterware logs.
For every service, we have a log in grid home.
Where does the clusterware logs locate?
cd $GRID_HOME/log path
eg: /u01/app/11.2.0/grid/log/rac1
For node2
/u01/app/11.2.0/grid/log/rac2
Will find
folders
crsd
cssd
evmd
How to read the log file ?
# tail -100f crsd.log
also
# tail -1000f crsd.log | more
Using diagcollection script
diagcollection.pl
Steps: #
# ./diagcollection.pl --collect
once the zip files generated , will use winscp/ftp to pull in to desktop and then upload to metalink/support.
If oracle support needs , to collect diagnostic information from cluster logs to identify the issue. Will need to provide logs using diagcollection script
Diagnostic files - Retention policy - housekeeping
----------------------------------
$GRID_HOME/log/rac1
crsd.log
cssd.log
ohasd.log
read by tail command
Retention Policy for these logs.
# Rotation / retention Policy
10x10 rules as part of automatic rotation/retention policy
and governed automatically.
10 copies of cssd.log files with 50M retained and rotated
subsequently
ohasd,evmd,crsd etc logs also retain 10 copies with 10m size.
The policy doesnt applies to the alerthostname.log file.
If the file got removed , need to stop,start service.
Changing parameter file :
--------------------------
Use sid to apply on nodes in a cluster.
sql>alter system set undo_retention=2600 scope=both sid='*';
------------------------------------
About -
Olsnodes,oifcfg,cluvfy
olsnodes
On a cluster - to list active nodes
# cd /u01/app/11.2.0/grid/bin/
#./olsnodes
# ./olsnodes -i
# ./olsnodes -n
rac1 1
rac2 2
rac3 3
# ./olsnodes -s
rac1 Active
rac2 Active
rac3 Inactive
---------------------------------------------------------
Oracle Interface Configuration Tool (OIFCFG)
to adminstrate the network interfaces.
# oifcfg iflist (For listing interfaces)
# ./oifcfg iflist
eth0 192.169.2.0
eth1 10.10.10.0
# ./oifcfg getif ( verify - list of public/private interfaces)
eth0 192.169.2.0 global public
eth1 10.10.10.0 global cluster_interconnect
#./oifcfg setif xxxxxx private
Verify - scan configuration
---------------------------
su - oracle
. grid_env
$ srvctl config scan
Cluster Verification Utility – Environment Integrity Check tool
--------------------------------------------------------------
Before installation , how to ensure the cluster integrity ...
how we verify cluster configuration
using cluvfy tool....
comes with in grid software as well. in installed binaries.
two Stages
pre
post - failed - no DNS Server - cluvfy failed.
cd /u01/app/11.2.0/grid/bin
ls –ltr cluvfy*
cluvfy
to be ran with oracle user *
Verifying shared storage accessibility
$cluvfy comp ssa -n all -verbose
Cluster Verification Utility Connectivity Verifications
$cluvfy comp nodereach -n rac1,rac2,rac3 -verbose
$cluvfy comp nodecon -n all -verbose
Post - stage
cluvfy stage -post crsinst -n rac1,rac2 -verbose
Cluster integrity
./cluvfy comp clu
Crsctl to manage grid services
srvctl to manage/monitor resources - db,listener,scan,vip.
Check the Status of the RAC
---------------------------
$ srvctl config database -d prod
$ srvctl status database -d prod
SQL> SELECT inst_id,inst_name FROM v$active_instances;
Verify instances status using views in grid.
--------------------------------------------
SQL> select inst_id,instance_name,status,thread# from gv$instance;
$srvctl status instance -d prod -i prod1,prod2
$srvctl status instance -d prod -n rac1
Note : Info on olsnodes it may be differ from your environment prod,dev,test,naming conventions and directories etc
THANKS FOR VIEWING MY BLOG FOR MORE UPDATES FOLLOW ME
