Paweł Domański
6.7 KiB
Steps followed to shutdown:
-
Shutdown Oracle Home process accessing database.
> emctl stop dbconsole > srvctl stop listener -n node1 -
Shutdown RAC Database Instances on all nodes.
> srvctl status database -d oradb > srvctl stop database -d oradb -
Shutdown All ASM instances from all nodes.
> srvctl status asm > srvctl stop asm -n node1 -f > srvctl stop asm -n node2 -f -
Shutdown Node applications running on nodes.
> srvctl status nodeapps -n node1 > srvctl stop nodeapps -n node1 -f -
Shut down the Oracle Cluster ware or CRS. ( Run as root user)
> crsctl check cluster -all > crsctl stop crs
Steps followed to startup:
-
Start Oracle Clusterware or CRS. (Run as root user)
> crsctl start crs > crsctl check cluster -all ( DBA can see "CRS-4639: Could not contact Oracle High Availability Services" or "CRS-4535: Cannot communicate with Cluster Ready Services" messages. Wait 5 minutes and then again check with "crsctl check cluster -all" command. This time Database administrator will get "CRS-4537: Cluster Ready Services is online". If still same issue DBA can start ora.crsd process to resolve this issue. Below is the command > crsctl start res ora.crsd -init -
Start Node applications running on nodes.
> srvctl start nodeapps -n node1 > srvctl status nodeapps -n node1 -
Start All ASM instances from all nodes.
> srvctl start asm -n node1 > srvctl status asm -n node1 -
Start RAC Database Instances on all nodes.
> srvctl start database -d oradb > srvctl status database -d oradb -
Start Oracle Home process accessing database.
> srvctl start listener -n node1 > srvctl status listener -n node1 > emctl start dbconsole
Start and Stop the One node services: ------------------> one by one node down query.
srvctl stop listener -n node_name srvctl stop instance -d <db_name> -i <instance_name> -o immediate srvctl stop asm -n node_name srvctl stop nodeapps -n node_name
srvctl start listener -n node_name srvctl start instance -d <db_name> -i <instance_name> -o immediate srvctl start asm -n node_name srvctl start nodeapps -n node_name
Check scan IP using OS command nslookup
nslookup rac-scan.localdomain
Server: 192.168.0.4 Address: 192.168.0.4#53
Name: rac-scan.localdomain Address: 192.168.0.121 Name: rac-scan.localdomain Address: 192.168.0.122 Name: rac-scan.localdomain Address: 192.168.0.123
nslookup 192.168.0.121
Server: 192.168.0.4 Address: 192.168.0.4#53
121.0.168.192.in-addr.arpa name = rac-scan.localdomain.
nslookup 192.168.0.122
Server: 192.168.0.4 Address: 192.168.0.4#53 122.0.168.192.in-addr.arpa name = rac-scan.localdomain.
nslookup 192.168.0.123
Server: 192.168.0.4 Address: 192.168.0.4#53 123.0.168.192.in-addr.arpa name = rac-scan.localdomain.
============================================================================================================================================
CRSCTL CheatSheet Amit Bansal / 21 May, 2008
You can find below various commands which can be used to administer Oracle Clusterware using crsctl. This is for purpose of easy reference.
Start Oracle Clusterware
#crsctl start crs
Stop Oracle Clusterware
#crsctl stop crs
Enable Oracle Clusterware
#crsctl enable crs
It enables automatic startup of Clusterware daemons
Disable Oracle Clusterware
#crsctl disable crs
It disables automatic startup of Clusterware daemons. This is useful when you are performing some operations like OS patching and does not want clusterware to start the daemons automatically.
Checking Voting disk Location
$crsctl query css votedisk
- 0 /dev/sda3
- 0 /dev/sda5
- 0 /dev/sda6 Located 3 voting disk(s).
Note: -Any command which just needs to query information can be run using oracle user. But anything which alters Oracle Clusterware requires root privileges.
Add Voting disk
#crsctl add css votedisk path
Remove Voting disk
#crsctl delete css votedisk path
Check CRS Status
$crsctl check crs
Cluster Synchronization Services appears healthy
Cluster Ready Services appears healthy
Event Manager appears healthy
You can also see particular daemon status
$crsctl check cssd
Cluster Synchronization Services appears healthy
$crsctl check crsd
Cluster Ready Services appears healthy
$crsctl check evmd
Event Manager appears healthy
You can also check Clusterware status on both the nodes using
$crsctl check cluster
prod01 ONLINE
prod02 ONLINE
Checking Oracle Clusterware Version
To determine software version (binary version of the software on a particular cluster node) use
$crsctl query crs softwareversion
Oracle Clusterware version on node [prod01] is [11.1.0.6.0]
For checking active version on cluster, use
$ crsctl query crs activeversion
Oracle Clusterware active version on the cluster is [11.1.0.6.0]
As per documentation, multiple versions are used while upgrading.
There are other options for CRSCTL too which can be seen using
$crsctl
Or
$crsctl help 11.2 Reference
11.2 introduced few changes to crsctl usage. Most important is clusterized commands which allows you to perform remote operations. They are
crsctl check cluster
crsctl start cluster
crsctl stop cluster
All these commands allow following usage
Default Stop local server -all Stop all servers -n Stop named servers server [...] One or more blank-separated server names -f Force option
Let's see usage
% crsctl check cluster -all
prod01: CRS-4537: Cluster Ready Services is online CRS-4529: Cluster Synchronization Services is online CRS-4533: Event Manager is online
prod02: CRS-4537: Cluster Ready Services is online CRS-4529: Cluster Synchronization Services is online CRS-4533: Event Manager is online
crsctl pin css is used to associate node name with node number. i.e if olsnodes shows prod01 as 1, then it should persist. This is helpful if you intend to run pre 11.2 database ? 1 2
#crsctl pin css -n prod01 #crsctl pin css -n prod02
To check daemon status, following commands need to be used
Check crsd - crsctl check crs
Check cssd - crsctl check crs
check evmd - crsctl check evm
crs_unregister is replaced by crsctl delete resource <resource_name>
crs_stat has been deprecated (though still works) and you need to use