RAC daily management (10G)

RAC daily management (10G)

1. RAC concept (Real Application Clusters)
multiple instances run on multiple servers,
a database is placed on shared storage, all instances can access
, data and information are exchanged between instances through an intranet

2. Features of RAC
In addition to ordinary database features:
each node instance has its own SGA,
each node instance has its own background process,
each node instance has its own redo logs,
each node instance Have their own undo tablespace

3. RAC daily management

1. The daemon process of the
   RAC – /etc/init.d/init.*
   ll init.*-fuzzy search,
   including: init.crs / init.crsd /init.cssd / init.evmd is generally started with the system
   init. crs starts the next three processes

Case operation:
Allow crs to start with the operating system
#/etc/init.d/init.crs enable

Disable crs to start with the operating system
#/etc/init.d/init.crs disable

2. Application layer management
crs_stat -t
crs_stat This command It is used to view the running status of all resources maintained by CRS. If there is no parameter, the summary information of all resources is displayed.

srvctl config
can view the resource information of the database and instance registered in CRS.
srvctl config database -d dbname

srvctl config listener- n rac1
srvctl config asm -n rac1

The ocrcheck
command is used to check the consistency of OCR content. This command does not require parameters

crsctl check crs
can be used to check the status of each crs process, manage Votedisk, track CRS process function

RAC stop script
1. Check the RAC environment
crs_stat -t / crs_stat- ls

2. Shut down the database
srvctl stop database -d PEDBDB - stop all
srvctl stop instance -d PEDBDB -i PEDBDB1 - single instance stop
3. Stop 1 node ASM instance
srvctl stop asm -n rac1
4. Stop 1 node NODEAPPS instance
srvctl stop nodeapps -n rac1
5. Stop 2 node ASM instance
srvctl stop asm -n
rac2 4. Stop 2 node NODEAPPS instance
srvctl stop nodeapps -n rac2

RAC startup script
1. Check the RAC environment
crs_stat -t / crs_stat -ls
2. Start 1 Node NODEAPPS instance
srvctl start nodeapps -n rac1
3. Start 1 node ASM instance
srvctl start asm -n rac1
4. Start the 2-node NODEAPPS instance
srvctl start nodeapps -n rac2
5. Start the 2-node ASM instance
srvctl start asm -n rac2
6. Start the database
srvctl start database -d PEDBDB - all start
srvctl start database -d PEDBDB -o mount- -All start to mout

Note:
1. EDBDB is db_name, PEDBDB1 is the instance name, rac1/rac2 is the hostname
2. After crs is set and the operating system is started, the RAC will start automatically when the server is turned on

4. Manually register and listen to the cluster
1. Delete the invalid listener resource
crs_stop ora.rac1.LISTENER_RAC1.lsnr in crs 2.
unregister (unregister from the resource)
crs_unregister ora.rac1.LISTENER_RAC1.lsnr

The location of the parameter file:
$CRS_HOME/crs/public or /u01/crs/oracle/product/10.2.0/crs/crs/public
3. Manually register
crs_profile -create ora.rac1.LISTENER_RAC1.lsnr -t application -a / u01/app/oracle/product/10.2.0/db_1/bin/racgwrap --Create resource. The

modification content is as follows:
DESCRIPTION=CRS application for listener on node
HOSTING_MEMBERS=rac1
PLACEMENT=restricted
REQUIRED_RESOURCES=ora.rac1.vip

4. Register and start the monitoring, later will be automatically
crs_register ora.rac1.LISTENER_RAC1.lsnr
--register crs_start ora.rac1.LISTENER_RAC1.lsnr --start with the server