842366 wrote:You want a solution that is free of charge. right?
is there a method to create a failover with single instance database 11.2 with oracle cluster without using script to start/stop/check?
AJ_23 wrote:Have you an example?
Instead of giving options you didn't ask for, like RAC and DG...here are a couple of things to look at that may help with a single instance failover in a Grid Cluster w/o other licensable features....
http://www.oracle.com/technetwork/products/clusterware/overview/si-db-failover-11g-134623.pdf - Old white paper, but worked for me in 22.214.171.124 ...may need some script tweeking
Marko Sutic wrote:Datafiles are on ASM so there's no problem with shared storage, but i must duplicate instance on others node with consume of ram and cpu.
With DG you have to think about additional storage for standby database and with failover cluster you can use shared storage.
(start/stop/check) scripts are very simple and very reliable with managing resources.
So if i perform a relocate:
crsctl modify serverpool ora.TEST -attr "MIN_SIZE=2,MAX_SIZE=4,SERVER_NAMES=rac1 rac2" -f
after a crs stop:
[root@rac1 ~]# crsctl relocate res ora.test.db -n rac2 CRS-2673: Attempting to stop 'ora.test.db' on 'rac1' CRS-2677: Stop of 'ora.test.db' on 'rac1' succeeded CRS-2672: Attempting to start 'ora.test.db' on 'rac2' CRS-2676: Start of 'ora.test.db' on 'rac2' succeeded ora.test.db 1 ONLINE ONLINE rac2 Open
[root@rac2 ~]# crsctl stop crs ... CRS-2676: Start of 'ora.rac2.vip' on 'rac1' succeeded CRS-2676: Start of 'ora.test.db' on 'rac1' succeeded ...
ora.test.db 1 ONLINE ONLINE rac1 Open