Disk Enclosures
1751939 Members
5022 Online
108783 Solutions
New Discussion

Re: Oracle database replication using Business Copy EVA

 
Jozef_Novak
Respected Contributor

Oracle database replication using Business Copy EVA

Hello everyone,

 

I have an environment with two systems running HPUX 11.31 and an HP EVA6400 licensed with BC. One of the servers is hosting a production Oracle database instance, the other one several test/acceptance instances. The idea is to regularly make several copies of production data using Business Copy onto the Vdisks owned by the test/acceptance instances in order to keep the non-production data up to date. Before I start writing shell scripts with RSM CLUI jobs, I would like to ask if there is any advice about the best concept or standard implementation of this ? 

 

Many thanks,

Best Regards,

Jozef

2 REPLIES 2
Jan Soska
Honored Contributor

Re: Oracle database replication using Business Copy EVA

Hello Jozef,

probably most important thing is how oracle will work with the data. From SAN point of view it is simply to make snapshot/clone of lun where are db files, but first you have to push oracle to accept them.

My sugestion is to make some manual tests first, only then try to automatize it.

 

Jan

Jozef_Novak
Respected Contributor

Re: Oracle database replication using Business Copy EVA

Hello Jan,

 

the process is to be repetitive, i.e. the snaps / clones will be recreated as needed by the database admins, and in order to automate the whole procedure it is mandatory that the device files on the secondary host will not change. I suppose that this can be achieved by using clones which will be converted to containers right before the replication starts, as the container retains properties of the original virtual disk (this approach cannot be used for snapshots). In terms of Oracle accepting the cloned data, this is not a concern here. The task is only to provide the database with the updated files, so the idea is to halt the secondary instances, export their volume groups, replicate the production data and import the volume groups with the new data using unchanged device files. I am trying to find out whether anyone is using this (or an alternative) approach.

 

Jozef