No description, website, or topics provided.
Shell
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
12cR1
stagefiles
.gitignore
.gitmodules
README.md
Vagrantfile

README.md

racattack-ansible-oracle

This is a joint work between 2 projects.

https://github.com/racattack/vagrantfile

and

https://github.com/oravirt/ansible-oracle

The idea is provide a fully automated racattack build from vm to database, 100% hands free

Before start

If you are new to vagrant, the suggestion would be test the standalone vagrant project, which is documented here: http://en.wikibooks.org/wiki/RAC_Attack_-_Oracle_Cluster_Database_at_Home/RAC_Attack_Automation

This project

Clone as: git clone --recursive https://github.com/racattack/racattack-ansible-oracle

cd racattack-ansible-oracle

edit Vagranfile and adjust number of nodes, ram and size of shared disk

#############################
##### BEGIN CUSTOMIZATION #####
#############################
#define number of nodes
num_APPLICATION         =   0
num_LEAF_INSTANCES      =   0
num_DB_INSTANCES        =   1
#
#define number of cores for guest
num_CORE                =   1
#
#define memory for each type of node in MBytes
#
#for leaf nodes, the minimun can be  2300, otherwise pre-check will fail for
#automatic ulimit values calculated based on ram
#
#for database nodes, the minimum suggested is 3072
#
memory_APPLICATION      =   1500
memory_LEAF_INSTANCES   =   2300
memory_DB_INSTANCES     =   3072
#        
#size of shared disk in GB
size_shared_disk        =   5
#number of shared disks
count_shared_disk       =   4
#
#############################
##### END CUSTOMIZATION #####
#############################

place oracle binaries at 12cR1/

For 12.1.0.2:

    linuxamd64_12102_database_1of2.zip
    linuxamd64_12102_database_2of2.zip
    linuxamd64_12102_grid_1of2.zip
    linuxamd64_12102_grid_2of2.zip

For 12.1.0.1:

    linuxamd64_12c_database_1of2.zip
    linuxamd64_12c_database_2of2.zip
    linuxamd64_12c_grid_1of2.zip
    linuxamd64_12c_grid_2of2.zip

For 11.2.0.4:

    p13390677_112040_Linux-x86-64_1of7.zip
    p13390677_112040_Linux-x86-64_2of7.zip
    p13390677_112040_Linux-x86-64_3of7.zip

For 11.2.0.3:

    p10404530_112030_Linux-x86-64_1of7.zip
    p10404530_112030_Linux-x86-64_2of7.zip
    p10404530_112030_Linux-x86-64_3of7.zip

First boot

vagrant up will create the machines This process can take time as here is where the shared disk will be created.

If you wish to reinstall, and avoid waiting on this step, you can use setup=clean vagrant provision

Installation

setup=standard vagrant provision will setup an standard 12.1.0.2 oracle cluster

setup=flex vagrant provision will setup an flex oracle cluster

if any leaf node is defined, cluster will be defaulted to flex

At this moment, defaults are giver and dbver 12.1.0.2

It is also possible to have different combinations of GI/DB installations. The supported versions are (at the moment): 12.1.0.2, 12.1.0.1 and 11.2.0.4, 11.2.0.3

It works like this:

setup=<standard|flex> [giver=<12.1.0.2|12.1.0.1|11.2.0.4|11.2.0.3>] [dbver=<12.1.0.2|12.1.0.1|11.2.0.4|11.2.0.3>] vagrant provision

so if you want a 12.1.0.2 GI installation and a 11.2.0.4 database, it should look like this:

setup=standard dbver=11.2.0.4 vagrant provision

Clean up for reinstall

setup=clean vagrant provision will delete /u01 , clean up previous installation, bring shared disk to clean state, then just stop the machines. vagrant halt.

then a vagrant up will start a clean environment. This is useful to avoid recreating the shared disk since is time consuming task.

Delete

vagrant destroy -f will delete all the VMs and the shared disk

alvaro @ kikitux.net

oravirt @ gmail.com