Time Required: 60 minutes
Class Materials:
Next we are going to create two networks for our RAC system to talk on.
The Oracle RAC system requires at least two network connections between cluster nodes. One network will be the public IP and the second will be a private IP reserved for inter-cluster traffic.
Time Required: 20 minutes
Class Materials:
Next we will install the ASMLib drivers that simplify management of ASM on Linux.
ASMLib is an interface between block devices presented to Linux, and the Oracle ASM software that executes as the Oracle user.
Time Required: 20 minutes
Class Materials:
Next we need to create the Oracle user account, OS groups and mount point.
The Oracle software will be installed and owned by the oracle user account. In additional several OS groups are created that allow other non oracle OS users privileged access to the database and grid resources.
Time Required: 20 minutes
Class Materials:
Next we need to configure the Linux kernel to support Oracle 11gR2.
Before we can install Oracle 11gR2 on our new VM, we need configure the Linux kernel. The following steps modify key settings to allow Oracle to execute. These steps are taken from the Oracle install guide located at the following URL:
Time Required: 20 minutes
Class Materials:
Next we need to add additional RPMs to support the Oracle database and grid installs.
Before we can install Oracle 11gR2 on our new VM, we need to load additional RPM packages that are required by the Oracle database and grid. Oracle publishes a list of the required VMs broken down by the supported Linux distributions at the following URL:
Time Required: 60 minutes
Class Materials:
To create our Oracle RAC database on VMware Workstation 8, we first need to create a Linux VM.
In this example I have the Red Hat Enterprise Linux 5.5 64-bit install software as an ISO image of the DVD install media. Having a single ISO image allows VMware to use its Easy Install feature that pretty much automates the OS install entire process.
It has long been an ambition of mine to have my very own RAC cluster database.
My former TUSC colleague and good friend John Smiley over at Amazon.com created a white paper some years back where he showed how to create a two node RAC cluster out of a couple of spare x86 Linux boxes and as I recall, some scotch tape, glue and pixie dust. It really was very cool and definitely worth a Blue Peter badge.
Once a new physical disk is added to a Linux host, it needs to be presented to the OS.
This is a five step process:
When new SCSI devices are added to a Linux system, it is necessary to re-scan the SCSI bus so that the OS can add them to the device list:
[root@localhost ~]# ls /dev/sd* /dev/sda /dev/sda1 /dev/sda2 /dev/sda3 [root@localhost ~]# su -c 'for hst in $(ls /sys/class/scsi_host) ; do echo "- - -" > /sys/class/scsi_host/$hst/scan ; done' [root@localhost ~]# ls /dev/sd* /dev/sda /dev/sda1 /dev/sda2 /dev/sda3 /dev/sdb [root@localhost ~]#
Back in 2005 I got interested in reverse engineering the Oracle data block.
I had already spent far too much time breaking apart the redo blocks, and had published an interesting paper on the matter called Disassembling the Redo Block which Frank Naude over at orafaq was kind enough to host for me. This was back before “blogging” became the ubiquitous pass time of Oracle DBAs it appears to be today.
Rich Niemiec at TUSC forwarded to me an interesting article on the Terilingua Block Viewer for Oracle, and so after a lot of searching around I found the BBED tool. What’s more, BBED was included in the standard Oracle distribution for Windows and Linux. On Linux all you had to do was link it.
And so armed with an old IBM Aptiva running Linux 8 Pro and a copy of Oracle 9i, I set about trying to figure out how the blocks of a database held together.
The result was a detailed analysis of BBED and how to use it. Although obsolete today, the article is still interesting to those DBAs who want to understand the inner working of Oracle. At the time it was quite popular, being cited by Pete Finnegan and Don Burlseon among others.
I include it here for posterity.
And no, you still have to work out the password for yourself 🙂
the gruffdba 