Redstream Blog » Linux/Mac OS

Using Oracle XE on your Mac (part 3): Improving the basics

Pascal Alma — Mon, 08 Dec 2008 15:30:51 +0000

In my previous posts I installed Debian as a virtual server and installed Oracle XE on it. In this post some minor improvements are being described. The first is to make use of a fixed IP address for the Debian instance (in my default setup I made use of DHCP to obtain an IP address). This is done in Debian by editing the file ‘/etc/network/interfaces’.
Originally I had this in the file:

# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interfaces
allow-hotplug eth0

iface eth0 inet dhcp

And I changed it into:

# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interfaces
allow-hotplug eth0

iface eth0 inet static
address 192.168.2.16
netmask 255.255.255.0
network 192.168.2.1
broadcast 192.168.2.255
gateway 192.168.2.1

Now when you restart the virtual server the Debian server will have the IP address 192.168.2.16. You can test this first by performing the ‘ifconfig’ command in Debian. And the real test is to open up a terminal in Mac OS and ping there the set IP:

Next step is to match the name ‘xe-server’ to this IP address in Mac OS. You can do this by editing the hosts file with:
sudo nano /private/etc/hosts
In my case I added the line:
192.168.2.16 xe-server
Now you can access the Debian server with ‘http://xe-server:8080/apex’ or ‘system@//xe-server:1521/xe’ in SQLDeveloper.
This was the last post about this issue. I hope it helps you in some way. Otherwise it was fun to do anyway :-)

Using Oracle XE on your Mac (part 2): XE Installation

Pascal Alma — Sun, 07 Dec 2008 13:20:00 +0000

In this post we will install Oracle XE on the virtual Debian server that we installed in part 1. This installation is described in detail on several sites. I started with the instruction from this site.
With the production release you only need the instructions in the grey box:
First log in as root and add to the /etc/app/sources.list:
deb http://oss.oracle.com/debian unstable main non-free
Perform an update with:
apt-get update
Then we install the oracle-xe package with:
apt-get install oracle-xe
This may take a while since it has to download the Oracle XE package. After the installation you will get this:

Like it says, the next step is to perform the command
/etc/init.d/oracle-xe configure
I accept all defaults so the database is started automatically when the Debian instance is started, but hey, that’s the only reason I made this instance.
When the database is installed, we need to make it accessable from outside the Debian server. There are several ways to this this. The way I did this is to setup a ssh tunnel to the Debian server from my Mac OS so I can access the Apex application with a browser (I found this tip here).
To setup the tunnel (this is done on the host system, Mac OS in our case) open up a terminal window and enter the command:
ssh 192.168.2.32 -L 8081:localhost:8080
(Btw, to obtain the ip address of the Debian server you perform the command ‘ifconfig’ and you get something like the following where ‘eth0′ is the adapter with the needed IP address:

So when the ssh command is issued we login into the Debain box as shown here:

While this ssh connection is open, we can open up a browser in the Mac OS and go to the url ‘http://localhost:8081/apex’ and we end up on the Debian host:

In the Apex application choose ‘Administration’ and on the right side of the screen choose the option ‘manage http access’. In the screen that then appears, select the option ‘Available from local server and remote clients’ and click ‘Apply Changes’.

When this change is done, we can access the Apex application directly from the host with the following url:’http://192.168.2.32:8080/apex’. The ssh tunnel is no longer needed.
Now the Oracle XE installation is ready to be used in the Mac OS and it will take a minimum of the available memory.

Using Oracle XE on your Mac (part 1): Debian installation

Pascal Alma — Sat, 06 Dec 2008 20:51:21 +0000

Since a week I am the proud owner of a MacBook. Besides all new nice features and tools I was getting with the MacBook I knew there was (at least) one thing I would miss: Oracle XE installation for Mac OS. Since I use this XE a lot for research and development I had the workaround for this high on my list of things to do.
While setting up my workaround, I noticed there are already a lot of articles about this issue, but nevertheless I decided to post my solution, worst case it would only be for my own reference in future :-)
Like I said my goal was to setup a virtual server running Oracle XE so I could access the database from Mac OS. My most important goal was to keep the memory footprint as low as possible (and cheap as possible). So setting up a virtual server running Windows was not an option! I choose to use Debian Linux, it’s free and without the graphical user interface it has a very low memory footprint. The rest of this post I will describe how I set up Debian with Sun’s Virtual Box. In the next post I will describe how to install Oracle XE on the Debian server.
First step is to download Sun xVM Virtual Box for Mac OS and install it on your MacBook. Next step is to download the Debian software here.
When this download is done we can create a virtual machine in VirtualBox and choose Debian as operating system. Make sure you map the downloaded iso for Debian as CD/DVD drive like shown here:

Another important thing is to set the Network Adapter to ‘host interface’ so you can use your current internet connection that you have in Mac OS automatically in the Debian instance (assumed you have an internet connection!). The virtual hard disk is set to max 8 GB. This will be sufficient for me to run Oracle XE but you can create a bigger one of course if your going to do more with this server. At the end you must have something like this:

When this is done you can start the virtual machine and you will be asked to setup Debian. The most of the installation is quite simple, only one thing I did especially for Oracle XE was define a 1.5 GB swap disk space at my virtual hard disk (I was told this was mandatory but I am not sure about that). This is how I did this:
Somewhere during the setup you will be asked to partition the hard disk. When this appears you choose ‘Manual’ from the menu:

Next choose the hard disk to partition (for the virtual machine newbies: don’t worry about your Mac OS installation: this hard disk is your virtual one and will not affect your Mac OS installation!)

Continue the setup to partition the first part of the disk as normal space. Just make sure you reserve 1,5 GB for the swap space. Here are the steps to perform:

width=”600″ />

Make sure here you’ll keep 1.5 GB available

Now the main part is setup. Now select the 1.5 GB free space to create the swap space.

Here is the important part. Select the ‘Use as’ and give to make the other option available.

Pick the ‘swap area’ as your choice and proceed.

Finally choose ‘yes’ to perform the partitioning of the disk.

Next question is to add a network mirror, which I didn’t to keep the installation as minimal as possible:

If you are working behind a proxy you will get the following screen that takes a while to disappear:

And after a while you will get (but we will fix this later, but only if you are after a proxy):

The last choice you have to make is the following:

I deselected both options, again to keep the footprint as minimal as possible. After this, proceed with the process and fill in the mandatory answers. At the end of the installation the system will restart and the installation is finished.
Now it is time to shutdown the guest system and modify the virtual machine settings in VirtualBox. If we don’t do anything, the system will start to install again after we start it again. In the VirtualBox change the boot order so the hard dsik preceeds the cd-rom from now on, like shown here:

After this is changed, startup the Debian server again and login as ‘root’ user (normally you don’t do this but for the following task it make things a little easier).
If you are working behind a proxy then give the following command to enter the proxy:
nano /etc/profile
Add the following line to the file:
export http_proxy=http://user:password@my.proxy.host:proxy_port
Save the file and perform the following command to load the profile:
source /etc/profile
Next step is to modify the sources.lst file. This file contains the repositories which Debian uses for downloading necessary tools, libraries, etc. To open the file go to the directory ‘/etc/apt/’ and open the file ‘sources.list’ with the ‘nano’ command:
cd /etc/apt nano sources.list
If you are behind a proxy you have to uncomment the lines:
deb http://security.debian.org/ etch/updates main contrib deb-src http://security.debian.org/ etch/updates main contribAnd add the following lines:
deb http://ftp.debian.org/debian unstable non-free deb http://ftp.nl.debian.org/debian/ etch main
Where the last line is depending on the location where you are.
Here is the complete content of the file:

Now we have set the repositories and set the proxy, if needed, we can get the necessary updates by performing the command:
apt-get update
If everything went well you will see the system downloading several packages. When the update is finished, we install one package, which we will need later and that is the ‘ssh’ package. With this package also other dependent packages are installed. To install the package we issue the command:
apt-get install ssh
And that’s it. We now have a Debian installation ready to have Oracle XE installed on it, which I will show in the next post.

Monitoring your Linux server

Pascal Alma — Fri, 28 Sep 2007 11:13:32 +0000

As I stated a post ago I am not very familiair with the Linux operating system. I know some basic commands to browse the system and to perform some filesystem actions. Most of the time that is sufficient. Usually I do the development of the software on a Windows workstation and then deploy it to a J2EE container running on some version/distribution of Linux. However this time I had to perform some performance tests on our applications while it was running at a (virtual) Linux server (RedHat 4 distribution, to be more specific). Usually I am able to find someone else to do the Linux thing :-) but this time I had to do it myself. Luckily there was my great friend Google that helped me in this matter! I quickly found out there is a ‘sar‘ command in Linux which will log all kinds of parameters about the system resources in a daily log file. And you can also use this command to create reports of the collected data. Although that was what I needed, what I really wanted was ‘ksar‘. This is a Java tool that you can run on your Windows machine (it’s Java so you can run it anywhere, right?) and have it creating graphs based on the ‘sar’ output. You should really check it out if you’re interested in the use of the resources on a Linux system. Here you see the tool at work: