Scott Pilkinton's Information Technology site

If you have been exposed to VMware, you know what I’m talking about when I start talking about how great it is.

In a nutshell, VMware will load on top of your Windows operating system and provides a ‘Virtual’ PC inside of your existing computer. It emulates everything to the point where an OS loaded in a Virtual Machine thinks it’s installed on any regular computer. One of the best things about it for I.T. Professionals is that it can be downloaded for free! There are two VMware products of interest that I will mention:

The first is the Server software which you can download here. This software will allow you to create the Image needed before an OS can be loaded on a Virtual Machine. During the image creation process, you will define how big you want the disk, memory, etc… Keep in mind when you create these images and define the resource parameters available, that is what will be taken away from your host Operating System for the Virtual machine. These resources are allocated dynamically, when needed, and only while the Virtual Machine is actually running.

Once you have created an image and loaded your Operating System of choice, you can get by with just the VMware player – which is the second product I alluded to. The player basically allows your VMware images to become portable. You can take a VMware image along with the small player program to any machine and boot the VM up for use.

I could go on about VMware, but that’s not really the point of this post. The nice thing all this has brought along with it is what VMware calls ‘Virtual Appliances‘. These are basically images that someone else has created (usually for specific purposes) and made them available to anyone who wants to download and use them. The Virtual Appliances can then be booted on any computer with the VMplayer installed. These images will typically have an Open Source operating system loaded with various software to serve a specific purpose. For example, you can download a fully installed version of a new Linux distribution. This would allow you to boot a Linux OS up just as if you had just went through the entire OS installation. With this work already done – it is much simpler to download and run something that has already been installed and ready to use.

Here are some examples of what is available, ready to use:

Virtual MYSQL Server
Virtual Appliances MySQL Server is a free to use nano-sized virtual machine hosting a MySQL server, and phpMyAdmin for easy web based administration.
The Virtual Appliances MySQL Server is just 22MB to download but provides almost 1.8GB of internal space for use by MySQL.

Network Security Toolkit (NST) Virtual Machine
The Network Security Toolkit (NST) is a system that provides easy access and management capability to best-of-breed Open Source Network Security Applications. The purpose for development of the toolkit was to provide the network security administrator with a comprehensive set of Open Source Network Security Tools for testing, assessing, analysis, monitoring and validation of security architectures.

rPath Port 25 Mail Server Appliance
The Port25 Appliance brings the best open source mail tools into one easily-deployable form factor. Whether your mail protocol of choice is SMTP(S), IMAP(S), or POP3(S), Port25 can deliver. In addition, Port25 includes mail filtering, anti-virus protection, and spam filtering capabilities.

OSCAR Cluster Headnode
This virtual appliance is based on the OSCAR (Open Source Cluster Application Resources) cluster infrastructure, a collaborative effort to make beowulf-type clusters easy to use and manage. The author is core OSCAR developer and uses VMware’s VMplayer on a regular basis for developing and testing OSCAR.

Again, keep in mind, these are ready to run platforms. This saves having to go through the sometimes complex installation and configuration of software, allowing you to get right to using them. The images can be used just in place of traditional server deployments!

Ok, this post has gotten long – I’ll follow up with the ‘nice network tool’ in a follow up post… (As you can probably guess, it is available for use as a Virtual Machine download!)

This is a post from the new feature in Word 2007 that allows you to create and post blog entries directly to your (supported) blog.

I just downloaded several of the new Beta 2 Office Family products and performed an upgrade to my existing Microsoft Office 2003 Installation.

Everything is looking good so far, I have to run for now. Look for more information soon!

 

good grief….  My domain name expired and earlier today I lost resolution to my domain through my ISP.  Network Solutions had transferred it to one of their pages with an explanation stating that it had expired.  Well, so much for transferring it to godaddy this year (where all my other domains are registered at).  You can’t transfer a domain to another registrar after it has reached the expired status.  So, I bit the bullet and paid full price to NS for another year.

Oh, and yes, they did send me three or four reminders during the month of January.  However, it didn’t help me since I wasn’t monitoring the account the emails were coming to.

Lesson learned:  If you are going to use more than one email address – for whatever reason – check them for goodness sakes!!!

There has been tremendous noise coming from Linksys lately regarding VOIP products.  A press release was release today between Airband and Linksys detailing a trial that will begin in Dallas using the Linksys One Solution (pdf).  Airband will use a product called the Cisco Systems Service Node to provision and route calls for the Linksys One Solution.  The services offered by this solution are meant  to bring technology to small business with minimal fuss.  The goal is to provide end-user customers with a turnkey and easy-to-install solution that supports business-quality voice, video, data networking, business applications and Internet access through a single high-speed connection.

Along the same lines – Linksys has announced several small office SIP based VOIP products that includes a PBX Key system capable of supporting 16 users. Along with the Linksys IP PBX(SPA9000), several phones (SPA901, SPA921, SPA922, SPA941, SPA942), and an Analog PSTN Gateway + Phone Adapter(SPA3000) make up the total solution.  This makes for a very nice (and complete) phone system package for the small/home office.

I’m anxious to hear more about these products as the early adopters deploy them…

Well, after 6 years I am leaving my current employer and going to work for another company.  I am excited about the new opportunity and look forward to starting.  In the meantime, I am in the process of trying to get everything wrapped up – so that I can leave.

I purchased a Dell Axim x50v handheld last year and was generally happy with it from the start.  The default OS at the time was Pocket PC 2003 SE, with an available upgrade to Windows Mobile 5 promised in the future.

Dell charged 40 bucks for the version 5 upgrade a month or so ago when it was generally released.  Of course, I immediately performed the upgrade after its release – only soon to be sadly disappointed.  The WM5 OS clearly wants more hardware resources than the previous 2003 version.

My one year old PDA had now become close to unusable.  The unit had become generally sluggish in any application I tried to run.  Apparently, there has been a discovery identifying the process that eats up all the cpu which causes it to become extremely slow.

There are several ‘tricks‘ attempting to bring the little booger out of its coma – however, it appears to produce limited results.  Thankfully Dell has acknowledged it is not ready for prime time and released a utility to downgrade the Axim back to Windows Mobile 2003.

Personally, I haven’t seen to many reasons to upgrade.  I do like the idea of persistent storage – which ends the chance of loosing your data in the event of a complete battery loss. That however, it not enough for me to deal with all the other problems in upgrading.

My opinion – an Axim x51v probably runs WM5 fine.  However, if you have a x50v – you are better off running 2003 SE…

Want to keep your Redhat/Fedora/RPM Based Linux system up to date? Well, here is ‘yum’ to the rescue.

yum (yellowdog updater, modified) has been developed by Duke University. Here is their definition of yum:

Yum is an automatic updater and package installer/remover for rpm systems. It automatically computes dependencies and figures out what things should occur to install packages. It makes it easier to maintain groups of machines without having to manually update each one using rpm.

For the curious, you may type ‘yum’ and press enter to see a list of available arguments to run this program with. So, we want to check our system to see if there are any updates available. In order to ‘view only’ available updates, you can type: yum check-update

You will probably get a list of packages a mile long – these are updates to packages that are currently installed on your system.

Note: If you get an error message like: “Cannot find a valid baseurl for repo: updates-released” – this most likely is because you are behind a proxy server and can’t reach the Internet directly. I just love those straightforward error messages :>) … To fix – type the following in at the prompt: export http_proxy=http://proxy:port and this will set an environment variable that many programs can use, including yum.

To go ahead and update your system type at the prompt: yum –y update . Note: If you don’t use the –y you will have to answer several prompts manually, which is not necessary.

After the update is finished, reboot the system by typing init 6 .

After the machine comes back up, you should now have a current, patched system. You can keep your system current by enabling nightly yum updates. To do this first you want to enable the yum service by entering the command: chkconfig yum on . Next, start the yum service by entering the command: service yum start .

This documentation is provided so that you may use to successfully install a very ‘core’ installation of Fedora 4. Most people that aren’t Linux guru’s/experts, quickly get lost in all the package selection menus, etc…. It is my intention to provide instructions on how to build a very ‘light’ installation of Fedora for a server application.

To get started, I will just mention hardware specs just for the sake of being complete in my documentation. I am starting with a Dell 1650 1U Server with (1) 1200MHz Processor, 1Gig Memory, and a hardware Raid 5 configuration with approx. 33 gig of usable disk space.

First thing I did was reboot the server with the CD 1 of Fedora inserted in the drive. When the Fedora Core install or upgrade menu appeared, I pressed the Enter key to start the installation in graphical mode. At this point the computer will begin booting from the CD-ROM.

The first prompt that appears is telling you that it found the Fedora Installation CD in the drive and it wants to test the CD to make sure all the files are intact that will be used to install your system. I have found this can take a while (since it checks each individual file), and do not see much benefit from it. This is of course assuming that you successfully downloaded the .iso and burned the image to CD without any problems. So at this point, I hit the tab key to select the ‘Skip’ button and press Enter.

The next thing you will see is the ‘graphical’ part of the installation, which is the Anaconda installer. The first page is the welcome page, which you may click on Next (Hopefully your mouse works, if for some reason it doesn’t – you can use tab/enter). Don’t get impatient and think you didn’t click Next – just hang out a sec and wait for the system to respond (it is reading from the CD).

At the Language Selection screen, highlight the appropriate selection by clicking once and then click on Next.

Same thing with the next screen, Make the appropriate selection for the Keyboard Configuration (U.S. English in my case) and click on Next.

Now, the next process that runs will determine if your system has any existing Linux installations. If it finds Linux already installed (which it was in my case), the radio button will be selected to upgrade an existing installation. Note: If you have Windows on the machine already – you will not see this mentioned screen, so just skip this paragraph. Since I am wanting to re-partition/format/etc and start with a fresh machine, I will click the Install Fedora Core radio button and then click next.

The next question will ask which installation type you would like to choose. Since we are going with a minimal install, select the Custom radio button and click on Next.

Disk Partitioning will be the next task that appears, I like to use the automatically partition option (in most cases) and then you have the opportunity to make any adjustments on the next screen. So, with automatically partition selected – click on Next.

We will then have Automatic Partitioning ask us the question of what we want to do with any existing partitions. Again, this is a ‘clean’ install so I select ‘Remove all partitions on this system’. NOTE: It is not my intent to document setting up a ‘dual boot’ configuration where there may be another operating system installed that you want to keep on the system. The selected default on this screen is ‘Remove all Linux partitions on this system’ – This will not remove any partitions that are not Linux related. So, if you had a Windows partition it would leave it intact, thus leaving you with less disk space (or maybe none at all) to setup your new Linux installation. So, be sure to select the radio button to remove all partitions (which will make your entire disk available for this installation). I only have one drive to select from in this installation, which is a DELL PERCRAID RAID5 selection. Machines with multiple standalone disks will have more than one option. Most of the time you will select the first drive listed. Leave the Review the partitions created box checked and click on Next. At the warning, confirm by clicking yes that you want to destroy your current (old) disk configuration.

On the next screen, observe the auto partitioning configuration for the installation. The default partitioning will leave you with three partitions. One will be your swap partition, the operating system will use this space for paging. The size of your swap partition is decided based on how much RAM is installed on your system. The next will be your boot partition and will be mounted as ‘/boot’. The boot partition is where the Linux Kernel and other core system files are stored that are needed for your system to boot (Typically creates a 100Meg partition). The other partition will be used for all your files and will be mounted as ‘/’. This basically provides you with one big file system to use for all your Linux applications and data. As mentioned earlier, I usually don’t change anything here and just click on Next.

You must have a boot loader (next screen to appear) in order for any operating system to boot on your computer. In most all other types of Operating Systems (besides *NIX), this is a ‘hidden’ step that is not mentioned in the installation. However, since Linux is the great and customizable operating system that it is – we have the option to customize. I don’t want to change anything on this screen, so I just click next.

Following the boot loader screen will be a Network Configuration screen. Here you have the option of selecting which network card(s) in your system will be used. Make sure the card you have intended to use has a checkbox marked in the ‘Active’ on Boot column. If you don’t change anything else on this screen, your computer will automatically request an IP address using DHCP when it boots. This might be fine in some cases where you use DNS resolution to access the machine, but in most cases you still want static IP addresses defined for Server class machines. So, since that is what this document is intended for, we will set static information. Highlight the Adapter being used (in my case and probably yours eth0) and click the Edit button.

In the popup window, uncheck Configure using DHCP and then manually enter the desired IP address/netmask and click on OK. Back at the Network Configuration window, we then set the hostname manually by entering the fully qualified name we want the machine to be known as.

Since we are setting a manual configuration, the Miscellaneous Settings must be entered as well. Enter the Default Gateway for your machine and at least one DNS server entry – click on Next.

The next step is the Firewall configuration – I would highly recommend leaving the Firewall Enabled. We will have a simple interface when we are finished that will allow us to add any necessary additional ports to be opened. I go ahead and check Remote Login (SSH – which is TCP port 22) along with Web Server (HTTP,HTTPS – which is TCP port 80,443 respectively). SELinux is the last configuration option on this page – for now I will change the default setting to disable SELinux by clicking on the drop down mention and selecting Disabled. Note: SELinux is something new for Fedora to help prevent malicious programs from running on the system. This feature can sometimes cause problems with legitimate programs that are installed. However, in most cases leaving SELinux active will not cause a problem. Click on Next.

In the time Zone Selection, you may find an area that uses the same time zone as your location and then click Next.

The root password is something that should be selected carefully. Please do not enter something here that is easy to guess. The root account allows access to anything and everything on your system. We will setup additional user accounts later in the document. The root account should only be used when temporary elevated privileges are needed on your system. Click on Next after you have confirmed the root password.

You will see a window that follows stating that the installer is Reading package information. This is basically used to provide you with that big menu of selections regarding what you want on the system. So after all the package information is read, we are given the opportunity to go through all the available selections. I will make this one easy, scroll all the way to the bottom of the page and under the miscellaneous section check the Minimal box. You will notice that will grey out any other options that are on this menu and will only install the files that are necessary for a base Fedora system to operate. This should consume approx 628Meg, click on Next.

You will see one last screen telling you it is about to start the installation, you make click on Next to get the party started.

The installation process took about 30 minutes, mileage will vary. At completion, your CD tray should open, remove the CD. At the Congratulations, the installation is complete window – you may click on Reboot.

At this point, your system should reboot and automatically boot after x seconds to the new Linux installation that was just performed. This particular installation configuration provides a command line interface only – which will be sufficient for a server application. Instructions will be provided in a separate post that describes how to completely manage your Linux machine from a web based interface.

After the machine is finished booting, you should see a login prompt. Since the root userid is the only user that currently exists, we will use it. Login with root as the username and the password is what you set previously during the installation.

At this point you are provided with a shell prompt waiting for you to take it somewhere.

Congratulations!  You have successfully installed a minimal Fedora Core OS.  The next article titled “Update your Fedora OS using yum” is the next in the series.  This next article will help you keep the Operating System up to date with all the latest software.

In a previous post, I talked about my new love for Trustix. I will attempt to document another installation using Trustix to act as a Windows Server for Microsoft Domain Controller Services, File Sharing, and a web server. Even though I am going through this process for my home server – this is the exact route one might take to create an ‘all-in-one’ server for a small business.

First the installation was a PXE install – except this time I installed additional packages. During the install, I selected the following: Minimal install with ssh access, Web Server, and Domain Services. The Domain Services selection refers to Windows File and Print sharing.

One quick command installed Webmin from a shell prompt. I went to www.webmin.com and clicked on the RPM download. After picking out a mirror from the sourceforge mirrors – I copied that shortcut to the clipboard for pasting into the terminal session open to the Trustix box. Use the following command to install: rpm –install (paste mirror here). The install took just a couple of minutes and then it was ready for connecting to http://host:10000. Webmin uses root as the user name along with the associated password to access the control panel.

To configure SAMBA click on Servers, then SAMBA Windows File and Print Sharing. You are now in the SAMBA configuration, so find the Global Configuration section and click on SWAT. You must provide the root user log-in information again to access the SWAT configuration.

From the Samba Web Administration Tool, click on the Wizard button. In the Samba configuration Wizard, click on the Edit Parameter Values button. On the Wizard parameter Edit page use the following parameters:

Workgroup – This is going to be the domain used to name your environment
realm – This is going to be fully qualified domain name of the Trustix box.
netbios name – This is going to be the same as the hostname of the Trustix box

Security Options – user

WINS Support – Yes

Now click on Commit changes to return to the Wizard.

Set the Server type to: Domain Controller
Configure WINS as: Server for client use
Expose Home directories: Yes

Click on the Commit button. After the changes are made – click on the Status button and restart all the services. Down at the bottom right of the page, click on Logout of SWAT and you will be carried back to the Webmin SAMBA section.

Down toward the bottom of the SAMBA section click on Configure automatic Unix and Samba synchronisation. Here you can setup automatic SAMBA user creation when users are added from the Webmin console. If you have existing users that were setup during the OS installation or before Webmin was installed – you will want to run the Convert Unix users to SAMBA users. This will only need to be performed once if you are going to configure automatic sync between UNIX and samba users.

At the very minimum, you should now be able to browse to the Linux server from a Windows machine using Network Neighborhood or click on Start –> Run and type in \\trustixserver and you should get prompted from a username and password. Once authenticated – the default shares you should see are netlogon, your home directory, and a Printers folder.

Yes, the environment around the I.T. department continues to change. Soon it will be a distant memory of the past – where the definition of a ‘good’ I.T. employee was a knowledgeable techie in their area (I don’t like the word geek).
 

Information Technology departments are looking for more than someone who just is good at ‘tech stuff’.  Things like project management, professional development, and manager related skills (even if your not a manager or even headed in that direction) are in more current I.T. job descriptions than not.

No matter what industry you are in: there is always another side to the I.T. department – which is commonly known as ‘the business’.  Now granted, without the business side there wouldn’t be an I.T. department.  At the same time, it is still probably one of the hardest areas to interact with.
 

For the most part, ‘the business’ doesn’t understand what I.T. does and for the most part – I.T. doesn’t understand what ‘the business’ does.  However, the last statement is rapidly changing.  In order to educate or even interact with business side – most all I.T. employees are being molded to interact in some way with the business side.  Don’t expect the opposite trend anytime soon – business people only want to share their needs and let someone else figure out how to get it done.  Their job is about improving the bottom number – nothing else.
 

Often, I.T. employees get accused of talking in another language to people outside of their department.  I think that after being in I.T. for a while – many people forget how to talk ‘normal’.  Even using ‘normal talk’ would be better than using all the new words learned in the tech world while talking to people on the business side.
 

Often this only temporary – especially for I.T. people who are seeking advancement of any kind.  They finally start to learn the next phase by adding business ‘lingo’ to their vocabulary.  Before long, the business vocabulary starts flowing out much like the tech talk once did.  One might argue (I sure would) that tech talk and business lingo can both be very annoying when overused.
 

That would bring us to the ‘peak’ where the pursuit of a well-rounded vocabulary takes place.  This is basically the process where you try and regain your normalness (which is where you started), only now you have all that tech and business knowledge/vocabulary that you could draw and fire at any time.
 

Caution: Many times people stop at the business part and never come off that high.  From an I.T. employee point of view, I don’t think any of us want to turn into a ‘business guy’ that now is viewed by others as someone who doesn’t understand how I.T. works.  If you never come off that peak – it is hard to relate to other I.T. employees anymore; no matter how many decades ago you started in I.T.
 

It takes the ‘down to earth’ personality to win over others.  A lot of times this is tough for people to achieve.  There is always someone else higher on the food chain.  An I.T. leader eventually answers to the business people.  Those people wind up spending all their time trying to relate (or please) those business leaders and do not spend enough time on the ‘flipside’ relating to I.T. employees.
 

This relating to I.T. employees I talk about is more than just ‘interacting’ with others in the department.  You can be someone in a leadership role and attend all the ‘fun’ events that take place and still not be perceived at someone approachable or who ‘really understands’.  I.T. workers want someone who can relate to what they are doing, someone who can feel their pain, someone who truly understands our problems.  Granted, we all like the ‘fun’ events with all the food and partying – but what we really want is someone who serves as the cheerleader for the average employee.
 

In summary, in the pursuit of advancement inside your department, whether you agree with the statements I have made or not – keep this in mind:  Always be aware of your current surroundings.  Make notes of the things you like and don’t like.  Don’t become someone in the future that creates the very atmosphere you despised years earlier.  An upcoming leader will be able to observe others and learn from the good actions along with the bad.  We take these experiences in to shape ourselves into better future leaders.