Microsoft Windows 7

reBlog from Kalil: System Admin Tools

Image via Wikipedia

I found this fascinating quote today:

ForeFront Client Security Definition Updates Publisher's DescriptionInstall the latest Windows Defender definition updatesspaceIf you encounter problems while trying to install definition updates for Windows Defender, you can download the latest definition update for computers running a 32-bit (x86-based) version of Windows using this link:Kalil, System Admin Tools, Sep 2009

You should read the whole article.

ReadyBoost with Windows 7's Performance Monitor

There are some tasks where the performance gain will be evident, and there are others where it will not. In other words, it is very difficult to be plugging away at some task on your system and just feel it kick in and say to yourself “Wow, there goes ReadyBoost! What a rush!”
To see ReadyBoost in action, you’re going to have to keep tabs on it with Windows 7’s Performance Monitor. In this edition of the Windows Vista and Windows 7 Report, I’ll show you how to do so.

A quick review

In a nutshell, ReadyBoost allows Windows 7’s SuperFetch cache management technology to use flash-based devices to maintain a copy of the disk cache. Once ReadyBoost is enabled, it essentially keeps tabs on hard disk operations and will go into action reading and delivering files from its copy of the cache only when doing so will boost performance.
For example, during sequential read operations, ReadyBoost will sit back and allow SuperFetch to use the cache on the hard disk since the hard disk can outperform a flash-based drive for these types of read operations. During nonsequential read operations, ReadyBoost will jump in and essentially redirect SuperFetch to use the cache on the flash-based drive since a flash-based drive can outperform a hard disk for these types of read operations.
Furthermore, fast hard disks pushing 7200-RPM or higher will in many cases be able to perform some, but not all, nonsequential read operations faster than a flash-based drive. As such, ReadyBoost won’t provide as significant of a performance gain as it would if your hard disk is running 5400-RPM or lower, such as those typically found in laptops. However, even if you have a fast hard disk, there are situations where ReadyBoost can make a big performance contribution. So don’t write off ReadyBoost just yet.

ReadyBoost does NOT equal RAM

And one more thing that I should point out: ReadyBoost does NOT equal RAM. I’ve heard plenty of people comment “Oh, I have 8GB of RAM in my system, I don’t need ReadyBoost. That’s only for systems with 1GB of RAM.” The truth of the matter is that ReadyBoost is about improving hard disk performance, not enhancing or adding RAM to the system.
Sure, a system with only 1GB of RAM can use all the help it can get, but it is important to keep in mind that with ReadyBoost, the performance gain is coming from improved hard disk performance, not from an addition to RAM.
So, again, even if you have 8GB or more of RAM, there are situations where ReadyBoost can make a big performance contribution. So don’t write it off.
(I admit that in some of my previous posts about ReadyBoost, I have inadvertently contributed to this confusion, and I apologize.)

Configuring Performance Monitor

As you may know, Performance Monitor is a visualization tool that allows you to view performance data or counters, both in real time and from log files. You can view performance data in graph, histogram, and report form. To help you monitor ReadyBoost, Performance Monitor provides a special set of counters under the heading ReadyBoost Cache. Let’s take a closer look.
To launch Performance Monitor, click on the Start button, type perfmon in the Start Search box, and press [Enter]. Once the tool launches, expand the Monitoring Tools branch, if it isn’t already, and click on Performance Monitor. When the graph appears, as shown in Figure A, you’ll see that in its default configuration, Performance Monitor is configured to monitor Processor Time. Since we won’t need this for the task at hand, the first order of business is to click the Red X or Delete Key icon.

Figure A

In its default configuration, Performance Monitor is configured to monitor Processor Time.

Now, click the Green + or Add icon. When you see the Add Counters dialog box, locate and select the ReadyBoost Cache header and click the Add button. When the ReadyBoost Cache header appears in the Added Counters section, as shown in Figure B, click OK.

Figure B

When the ReadyBoost Cache counter appears in the Added Counters section, click OK.

When you add the ReadyBoost Cache header to Performance Monitor, you are actually adding a set of 10 counters to the monitor, as shown in Figure C. These counters are listed and described here:

• Bytes Cached: The total amount of uncompressed data currently stored in the cache.
• Cache Space Used: The total amount of space currently being used by the cache.
• Compression Ratio: The actual size of the data in the cache divided by the uncompressed size of the data in the cache.
• Total Cache Size Bytes: The total amount of space reserved on the device for the cache.
• Cache Reads/Sec: The number of times data is read from the cache per second.
• Cache Read Bytes/Sec: The number of bytes read from the cache per second.
• Skipped Reads/Sec: The number of read operations skipped per second.
• Skipped Read Bytes/Sec: The number of bytes not read from the cache per second.
• Total Reads/Sec: The number of read operations directed to ReadyBoost. (Includes both satisfied and skipped read operations.)
• Total Read Bytes/Sec: The number of bytes in the read operations directed to ReadyBoost. (Includes both satisfied and skipped read operations.)

Figure C

When you add the ReadyBoost Cache header to Performance Monitor, you are actually adding a set of 10 counters to the monitor.

As you can see, having 10 counters on the monitor at one time can be a bit confusing. Furthermore, these 10 counters monitor the full spectrum of ReadyBoost’s activities and not all of them are pertinent when measuring performance gains. As such, you will want to whittle down the counters to just the ones that give you a good view of what is happening.
For example, what I really want to know is how much the ReadyBoost cache is being used to deliver data, so I chose Cache Reads/Sec and Cache Read Bytes/Sec. These two counters will help you gauge the amount of read activity occurring in the cache and will let you know how much data is being delivered from the cache. To balance this view I also want to know how often ReadyBoost is deferring to the hard disk for reading data from the cache. So, I also chose Skipped Reads/Sec and Skipped Read Bytes/Sec.
I then initiated several large file copy operations involving multiple gigabytes of data and let the monitor run in the background. (I’m using a Western Digital 500GB SATA drive that pushes 7200 RPM.) When I returned to Performance Monitor, I discovered that there was a lot of activity displayed on the graph, as shown in Figure D.

Figure D

After several large file copy operations, the Performance Monitor graph displayed a lot of activity.

If you look closely, you’ll see that the amount of data being read from the ReadyBoost cache is roughly equal to the amount of data being read from the hard disk cache. However, this is but a snapshot of ReadyBoost in action under a specific operation. To really gauge how much and under what circumstances ReadyBoost is actually benefiting your system, you’ll want to configure Performance Monitor to create a log file over a period of time. For instance, you might want to log ReadyBoost activity over a period of a couple hours during which you perform the types of tasks that you regularly use your system to do.

Creating a Data Collector Set

If you want to log ReadyBoost activity over a period of time, you’ll need to create a Data Collector Set. Doing so is pretty easy once you know the steps.
To begin, expand Data Collector Sets branch in the Performance Monitor navigation pane. Then, right-click User Defined, select New, and click Data Collector Set. When the Create New Data Collector Set wizard appears, type a name for your Data Collector Set and select the Create Manually option button, as shown in Figure E.

Figure E

This wizard will walk you through the steps of creating a Data Collector Set.

You’ll then select the Performance Counter check box and on the next page use the Add button to access the Add Counters dialog box where you can individually select the four pertinent counters. These steps are condensed in Figure F.

Figure F

You’ll work through the next two steps in the wizard to add the four pertinent counters to your ReadyBoost Monitor Data Collector Set.

Even though there are other settings, click Finish at this point to complete the wizard.
When you return to the Performance Monitor window, you’ll see your ReadyBoost Monitor Data Collector Set. When you are ready to begin logging ReadyBoost activity, just click the green Play icon, as shown in Figure G.

Figure G

To begin logging ReadyBoost activity, just click the Green Play icon.

When you have used your system for what you would consider a substantial amount of time, you can open you Data Collector Set and view the log. Click the Stop icon, and then select Performance Monitor in the navigation pane. Now, click the View Log Data icon and when the Performance Monitor Properties dialog box appears, click the Add button. You can then locate and select your log file, as shown in Figure H.

Figure H

Click the Add button to locate and select your ReadyBoost Monitor log file.

When you click OK, you’ll see your log file and can determine how ReadyBoost behaved during your test period. During my 15-minute sample test, you can see in Figure I that ReadyBoost made a very minimal contribution to the tasks that I was performing.

Figure I

During this particular test, ReadyBoost made a very minimal contribution.

Remember that you will most likely want to conduct multiple tests under various usage circumstances in order to get a real handle on how much ReadyBoost benefits your particular system. As always, keep in mind that your mileage may vary.

OpenSuse's KDE/Firefox Integration

Just a quick nod to the OpenSuse team, which looks to have a great looking KDE desktop for their new 11.2 Release. Congratulations!

Now, about those changes to make Firefox integrate better with KDE — you'll be releasing those upstream to KDE for the inclusion to all distros, right? :D

=-=-=-=-=
Powered by Bilbo Blogger

E-mail Attachment | Ubuntu Command Line | Howto

Here is a very usefull command on sending files as attachement on Linux/Unix command line, this is specifically very usefull when using it in script to send daily report, or just for mailing client certificate directly from the server, there are two ways I know that you can use to send a file as attchemnt using command line, one is using uuencode utility with the "mail" command and the other one using "mutt", I find using mutt command to be much more reliable compared to using uuencodeutility with mail command.

uuencode utility writes an encoded version of the named input file, or standard input if no file is specified, to standard output. The output is encoded using the algorithm described in the STDOUT section and includes the file access permission bits of the input file and the decode_pathname, for re-creation of the file on another system that conforms to this specification.

If you don't have uuencode command on your system, and your running Cent OS or Redhat distribution you can find the uuencode and uudecode on the “sharutils" package.

To send a file attachment using "mail" command
$ uuencode filename.jpg filename.jpg | mail -s 'subject' recipient@email.com

To send a file attachemnt using "mutt" command
$ echo | mutt -a filename.jpg -s 'subject' recipient@email.com

Mutt is a small but very powerful text based program forreading electronic mail under unix operating systems,including support color terminals, MIME, and a threadedsorting mode.

How to Monitor RAID Devices on Linux

Here is a simple howto on using mdadm to alert you of failing array Linux Software RAID. this is important becasue failed or failing devices needs to be replaces as diffrent RAID level have different capabilities when it comes to dealing with disk failure.

Lets us try to send a test e-mail to make sure that mail is working on your server
$ mdadm --monitor --scan --mail= YourEmailAddress /dev/md0 -t -1

If your mail is working you will receive a test message from your server

Now we set up the monitoring, to check the RAID array every 15 minutes using cron do "crontab -e" to edit cron
*/15 * * * * /sbin/mdadm --monitor --scan --mail= YourEmailAddressHere /dev/md0 -1

Note: /dev/md0 is the RAID array you want to monitor

KDE 4, please integrate 10 Administrative tools!

I am a big fan of the KDE desktop but their current KDE 4.3.2 release is still missing tools that are required for a 'complete desktop experience', specifically for system administration tasks with a Graphical User Interface.

Reading around the web, there are many requests and suggestions about ways to improve the KDE Desktop. It is very important to note that KDE 3.5.x was years in development to get to the functionality and stability that the 3.5.x release had; KDE 4.3.2 has only been in development for a bit over two years and its capabilities are already very impressive.

KDE 4.3.2 may have an impressive list of capabilities, but the desktop isn't yet complete for those Point-and-Click types. Here is my wish list for the near future of KDE development, to allow full system administration within the GUI of KDE.

1. KDESudo access to saving files

I would love to be prompted for a sudo password when trying to edit a file which I have no write access to, such as if a desktop computer is going to be used for an application like Motion and the user doesn't know various command-line editors. Browsing to /etc/ and opening a file with Kate (or KWrite, or any app) should allow for read-only access, and prompt for a password if the user wants to edit and save that file.

2. Administrative Mode in System Settings

Various parts of System Settings require that the root user makes changes, such as for user auto-logins and for adding printers. I am not sure the status of this. Launching 'kdesudo systemsettings' allows a user to get around this, for now.

3. Printer setup tools

In Debian at least, printers can be added in KDE by launching 'kdesudo systemsettings' and then using the Printer Configuration module, assuming the user has installed the package 'system-config-printer-kde'. Access to this should be tied in with point number two, above. Also, any machine with CUPS (which is likely any Linux machine that prints) can set up printers in http://localhost:631 — but we all knew that one by heart, right?

4. User Account Tools

Adding users to a multi-user system like Linux is important, yet it is missing in KDE's System Settings. The application 'kuser' can be installed and launched with 'kdesudo kuser' but it isn't integrated into the System Settings as far as I can tell.

5. Services and Daemons

There should be an application to start/stop/setup system daemons that happen at boot; there is an addon located here which looks to do the job on Kubuntu, but this should be a standard module within System Settings, to allow a user to stop MySQL or to change when SSH is launched, etc.

6. FUSE File System Setups

The FUSE system is great; it allows users to mount a remote file system on their current setup, without needing root access that a regular filesystem mount needs.

For example, at home my Laptop gets its wireless signal and connects to my big computer's music collection, providing transparent streaming of my music to the laptop without having all of those files on the laptop itself. For someone familiar with the command line and mounting filesystems this is pretty easy; for a computer (or even Linux) newbie it isn't so clear what needs to be done.

I would really like a System Settings module which allows for certain file systems to be mounted at login depending on various conditions. For example, if my wireless access point is 'lefty@home' then my sshfs:/ and fusesmb:/ file systems should mount; if my access point is "corporation-WEP" when I am at work, then FUSE file systems which I have defined for my work needs should mount, etc.

7. Package installation integration

Using some software requires that additional packages are installed to make that first app work properly in certain situations. If I want to open a .doc file, for example, I would need OpenOffice or KOffice or some other application which can read the closed, binary Microsoft document format. If I try to open a .doc file and I have no applicable software available, KDE should offer a list of options and install the needed software (and not limit my options to KDE applications in the process).

8. Graphics setup configuration

Adding a second monitor to my home computer required setting up an XOrg config file (because Xorg doesn't use those by default any longer) and then defining a Virtual Desktop size that would be equal to, or larger than, the total resolution of my two monitors added together. I had to add this to my Screen section of my /etc/X11/xorg./conf file (completely outside of KDE), and the whole process isn't very intuitive:

SubSection "Display"

Viewport 0 0

Depth 24

Virtual 3048 2048

EndSubSection

KDE would do well to extend the capabilities of KRandR to allow these setups to work with minimal hassle to the end user, as well as to insert various modules into that xorg.conf file to allow for 3D rendering etc, regardless of the driver used (assuming the driver is capable of this with some module modifications). KRandR sits in the Tray of your plasma task bar, but it should be accessible within System Settings. There is a Disp[lay section with information about Multiple Monitors but it doesn't do anything to assist in the setup.

9. Hardware information

In KDE3 I could see my hardware, capabilities, and even the serial number of some of the hardware peripherals installed onto my computer. KDE4 could do this and so much more, but currently the Hardware module in System Settings only discusses software backends for HAL Power Management, for Networking, and for Bluetooth devices. Where is the information about my RAM? About my processor speed? This information is all inside KInfoCenter but there isn't a way to access that from System Settings. Granted, this is information and not something with settings, but it should still be within System Settings.

10. KDE4 Technologies need clarification

Akonadi? Nepomuk? If you're not a KDE user or developer these names mean nothing to you; as a KDE user I am still unsure what these do. They need more descriptive naming in the System Settings to describe what it is I am looking at. Don't call it Anokadi in system Settings; call it Anokadi Data Storage or whatever it does. Even Aaron Seigo, the KDE hacker, puts it this way in a recent post

Jargon Is Bad:

There's a lot more jargon in KDE, though: nepomuk (search service!), krandr (screen settings!), kwin compositing (desktop effects!), akonadi ... If we can keep the jargon out of what we see when using the software, it will help people immensely.

Notice even Aaron didn't tell us what Akonadi is? :D

All and all, I love KDE, its capabilities, its sensible defaults, its configurablity, its beauty, its speed... I just hope to see KDE 4 allow users to control their whole desktop better from an administrative perspective. Give it time, it will happen. The best part about KDE is that it keeps improving, by leaps and bounds.

Any Graphical Administration suggestions that readers can provide? Please, leave a comment, and enjoy your Free Software.

=-=-=-=-=
Powered by Bilbo Blogger

HowTo: VirtualBox on Debian Squeeze (Testing / Sid)

VirtualBox is the great, Free Software virtual machine software which compares to VMware and others. Unlike the closed-source competitors, VirtualBox has the needed source code 'kernel headers' which allow it to hook into your Linux kernel and do its networking setup, etc.

With a rolling-release distribution (which makes new software available as it gets released) such as Debian Testing or Debian Sid ('Unstable'), new kernels mean that the older apps and drivers which hook into the kernel no longer work. For example, if you have run a working VirtualBox virtual machine in the past but get errors like the following image, there is a good chance that the kernel has changed and you need to recompile the modules (drivers):

Debian has a great tool called "Module Assistant" which easily compiles needed modules (drivers) for various applications, including VirtualBox and some wireless network drivers.

To run VirutalBox successfully, make sure that you have 'module-assistant' and 'virtualbox-ose' installed, and optionally 'virtualbox-ose-qt' if you prefer a more KDE-like graphical interface.

Once those are installed you'll need to work at a command line. As root (or as a regular user with 'sudo' capabilities) run at the command prompt:

shell$ sudo m-a prepare

to get Module Assistant prepared with whatever it needs; likely it will download some kernel headers if you've not used Module Assistant before. Then we're going to 'auto install' (a-i) VirtualBox with Module Assistant:

shell$ sudo m-a a-i virtualbox-ose-source

You'll see some screens like this as Module Assistant is doing its work:

Once that is done, we'll need to load (modprobe) the kernel modules (drivers) that VirtualBox uses; you may want to add these modules to the file at /etc/modules so that they auto load at boot time. Note that because I had the older kernel modules loaded, I am going to remove (-r) them first, and then load in the newly built modules:

shell$ sudo modprobe -r vboxdrv
shell$ sudo modprobe -r vboxnetflt
shell$ sudo modprobe vboxdrv
shell$ sudo modprobe vboxnetflt

In this screenshot, we're getting warnings about an unrelated config file so we're going to ignore it as you may not have these issues on your own setup.

Now, VirtualBox is using the needed kernel modules for its current kernel and VirtualBox, and I am able to load my virtual machines again.

Happy VM'ing!

An open request to rename Ubuntu package formats

I have a .deb file, where can I install it?

If the software package was compiled for Debian, I can install it on nearly any recent Debian release. Currently I have XTightVNCViewer installed with the Debian package release 1.2.9-21 (from the Stable branch) even though I run a Sid desktop, because this older package has functionality which isn't available in the Sid package at version 1.3.9-5. Other than some dependency resolutions, mixing Debian software branches (Stable, Testing, and Unstable) isn't recommended but it isn't difficult to do, and my experiences with it have been a rather stable setup when it needs to be done. (Let me repeat, this is not recommended nor supported). Debian releases infrequently enough that this mixing probably isn't needed much; if a package is still relevant there will be a new release for it.

Ubuntu came onto the scene a few years ago with the goal to release Debian Sid (Unstable) as a stabilized set of packages. Not to start a flame war, but Ubuntu's success in this has been questionable, depending on who you ask or your own experiences.

The repackaging of Debian was at first welcomed by many in the Debian community, but early into the project the Ubuntu developers forked and changed the way Debian works for their own distribution. This is Free Software and there is nothing wrong with that approach, although there has been much said about Ubuntu's contributions upstream.

With that split from Debian, however, came no split with the package manager or its tools. Ubuntu now requires separate package releases for each OS release; a quick stop at GetDeb.net shows that you have to search for your specific version of Ubuntu to find a package which will work for your setup. These aren't repositories to keep your software up-to-date like Debian is designed for; these are downloadable packages to be installed much like a Windows .exe file would be installed: download and double-click.

When SuSE Linux took the Red Hat package format, RPM, it led to the increase of Dependency Hell for RPM-based distributions. Red Hat, Suse, and Fedora all use RPM files but they're incompatible; without knowing which release and distribution a package was created for, that package is nearly useless and attempting to install it will lead to a failure or massive headaches when trying to resolve its dependencies. O perhaps it will just luckily work, all depending on the software and its complexity.

Ubuntu is doing the same to the .deb package format. Their frequent OS releases means that every package needs to be recompiled for that release. Applications downloaded from GetDeb.net or one of the many other Ubuntu-focused software sites may no longer work when the OS is upgraded; without a repository there is no way to upgrade the 3rd party packages. (Note: These sites provide a valuable service to the Ubuntu community, but seem to disregard Debian completely; the purpose of this Open Request is not to stop or deride these sites at all.)

Debian, on the other hand, has always used .deb files and has encouraged using software repositories for accessing new software and for keeping it up-to-date. While mixing Debian released packages from one OS release to the next isn't recommended or supported, it does work. Downloading individual .deb files for installation works but it isn't the recommended method. Perhaps mixing Ubuntu packages (from one release to the next) would work also, I am not sure.

But the real problem here is that Ubuntu is using the .deb package format and giving it both a bad name (due to dependency breakages), and its packages are named .deb but don't work on Debian, the very system that the packages were designed for.

With Ubuntu's frequent release cycle and with the nature of Free Software, the package management tools that Ubuntu uses (dpkg, apt, aptitude) should be modified to accept packages in a Ubuntu-specific package extension: .ubu files. This could happen at any release; perhaps after this next release it could be worked on, to be ready in time for the next LTR or even the next 'short term' release in order to give time for testing.

While the packages themselves may be internally identical to a .deb file, these .ubu files would specify that the packages are for Ubuntu and not for Debian. SuSE should have done that and didn't; Ubuntu doesn't have to make this same mistake. Debian users wouldn't have to be confused trying to know if a .deb file will work on their system or not. Ubuntu would get the great package management that Debian has but with better branding of their packages and software. Debian would have its branding as well, and perhaps both Debian and Ubuntu could still install one another's packages with unknown results, just as it is now.

The new .ubu files wouldn't impart any release info in their file names as they are; either the name itself could be 'packagename.904.ubu' or maybe the extension should specify the release, for example 'packagename.u904'. I don't have the answers to this, but naming separate distribution packages the same isn't helpful at all.

Ubuntu may be here to stay, but please, keep Debian's package extension .deb for Debian.

=-=-=-=-=
Powered by Bilbo Blogger

Technical Interview Questions – Networking

• What is an IP address?
• What is a subnet mask?
• What is ARP?
• What is ARP Cache Poisoning?
• What is the ANDing process?
• What is a default gateway? What happens if I don't have one?
• Can a workstation computer be configured to browse the Internet and yet NOT have a default gateway?
• What is a subnet?
• What is APIPA?
• What is an RFC? Name a few if possible (not necessarily the numbers, just the ideas behind them)
• What is RFC 1918?
• What is CIDR?
• You have the following Network ID: 192.115.103.64/27. What is the IP range for your network?
• You have the following Network ID: 131.112.0.0. You need at least 500 hosts per network. How many networks can you create? What subnet mask will you use?
• You need to view at network traffic. What will you use? Name a few tools
• How do I know the path that a packet takes to the destination?
• What does the ping 192.168.0.1 -l 1000 -n 100 command do?
• What is DHCP? What are the benefits and drawbacks of using it?
• Describe the steps taken by the client and DHCP server in order to obtain an IP address.
• What is the DHCPNACK and when do I get one? Name 2 scenarios.
• What ports are used by DHCP and the DHCP clients?
• Describe the process of installing a DHCP server in an AD infrastructure.
• What is DHCPINFORM?
• Describe the integration between DHCP and DNS.
• What options in DHCP do you regularly use for an MS network?
• What are User Classes and Vendor Classes in DHCP?
• How do I configure a client machine to use a specific User Class?
• What is the BOOTP protocol used for, where might you find it in Windows network infrastructure?
• DNS zones – describe the differences between the 4 types.
• DNS record types – describe the most important ones.
• Describe the process of working with an external domain name
• Describe the importance of DNS to AD.
• Describe a few methods of finding an MX record for a remote domain on the Internet.
• What does "Disable Recursion" in DNS mean?
• What could cause the Forwarders and Root Hints to be grayed out?
• What is a "Single Label domain name" and what sort of issues can it cause?
• What is the "in-addr.arpa" zone used for?
• What are the requirements from DNS to support AD?
• How do you manually create SRV records in DNS?
• Name 3 benefits of using AD-integrated zones.
• What are the benefits of using Windows 2003 DNS when using AD-integrated zones?
• You installed a new AD domain and the new (and first) DC has not registered its SRV records in DNS. Name a few possible causes.
• What are the benefits and scenarios of using Stub zones?
• What are the benefits and scenarios of using Conditional Forwarding?
• What are the differences between Windows Clustering, Network Load Balancing and Round Robin, and scenarios for each use?
• How do I work with the Host name cache on a client computer?
• How do I clear the DNS cache on the DNS server?
• What is the 224.0.1.24 address used for?
• What is WINS and when do we use it?
• Can you have a Microsoft-based network without any WINS server on it? What are the "considerations" regarding not using WINS?
• Describe the differences between WINS push and pull replications.
• What is the difference between tombstoning a WINS record and simply deleting it?
• Name the NetBIOS names you might expect from a Windows 2003 DC that is registered in WINS.
• Describe the role of the routing table on a host and on a router.
• What are routing protocols? Why do we need them? Name a few.
• What are router interfaces? What types can they be?
• In Windows 2003 routing, what are the interface filters?
• What is NAT?
• What is the real difference between NAT and PAT?
• How do you configure NAT on Windows 2003?
• How do you allow inbound traffic for specific hosts on Windows 2003 NAT?
• What is VPN? What types of VPN does Windows 2000 and beyond work with natively?
• What is IAS? In what scenarios do we use it?
• What's the difference between Mixed mode and Native mode in AD when dealing with RRAS?
• What is the "RAS and IAS" group in AD?
• What are Conditions and Profile in RRAS Policies?
• What types or authentication can a Windows 2003 based RRAS work with?
• How does SSL work?
• How does IPSec work?
• How do I deploy IPSec for a large number of computers?
• What types of authentication can IPSec use?
• What is PFS (Perfect Forward Secrecy) in IPSec?
• How do I monitor IPSec?
• Looking at IPSec-encrypted traffic with a sniffer. What packet types do I see?
• What can you do with NETSH?
• How do I look at the open ports on my machine?

The next article in this series will cover Active Directory questions!