joeware - never stop exploring... :)

I had to work out a few things to do in a batch file the last couple of days so I thought I would share. Note: so this doesn’t become a huge powershell versus perl versus whatever discussion, I had no choice, I had batch available period and even that was possibly stretching it a little as the people involved seemed very concerned about scripts, etc. Also, yes I am aware of Tim–Toadie (TIMTOWTDI). So please feel free to comment with additional methods.

Getting date and time in a useful format

for /f "tokens=1-4 Delims=/ " %%i in (‘date /t’) do  set dt=%%l/%%j/%%k
for /f "tokens=1" %%i in (‘time /t’) do set tm=%%i
echo Date: %dt%-%tm%

Is this machine virtual or physical? It could be VMWare, HyperV or Virtual Server or Virtual PC. I don’t have the strings for other virtualization techs, if people are using them and can post the results of the command “wmic computersystem get model” then they can be added to the batch commands.

set VIRTUAL=NO
set VMWARE=NO
set HYPERV=NO
for /f "skip=2 tokens=1-2 delims=," %%i in (‘wmic computersystem get model /format:csv’) do set MODEL=%%j

: Microsoft Virtualization
if "%MODEL%"=="Virtual Machine" (
  set VIRTUAL=YES
  set HYPERV=YES
)

: VMWare Virtualization
if "%MODEL%"=="VMware Virtual Platform" (
  set VIRTUAL=YES
  set VMWARE=YES
)

Is this machine Vista or better?

SET VISTA+=NO
if exist %systemroot%\system32\bcdedit.exe set VISTA+=YES

Is this machine x64?

SET x64=NO
if exist "%systemdrive%\program files (x86)" set x64=YES

So I am forced to write some more .NET code, worse, some Windows Forms stuff. Couldn’t put me further away from my happy zone. I do have to say that Borland err, CodeGear Builder C++ still blows VS away for simple generation of GUI. If only CodeGear did x64…

So two things that I have found I am unhappy about in just the last hour.

1. RefreshCache() doesn’t appear to work for the LDAP provider in DirectoryEntry. WTF.
2. CTRL-A isn’t captured and used for SELECT ALL by default in a TextBox. I actually had to code a workaround for it with the following C# Code:

private void textBox1_KeyPress(object sender, KeyPressEventArgs e)
 {
  // CTRL_A Select All not handled by default
  if (Convert.ToInt(e.KeyChar) == 1) textBox1.SelectAll();
 }

If you are running Internet Explorer 6 SP1, it is probably a good time to upgrade. 🙂

http://portableapps.com

I’m sure you have had that question before or seen it before. It is a popular question. And of course as everyone knows or at least finds out (possibly after much painful troubleshooting) is that it is the DHCP Client Service that registers your dynamic DNS updates. And when you learn that you say, but my servers aren’t using DHCP so that can’t be right… Doesn’t matter, the DHCP Client is registering the names in DNS for you… So don’t shut it off unless you don’t want your servers magically registering themselves… 

That is what we have been learning and possibly relearning and recalling every time that question has been asked since Windows 2000 first went RTM. You may have forgotten it after you first heard it in class or from a casual conversation or saw it on an internet post but the first time you can’t resolve a server to name and track it down to the fact that someone disabled the DHCP Client Service because “the server wasn’t and wouldn’t ever use DHCP” you will not forget again. Right?

Well get ready to forget that piece of trivia. Microsoft was sneaky and changed the DNS Registration functionality from the DHCP Client Service to the DNS Client Service… WHOA! This change from what I understand started in Windows Vista. I don’t do much with client OSes so never ran into it. But I can absolutely confirm that this functionality is in Windows Server 2008. I glanced through the Source Code for Windows Server 2008 for the dynamic updates and performed about 30 minutes of testing with WireShark and it absolutely is moved into the DNS Client Service.

Now I admit, I may be a little behind the curve and some of you are almost certainly going to say, whoa joe… you ARE behind the curve… But I felt I would share this info because I just learned it and when I looked around I saw some confusing internet posts on the topic. 

One MSFT document I saw said that sometimes the DNS Client Service does the registration, sometimes the DHCP Client Service does the registration. When I looked at the source code, that didn’t seem to be the case, also my testing didn’t seem to bear that out. Even if I was using DHCP for the server, if I disabled the DNS Client Service and, obviously, had the DHCP Client running fine, it wasn’t registering its name at all. IPCONFIG/REGISTERDNS absolutely used the DNS Client Service as well.

I updated the Wikipedia article on Microsoft DNS to reflect this change in functionality.

    joe