|
don't cross-post pls
<< >>
|
|
|
|
|
Hello everyone,
We have a problem with our .net services. On most of the servers where these services are installed, they won't start automatically when the system gets rebooted. We've checked the event logs of these servers, but cannot find any events of these services regarding to why they didn't startup. Restarting the services manually never poses a problem.
These services are installed with the automatic startup option and run under the domain administrator account.
We've tried to add some dependencies but without any luck in letting them start automatically.
Has anyone got some suggestions in which direction we might need to look, they are more then welcome ...
Best regards,
Kristoff
|
|
|
|
|
I usually see this as a dependency issue. Are these services using something in their start up that has not started yet?
Perhaps you need to go through the start-up code of your services and look at what is done in there that might need something else to already be running. You can then set up a dependency between the services to make sure that things are started in the proper order.
At least this is a place to start.
George Carlin wrote:
"Don't sweat the petty things, and don't pet the sweaty things."
Jörgen Sigvardsson wrote:
If the physicists find a universal theory describing the laws of universe, I'm sure the a**hole constant will be an integral part of that theory.
My Blog[^]
|
|
|
|
|
Thx Ray for your reply.
We thought the same thing over here ...
One of our services ( a tcp/ip client with our own protocol ) does the following things at startup:
1. retrieve some settings from the registry
2. retrieve some settings from the database
3. check and/create the needed process folders
4. start a logwriter
5. start the tcp/ip client
So what we tried is to add like services concerning networking like lanmanworkstation ,lanmanserver, ... but wihtout any luck so far ... seems like I need to do some more research then :p
Anyway thx for your reply Ray and if you or anyone else comes up with something let us know, thx
Best regards,
Kristoff
--------
It's time to kick ass and chew bubble gum and I'm all out of gum.
|
|
|
|
|
Hello all,
after hours of banging my head on the wall I hope someone can help me!
I´ve got a network share on my Windows 2003 Server. On this share are tools like NAnt and NDoc.
I like to execute this tools on the server and integrate them in my build process. But I get a security exception.
I have adjusted the .NET Security settings: Created a new CodeGroup and set the condition to local
intranet and the permissionset to FullTrust. On my XP machine this works fine and I can execute the .NET applications directly from the share. But not on the server, there I get a security exception.
What´t the problem??
Thanks a lot!!!
Andy
|
|
|
|
|
Apologies if this is in the wrong forum.
I want to draw text on the screen and and be able to fill in the centers of letters, like kids do when they are bored in school. Essentially I want to achieve what I could achieve if I drew text with a paint program, then used the fill tool to fill in the background, leaving areas completely surrounded by the pixels that make up the letters untouched.
I remember doing this many years ago, back in generic Windows SDK days, but I have no idea how I did it.
Can anyone tell me how to do this?
Thank you,
bruce
|
|
|
|
|
Look at Graphics class. I think you'd wanna use Graphics.FillRegion().
Farhan Noor Qureshi
if (this == this) thow this;
|
|
|
|
|
I looked at this for a while and did some tests, but I can't figure out how FillRegion would get me there.
Any more ideas?
bruce
|
|
|
|
|
Hi,
I've been experimenting building a customer designer to allow users to build simple plug-in forms for our app at runtime.
The designer bit works splendid. However, I'm getting stuck building an event editor. The standard EventsTab implementation hooks up the events OK but what I'm after is the ability to click a button on the events to bring up an editor (to define the event code).
I've tried various attempts at deriving a class from EventsTab and messing with the event descriptor attributes (specifying a custom UITypeEditor etc.) but at best this has no effect and at worst it generally upsets the MS code. The general outcome of overriding the default GetProperties() method on EventsTab seems to be that my EventBindingService dissappears eventually.
Anyone now of any examples of how to do this?
Is it best to build a custom property page from scratch rather than hooking into the EventsTab code?
Thanks,
Rich Bamford.
|
|
|
|
|
Hey,
Does Windows NT4.0 support .Net framework 1.1?
If so what are the prerequisites?
Thanks!
|
|
|
|
|
Yes. The System Requirements page[^] lists:
Windows NT 4.0 Service Pack 6a
Microsoft Internet Explorer 5.01 or later is required
Microsoft Data Access Components 2.6 is required for data scenarios
NT 4.0 Terminal Server is not supported.
Stability. What an interesting concept. -- Chris Maunder
|
|
|
|
|
benqazou wrote: Does Windows NT4.0 support .Net framework 1.1
Yes, with SP6a.
Microsoft wrote: System Requirements
* Supported Operating Systems: Windows 2000; Windows 98; Windows ME; Windows NT; Windows XP
* Microsoft Windows NT 4.0 requires Service Pack 6a.
'A programmer ist just a tool which converts caffeine into code'
|
|
|
|
|
.NET runs on NT4 SP6a. However, I don't think you can host the Visual Studio .NET on NT4. Or, if you can, some web functionality will not be available IIRC.
Kevin
|
|
|
|
|
hi
I want to import a CSV file using .NET
i am using the following line for connection string
conn_csv_str = "Driver={Microsoft Text Driver (*.txt; *.csv)};Dbq=" + Path.GetFullPath("../../UPS");
this works fine and my application si working also, however it ignores the first line, becasue it assumes that its the header. where as i want to read from a csv file that does not have a header, can anyone help me out on this.
Vandan
|
|
|
|
|
You might try the Jet OleDb driver instead. The connection string should look like the following:
conn_csv_str = "Provider=Microsoft.Jet.OleDb.4.0; data source=" + Path.GetFullPath("../../UPS"); Extended Properties=\"Text;HDR=NO; IMEX=1;FMT=Delimited;\""
HDR states whether there is a header or not.
|
|
|
|
|
Hello, I have a .Net executable that calls a .Net Dll, which is strong named and registered in the GAC, so it won't have to be in the same directory as the executable.
I made some changes in the Dll (but didn't alter the methods headers) and recompiled it with the same StrongName Key.
Then I tried to replace the old Dll with the new one, both in the Dll directory and in the GAC. But now the executable doesn't recognize the new Dll!
Any help?
Thanks.
|
|
|
|
|
If you didn't change the version number, GACUTIL won't do anything. It looks at the metadata, says 'yup, I've already got that one' and doesn't overwrite. The same is true for any other methods of updating the GAC (since the work is done by fusion.dll - all methods of updating the GAC wrap fusion.dll).
You need to remove from the GAC first.
Note that you shouldn't issue a new DLL with the same version, culture and strong name key as before. You're supposed to either use Publisher Policy (yuck) or configuration (slightly less yuck) to redirect from the old version to the new, if required.
If the DLL is tied closely to a particular project, you should not use the GAC. IMO, the GAC is overused. You should only use the GAC if you have a shared DLL that you won't control all the clients of. It's generally safer and cleaner to stick to keeping private DLLs - not strongly named - in the same directory as the executable, or in a dll directory below the executable's directory.
You may feel that you need to have certain fixes apply to all clients of your DLL. If you've been able to test the changes to all clients sufficiently, go ahead. But it's generally safer to stick to using the version that the executable was tested with, and update the whole package if a bug is found.
Stability. What an interesting concept. -- Chris Maunder
|
|
|
|
|
Hi. Thanks very much for your advice.
I will describe the steps of what I did in more detail:
1- Compiled the Dll with a strong name key file, and version set to '1.0.*' (what happens here is that the version number changes everytime I compile)
2- Copied the Dll to 'c:\Components\MyComponent.dll'
3- Registered the 'MyComponent.dll' in the GAC
4- In the Client project I added a reference to 'c:\Components\MyComponent.dll'
5- Compiled the Client project and copied it to 'c:\Programs\MyProgram.exe'
(At this point all things work properly. A few days later I needed to do some changes in the Dll component)
6- I changed some lines of code in a method of the Dll project, then recompiled it (thus generating a new version number) with the same strong name key file.
7- Removed the old Dll from the GAC and deleted the file 'c:\Components\MyComponent.dll'.
8- Copied the updated Dll to 'c:\Components\MyComponent.dll' and registered it in the GAC.
(Now, from this point the client program at 'c:\Programs\MyProgram.exe' stopped working, saying it doesn't find the component.)
This component is to be shared between different applications, and I decided to put it in the GAC so I wouldn't have to copy it to all the applications directories everytime there is a change in the component.
I expected it was enough to just update the Dll from the 'c:\Components' directory and the changes would reflect to all applications that use it.
Am I missing something here?
|
|
|
|
|
When a program or DLL references a strongly-named DLL, all components of the DLL's identity are recorded - assembly name, version number, culture, and public key token. Only that version of the DLL will be loaded.
The GAC can contain multiple assemblies with the same base name if they have different version numbers, cultures, and/or public key tokens. That last allows different vendors to use the same base name independently, although this will be pretty confusing to a user!
When loading a strongly-named DLL, the GAC is searched first, then the application's directories. See How the Runtime Locates Assemblies[^] for more.
To cause a different version of an assembly to be loaded from that listed in the referencing assembly's manifest, you have two options: Publisher Policy[^], or <assemblyBinding>[^] elements in the application's configuration file. The former allows all uses of an assembly to be directed to a new version, but the documentation is a bit unclear. The latter has to be applied to every executable that will use the updated assembly.
I'll reiterate that for components that aren't going to be released to third-party developers, I'd recommend keeping your class library assemblies private, without strong naming. If the assembly isn't strongly named you can overwrite without regard to version numbers. It does make deployment slightly more difficult, since you have to copy the updated assembly to the directories of every application you want to use it, but then you should probably only update those applications which are experiencing the problem you've fixed in the newer version of the assembly.
Stability. What an interesting concept. -- Chris Maunder
|
|
|
|
|
client socket based app is in VB6 and server socket based is in VC++.NET.
Prob is i am doing encryption in VB6 and convert that encrypted byte array to string using copy memory.
when I get that data over socket , using UTF7 encoding,it have question marks and result is i cant do decryption on this data. for example VB6 send data like this.
Rê—ì:Eaœ!ƒcêÕ½2A€ ;ŽîOv|¼¤iN<ýåü4=ŠË¹¿•ðµ_ìq¾”&
then at VC++.NET it will b like this
Rê?ì:Ea♂§?!?cêO½¶2A? ;?îOv|¼?☼iN<→yåü4=?∟E1¿?▬?d↔µ←_ìq_?&
Please anyone can solve this problem. yr help wud be appreciated.
thanx
|
|
|
|
|
Strings in VB are basically BSTRs and strings in .NET are managed classes with lots of extra info attached to unicode char array. You could receive the encrypted data in an unmanaged buffer in C++.NET side, decrypt it and then marshal it into the managed side.
Farhan Noor Qureshi
if (this == this) thow this;
|
|
|
|
|
Thanx for yr reply.
on .NET side i am getting data in BYTE so i send this BYTE data as it is to decrypt function.but still problem persists.
is there anyway to workout this buffer BYTE. this is of managed type.
yr reply wud be really appreciated
|
|
|
|
|
Post some code.
I think the problem might be,
1. in using strings
2. in encoding UTF-7
Can you try with base64? http://en.wikipedia.org/wiki/Base64[^]
Farhan Noor Qureshi
if (this == this) thow this;
|
|
|
|
|
I tried base64 with no joy.......... 
|
|
|
|
|
Hello!
Usually I show errors to the user by handling the Application.ThreadException event. This works fine for the button1_Click() method (message box appears) but I do not get a message box from the DragDrop handler (see code below). Why?
public partial class Form1 : Form
{
public Form1() {
InitializeComponent();
textBox2.AllowDrop = true;
Application.ThreadException += OnThreadException;
}
void OnThreadException(object sender, ThreadExceptionEventArgs e) {
MessageBox.Show(e.Exception.Message);
}
private void button1_Click(object sender, EventArgs e) {
throw new ApplicationException("button1_Click");
}
private void textBox1_MouseDown(object sender, MouseEventArgs e) {
DoDragDrop(sender, DragDropEffects.All);
}
private void textBox2_DragEnter(object sender, DragEventArgs e) {
e.Effect = e.AllowedEffect;
}
private void textBox2_DragDrop(object sender, DragEventArgs e) {
MessageBox.Show("I'm here!");
throw new ApplicationException("You'll never see this :-(");
}
}
-----------------
Alex
|
|
|
|