WaitNamedPipe() Issue

Hi...

I'm trying to send a message to a server using a named pipe. I managed to send a message from the client application to the server application in the same machine. But when i try sending the message with the server application in another machine it returns false in the WaitNamedPipe() function :(

I have turned off both machines firewall, i can ping both machines and the ip address of the server is correct.

What can be the problem?
Here is the code for my Cliente and Server App:

CLIENT

Code:

---

int main(int argc, char* argv[])
{
    HANDLE hPipe;
    DWORD dwReadWritten;
        DWORD dwMode=PIPE_READMODE_MESSAGE|PIPE_WAIT;
        char buffer[1024];
    char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

    if (!WaitNamedPipe(pszPipe, NMPWAIT_WAIT_FOREVER))
        {
                printf("CLIENT: Error connecting\n");
                return 1;
        }
        printf("CLIENT: Connect to server\n");
    hPipe = CreateFile(pszPipe, GENERIC_WRITE|GENERIC_READ, 0, NULL,
                      OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);
        if (hPipe==NULL)
        {
                printf("CLIENT: Error a connecting \n");
                return 1;
        }
        if (!SetNamedPipeHandleState(hPipe,&dwMode,NULL,NULL))
        {
                printf("CLIENT: Error SetNamedPipeHandleState\n");
                return 1;
        }
        while (1)
        {
                printf("Frase: ");
                gets_s(buffer,1024);
                buffer[lstrlen(buffer)]=0;
                if (!strncmp(buffer,"end",3))
                        break;
                /*if (!WriteFile(hPipe, buffer, strlen(buffer)+1,
                                                                          &dwReadWritten, NULL)) {
                        printf("CLIENT: Error WriteFile\n");
                        return 1;
                }
                if (!ReadFile(hPipe, buffer, 1024,
                                                                          &dwReadWritten, NULL)) {
                        printf("CLIENT: Error  ReadFile%d\n",GetLastError());
                        return 1;
                }*/
                if (!TransactNamedPipe(hPipe, buffer, lstrlen(buffer)+1,
                                                                          buffer, 1024,
                                                                          &dwReadWritten, NULL)) {
                        printf("CLIENT: Error WriteFile\n");
                        return 1;
                }
                printf("\nResponse: %s\n",buffer);
        }
        printf("CLIENT: turn off pipe\n");
        CloseHandle(hPipe);
        return 0;
}

---

SERVER

Code:

---

void main( )
{
    SERVICE_TABLE_ENTRY  DispatchTable[] =
    {
        { NOME_DO_SERVICO, MyServiceStart      },
        { NULL,              NULL          }
    };
 
    if (!StartServiceCtrlDispatcher( DispatchTable))
    {
        ShowError(" [ServNT] StartServiceCtrlDispatcher error = %d\n", GetLastError());
    }
}

void WINAPI MyServiceStart (DWORD argc, LPTSTR *argv)
{
        HANDLE hPipe;
        OVERLAPPED ovl;

    MyServiceStatus.dwServiceType        = SERVICE_WIN32;
    MyServiceStatus.dwCurrentState      = SERVICE_START_PENDING;
    MyServiceStatus.dwControlsAccepted  = 0;
        MyServiceStatus.dwWin32ExitCode      = 0;
    MyServiceStatus.dwServiceSpecificExitCode = 0;
    MyServiceStatus.dwCheckPoint        = 0;
    MyServiceStatus.dwWaitHint          = 0;
 
    MyServiceStatusHandle = RegisterServiceCtrlHandler(NOME_DO_SERVICO, MyServiceCtrlHandler);
 
    if ( MyServiceStatusHandle == (SERVICE_STATUS_HANDLE) 0 )
    {
        MostrarErro(" [ServNT] RegisterServiceCtrlHandler failed %d\n", GetLastError());
        return;
    }
 
    if ( !SetServiceStatus (MyServiceStatusHandle, &MyServiceStatus) )
    {
        status = GetLastError();
        MostrarErro(" [ServNT] SetServiceStatus error %ld\n", status);
    }

    // Initialization code goes here.
        Event=CreateEvent(NULL,FALSE,FALSE,NULL); //semaforo binário

        if (Event==NULL)
    {
        MyServiceStatus.dwCurrentState      = SERVICE_STOPPED;
        MyServiceStatus.dwCheckPoint        = 0;
        MyServiceStatus.dwWaitHint          = 0;
        MyServiceStatus.dwWin32ExitCode      = ERROR_SERVICE_SPECIFIC_ERROR;
        MyServiceStatus.dwServiceSpecificExitCode = 0x1234;
 
        SetServiceStatus (MyServiceStatusHandle, &MyServiceStatus);
        return;
    }
 
    // Initialization complete - report running status.
    MyServiceStatus.dwCurrentState      = SERVICE_RUNNING;
    MyServiceStatus.dwCheckPoint        = 0;
    MyServiceStatus.dwWaitHint          = 0;
    MyServiceStatus.dwControlsAccepted  = SERVICE_ACCEPT_STOP/* | SERVICE_ACCEPT_PAUSE_CONTINUE*/;
 
    if (!SetServiceStatus (MyServiceStatusHandle, &MyServiceStatus))
    {
        status = GetLastError();
        ShowError(" [ServNT] SetServiceStatus error %ld\n",status);
    }
 
    // This is where the service does its work.
    ShowError(" [ServNT] Waiting for clients \n",0);
        flag=TRUE;

        while(flag)
        {
                hPipe = CreateNamedPipe("\\\\.\\pipe\\ServNT",
                          PIPE_ACCESS_DUPLEX | FILE_FLAG_OVERLAPPED | FILE_FLAG_WRITE_THROUGH, PIPE_WAIT |
                          PIPE_TYPE_MESSAGE | PIPE_READMODE_MESSAGE, PIPE_UNLIMITED_INSTANCES, 1024, 1024, 1000, NULL);

                if (!hPipe)
                        break;

                ovl.hEvent=Event;
                ConnectNamedPipe(hPipe, &ovl);

                WaitForSingleObject(Event, INFINITE);

                if (!flag)
                        break;

                CreateThread(NULL, 0, MyServiceProc, hPipe, 0, NULL);
        }
        CloseHandle(hPipe);
        CloseHandle(Event);

    MyServiceStatus.dwCurrentState      = SERVICE_STOPPED;
    MyServiceStatus.dwCheckPoint        = 0;
    MyServiceStatus.dwWaitHint          = 0;
 
    if (!SetServiceStatus (MyServiceStatusHandle, &MyServiceStatus))
    {
        status = GetLastError();
        ShowError("[ServNT] SetServiceStatus error %ld\n", status);
    }
 
    return;
}
 
VOID WINAPI MyServiceCtrlHandler (DWORD Opcode)
{
    switch(Opcode)
    {
        case SERVICE_CONTROL_PAUSE:
        // Do whatever it takes to pause here.
//            MyServiceStatus.dwCurrentState = SERVICE_PAUSED;
            break;
 
        case SERVICE_CONTROL_CONTINUE:
        // Do whatever it takes to continue here.
//            MyServiceStatus.dwCurrentState = SERVICE_RUNNING;
            break;
 
        case SERVICE_CONTROL_STOP:
        // Do whatever it takes to stop here.
            MyServiceStatus.dwWin32ExitCode = 0;
            MyServiceStatus.dwCurrentState  = SERVICE_STOP_PENDING;
            MyServiceStatus.dwCheckPoint    = 0;
            MyServiceStatus.dwWaitHint      = 0;
 
            if (!SetServiceStatus (MyServiceStatusHandle,
                &MyServiceStatus))
            {
                status = GetLastError();
                MostrarErro(" [ServNT] SetServiceStatus error %ld\n",status);
            }
                        flag=FALSE;
                        SetEvent(Event);
            ShowError(" [ServNT] Leaving MyService \n",0);
            return;
 
        case SERVICE_CONTROL_INTERROGATE:
        // Fall through to send current status.
            break;
 
        default:
            ShowError(" [ServNT] Unrecognized opcode %ld\n",
                Opcode);
                        break;
    }
 
    // Send current status.
    if (!SetServiceStatus (MyServiceStatusHandle,  &MyServiceStatus))
    {
        status = GetLastError();
        ShowError(" [ServNT] SetServiceStatus error %ld\n",status);
    }
    return;
}
 

DWORD  WINAPI MyServiceProc(LPVOID param) //alterar para o nosso codigo/serviço
{
        char buffer[1024];
        HANDLE hPipe=(HANDLE) param;
    DWORD dwReadWrite;
    BOOL ret = FALSE;

        ShowError("Vou entrar na Thread\n",0);
        while(flag)
        {
                buffer[0]=0;
                ret = ReadFile(hPipe, buffer, 1024, &dwReadWrite, NULL);

                if (!ret || !dwReadWrite)
                {
                status = GetLastError();
                MostrarErro(" [ServNT] ReadFile error %ld\n", status);
                        break;
                }

                lstrcat(buffer, "...Ok!!!");

                ShowError(buffer,0);

                ret = WriteFile(hPipe, buffer, lstrlen(buffer)+1, &dwReadWrite, NULL);

                if (!ret || !dwReadWrite)
                {
                status = GetLastError();
                ShowError(" [ServNT] WriteFile error %ld\n", status);
                        break;
                }

        }

        DisconnectNamedPipe(hPipe);
        CloseHandle(hPipe);
        ShowError(" [ServNT] Getting out of Thread\n",0);

        return 0L;
}

HANDLE hEventSource;

void ShowError(LPSTR str,DWORD val)
{
        char buf[1024];
        LPSTR msg[2]={buf,NULL};
       
        wsprintf(buf,str,val);

    if (!hEventSource)
        {
        hEventSource = RegisterEventSource(NULL,            // local machine
                                          NOME_DO_SERVICO); // source name
    }

    if (hEventSource)
        {
        ReportEvent(hEventSource,
                      EVENTLOG_INFORMATION_TYPE,
                      0,
                      0,
                      NULL,  // sid
                      1,
                      0,
                      (const char **)msg,
                      NULL);
    }

}

---

You may have an authentication issue with the remote computer using the target ip address in this manner:

Code:

---

char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

---

Change the Ip address to the computer name of the target computer:

Code:

---

char *pszPipe="\\\\TargetComp\\pipe\\ServNT";

---

hi BobS0327, thanks for the reply :)

i changed the the ip of the target machine to it's computer name, and it didn't work either :(

Code:

---

char *pszPipe="\\\\TSCII03CLT1\\pipe\\ServNT";

---

What could be the problem?

My gut feeling is that it is some sort of security issue on the remote computer.

Hopefully, you'll humor me a little bit and add the following function to your code. Then call DisplayError immediately after you encounter the error. This will display a detailed message about the error. Thus, we won't have to use the "pulling at straws" method to determine the source of your problem. Post the detailed error message on the forum.

Code:

---

VOID DisplayError(CHAR* pMessage)
{
        DWORD eNum;
        CHAR sysMsg[256] = {0};
        CHAR* p;

        eNum = GetLastError( );
        FormatMessage( FORMAT_MESSAGE_FROM_SYSTEM | FORMAT_MESSAGE_IGNORE_INSERTS,
                NULL, eNum,
                MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT), // Default language
                sysMsg, 256, NULL );
        // Trim the end of the line and terminate it with a null
        p = sysMsg;
        while( ( *p > 31 ) || ( *p == 9 ) )
                ++p;
        do { *p-- = 0; } while( ( p >= sysMsg ) &&
                ( ( *p == '.' ) || ( *p < 33 ) ) );
        // Display the message
        printf( "\n%s failed with error %d (%s)", pMessage, eNum, sysMsg );
}

---

EDIT I tested your code using a remote workstation and I didn't have any problems whatsoever. So, it really looks like a security issue on the remote workstation.

Perhaps the Windows Firewall is blocking this?

--
Mats

hi matsp.
It can't be the firewall, because i turned it off on both machines (local and virtual).

BobS0327: I'm going to try that and tell you what happened.

Quote:

---

I tested your code using a remote workstation and I didn't have any problems whatsoever. So, it really looks like a security issue on the remote workstation.

---

If it is a security issue on the remote workstation, how can i resolve it?

Quote:

---

If it is a security issue on the remote workstation, how can i resolve it?

---

Well, first of all my caveat, I'm not a network administrator by any stretch of the imagination. With that said, I would do the following:

Assuming that you are using Active directory, I would determine the name of the global SysAdmin account within AD. I would also verify that your userid is in this global account. Next, I would check the local admin groups on the target workstation.by right clicking My Computer, selecting Manage and then selecting Local Users and Groups. Verify that the global AD Sysadmin group is in the Local sysadmin Group.

If not, check the Local users group to verify that possibly the Network Administrator has only added one userid as SysAdmin on the target computer. If there is a sysAdmin type in Local users but your code still does not work, check to determine whether or not the password has expired on the target workstation for this local Sysadmin. I've recently run into a situation where the Network Administrator added a sysAdmin to the local users but set the password to expire every 30 days. This is not good when you try to use the account on the 31st day.

This is just a starting point since there can be so many variation determined by what the Network/Security Administrator wishes to accomplish.

I would determine the exact problem by reviewing the DisplayError output message before tinkering with an global\local groups\users.

Finally, it would behoove you to discuss any changes you would like to make with your Network Administator. They normally have a very good reason for everything.

Quote:

---

Originally Posted by BobS0327

Finally, it would behoove you to discuss any changes you would like to make with your Network Administator. They normally have a very good reason for everything.

---

Or at least what THEY THINK is a good reason... And it may not be easy to change their decisions and reasoning here either.

--
Mats

BobS0327:
where do i add the function? On the server app or client app?
and were in the code do you want me to call it?

Quote:

---

Originally Posted by IndioDoido

BobS0327:
where do i add the function? On the server app or client app?
and were in the code do you want me to call it?

---

You add it somewhere in the code where you are experiencing the failing call(s) to the OS - and add calls after all system calls that may fail.

--
Mats

In the client:

Code:

---

DWORD dwReadWritten;
        DWORD dwMode=PIPE_READMODE_MESSAGE|PIPE_WAIT;
        char buffer[1024];
    char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

    if (!WaitNamedPipe(pszPipe, NMPWAIT_WAIT_FOREVER))
        {
                      DisplayError("waitNamedPipe error");
                printf("CLIENT: Error connecting\n");
                return 1;
        }

---

And as Matsp explained wherever else you're experiencing problems.

okay...
this is what i did:

Code:

---

char *pszPipe="\\\\192.168.1.80\\pipe\\ServNT";

if (!WaitNamedPipe(pszPipe, NMPWAIT_WAIT_FOREVER))
{
        DisplayError(pszPipe);
        return 1;
}

---

and this is the result error:

Quote:

---

waitNamedPipe error faild with error 3 (System cannot find the path specified)

---

I did testes with the firewall on\off and the error was the same :(

What now? I'm against violence, but do i have to beat my virtual machine up?

Do you really have \\ in all the places in that path?

sorry tabstop, i pasted the wrong thing :(
My previous post is now correct

Please don't take offense if I ask the obvious.

Have you verified that your remote workstation service is installed AND running prior to executing the client?

lol, no offense taken ;)

Yes. Every time i make changes to the server code i remove the old one and then install the newer version, then run it from windows services.

I think the problem is really in the authentication :S

This project uses windows authentication between the client and server.
If i'm in a specific account sending a message through a named pipe to a server, the server machine must accept my account.

How do i make the server accept my account? If this is actually the issue...

Quote:

---

Originally Posted by IndioDoido

I think the problem is really in the authentication :S

This project uses windows authentication between the client and server.
If i'm in a specific account sending a message through a named pipe to a server, the server machine must accept my account.

How do i make the server accept my account? If this is actually the issue...

---

Again, my caveat, I'm not a Network/Security administrator.

At the target workstation, right click on My computer, click on manage, click on Local users and Groups, Select Groups in the right pane. There should be an Administrators group listed. Add your userid to this group. Click on Ok etc to back out. Reboot the target workstation and go into control panel and select user accounts. you should see your user account listed. Reset the password of this account to match your password. Once this is done try your app.

One other thing. since I work with Active directory, I have always had to preface the added user account with the domain. For Instance, when I added my account to the Local Admin group, I had to use Domain\123456. I' m not sure if you will have to do that. You may just need to add 123456.

Quote:

---

At the target workstation, right click on My computer, click on manage, click on Local users and Groups, Select Groups in the right pane. There should be an Administrators group listed. Add your userid to this group.

---

When you say to add my userid to that group, do you mean the userid i use in the host machine? If so, how can i get my userid? :S

Quote:

---

When you say to add my userid to that group, do you mean the userid i use in the host machine? If so, how can i get my userid? :S

---

You should find your userid (user name) from the task manager. It's the Users tab. What type of network are you on? Is it a school network, home network or corporate network?

But anyway, I don't think it's an authentication issue because if it were, the returned error code would be 1326, Logon failure: Unknown user name or bad password.

The error 3 indicates that it's a poorly formed path. Humor me again and change this

Code:

---

char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

---

to this

Code:

---

char pszPipe[] = {"\\\\192.168.1.82\\pipe\\ServNT"};

---

oh! i thought that userid was different from user name, sorry :S

ok, i will humor you...
after changing:

Code:

---

char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

---

to this

Code:

---

char pszPipe[] = {"\\\\192.168.1.82\\pipe\\ServNT"};

---

i can know access the server :D
but........i know get a SetNamedPipeHandlerState error.
At least we are progressing :D

Quote:

---

Originally Posted by IndioDoido

oh! i thought that userid was different from user name, sorry :S

ok, i will humor you...
after changing:

Code:

---

char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

---

to this

Code:

---

char pszPipe[] = {"\\\\192.168.1.82\\pipe\\ServNT"};

---

i can know access the server :D
but........i know get a SetNamedPipeHandlerState error.
At least we are progressing :D

---

That change should make absolutely NO DIFFERENCE, other than as to how the memory is actually allocated for the variable. If it does, I'm guessing you are getting some sort of memory overwrite problem.

--
Mats

your right!

i changed it back to:

Code:

---

char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

---

and it "works"...

really strange :S
i might have been doing something wrong, but anyway, at least now i can access the server.

How can i resolve this other error?

Quote:

---

SetNamedPipeHandlerState error

---

What is the exact error code?

Quote:

---

Originally Posted by IndioDoido

your right!

i changed it back to:

Code:

---

char *pszPipe="\\\\192.168.1.82\\pipe\\ServNT";

---

and it "works"...

really strange :S
i might have been doing something wrong, but anyway, at least now i can access the server.

How can i resolve this other error?

---

The error that you were getting, error 3 "System cannot find the specified path" indicated that the path was poorly formed which indicated that there was something wrong with *pszPipe. But yet I kept staring at that variable and my mind told me that there wasn't anything wrong with it. I just didn't want to believe the returned error code because *pszPipe looked correct to me.

I can't recreate the error. Otherwise, I'd run it thru Ollydbg and tell you exactly what the issue is.

Quote:

---

Originally Posted by IndioDoido

your right!
and it "works"...

---

Are you sure? Your error checking code is contains errors :)

Code:

---

hPipe = CreateFile(pszPipe, GENERIC_WRITE|GENERIC_READ, 0, NULL,
                      OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);
        if (hPipe==NULL)
        {
                printf("CLIENT: Error a connecting \n");
                return 1;
        }

---

hPipe will never equal NULL as CreateFile returns INVALID_HANDLE_VALUE when it fails.

Quote:

---

Originally Posted by adeyblue

Are you sure? Your error checking code is contains errors :)

Code:

---

hPipe = CreateFile(pszPipe, GENERIC_WRITE|GENERIC_READ, 0, NULL,
                      OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);
        if (hPipe==NULL)
        {
                printf("CLIENT: Error a connecting \n");
                return 1;
        }

---

hPipe will never equal NULL as CreateFile returns INVALID_HANDLE_VALUE when it fails.

---

Good catch.

This probably is the source of his SetNamedPipeHandlerState error since CreateFile will always be considered to execute successfully whether or not it was actually successful using his current code.

hey everyone!

My client and server apps are working nicely on the same machine. I can send command lines from the client app to the server via pipe and the server runs 's the commands. How can there be an error in the code if i can accomplish these tasks?

About the issue of not connecting to my VM. My teacher said that to communicate from different machines in this project i have to be logged as the same user in both machines :S and for that i have to use a some "start" command line. For example, before a run the client app on the host machine i have to log-on to the VM, and for that i run this command:

Code:

---

start \\vmIP Username Password

---

After that i run my client app.

That's what i did...and know i can access the server.
But.....when i send a command line from the client app to the server, the server gives me a "ReadFile error 109" in the event viewer :(

Why is that happening?

Quote:

---

Originally Posted by adeyblue

Are you sure? Your error checking code is contains errors :)

Code:

---

hPipe = CreateFile(pszPipe, GENERIC_WRITE|GENERIC_READ, 0, NULL,
                      OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL);
        if (hPipe==NULL)
        {
                printf("CLIENT: Error a connecting \n");
                return 1;
        }

---

hPipe will never equal NULL as CreateFile returns INVALID_HANDLE_VALUE when it fails.

---

As adeyblue points out, you're incorrectly checking the return value of Create File

It should be:

Code:

---

if (hPipe==INVALID_HANDLE_VALUE)
        {
                printf("CLIENT: Error a connecting \n");
                return 1;
        }

---

The reason being is that INVALID_HANDLE_VALUE is 6 and if memory serves me correctly, NULL evaluates to 0 (zero). Thus, your CreateFile call will ALWAYS be successful whether or not it is actually successful. If CreateFile fails, it will return six not zero.

The 109 error translates to "the named pipe has ended". Normally, this should be an informational message. Named pipes work in the following manner:

The server makes a client connection, communicates with the client until the client disconnects casuing ReadFile to return FALSE. it then disconnects the server side connection which causes the "109 the name pipe has ended" message. and then the server attempts to connect to another client.

Thus, under normal circumstances the 109 message is just an informational message.

hi BobS0327

i've changed the error checking code to:

Code:

---

if (hPipe==INVALID_HANDLE_VALUE)

---

I still get the same error in the event viewer when i send a command to the server :(
But, in the client side i now "trigger" this checking:

Code:

---

        if (!SetNamedPipeHandleState(hPipe,&dwMode,NULL,NULL))
        {
                printf("CLIENT: Error SetNamedPipeHandleState\n");
                return 1;
        }

---

Quote:

---

Originally Posted by IndioDoido

hi BobS0327

i've changed the error checking code to:

Code:

---

if (hPipe==INVALID_HANDLE_VALUE)

---

I still get the same error in the event viewer when i send a command to the server :(
But, in the client side i now "trigger" this checking:

Code:

---

        if (!SetNamedPipeHandleState(hPipe,&dwMode,NULL,NULL))
        {
                printf("CLIENT: Error SetNamedPipeHandleState\n");
                return 1;
        }

---

---

As far as the event viewer error code, I would suggest that you call PeekNamedPipe before calling ReadFile to verify that there is actually some data in the pipe for the ReadFile function. If there isn't any data in the pipe then do NOT execute ReadFile.

How about executing the DisplayError function when SetNamedPipeHandleState fails and post the detailed error message?

Quote:

---

But, in the client side i now "trigger" this checking:

Code:

---

if (!SetNamedPipeHandleState(hPipe,&dwMode,NULL,NULL))
{
        printf("CLIENT: Error SetNamedPipeHandleState\n");
        return 1;
}

---

---

Sorry about my last this comment, i was wrong, it "triggered" the same checking:

Code:

---

if (hPipe==INVALID_HANDLE_VALUE)

---

I called the DisplayError function in that checking, and it showed me this error:
" \\serverIP\\pipe\\ServNT failed with error 5 (Access is denied) "

Quote:

---

Originally Posted by IndioDoido

Sorry about my last this comment, i was wrong, it "triggered" the same checking:

Code:

---

if (hPipe==INVALID_HANDLE_VALUE)

---

I called the DisplayError function in that checking, and it showed me this error:
" \\serverIP\\pipe\\ServNT failed with error 5 (Access is denied) "

---

Are you using an Administrator or non Administartor account on the client workstation? An Administrator login on the client workstation will eliminate the access denied error. Otherwise, if you have to use a non Admin account on the client workstation, you'll have to specify an ACL in the Createfile lpSecurityAttributes parameter. Currently, you're just using the default (NULL) security attributes which grants full control to the LocalSystem account and administrators. They also grant read access to members of the Everyone group and the anonymous account. You fall into the latter group if you're using a non Admin account.

I would first check with your instructor on this issue prior to fiddling with the ACL's since that may be beyond the scope of your project.

i already spoken with my professor, and he said that i need to have admin privileges in both host and virtual machines. And for the pipes to work i have to activate file and printer sharing in both machines, because named pipes work with the same protocol.

i'm using a admin account in my host and virtual machine. I can access both of them with admin privileges, and they both have file sharing activated.

The only difference between both machines is the OS. I'm running Vista x64 in the host and WinXP in the virtual.

:confused:

Quote:

---

The only difference between both machines is the OS. I'm running Vista x64 in the host and WinXP in the virtual.

---

I believe the source of your problem may be Vista. I've tested your app under XP Home, XP Pro and Win2k and have not encountered any problems . But when I tested under Vista, I repeatedly encountered the 109 error indicating that the pipe has ended. So, IMHO, it's a Vista issue not an application issue. I have almost no Vista experience, so I can't offer any solutions.

DAMMM VISTA!!! :mad:

Tomorrow i'm going to test my app between two winXP OS, and see the results.

By the way...isn't it strange i can run the client and server app both under vista without having any problems, but when i run the client on vista and the server on a virtual machine running winXP, they don't work?

To log your account into the virtual from your PC can use

net use \\<<IP>> /user:<<UID>> <<PWD>>

Quote:

---

Originally Posted by IndioDoido

By the way...isn't it strange i can run the client and server app both under vista without having any problems, but when i run the client on vista and the server on a virtual machine running winXP, they don't work?

---

Are you using remote desktop to connect to the virtual?

If so make sure you use the command 'mstsc /console' to open only ONE desktop, not one per connection. (not sure if this occurs in XP)
Other instances of your app may be running in other desktops and causing issues.

hey everyone!
My problem is finally solved :)

The problem was with my anti-virus software...yes, it's true. I wasn't able to connect to my virtual machine because i had bit defender installed, and bit defender blocks named pipe transfers...
I uninstalled the software, set the windows named pipe properties back to normal and...EVERYTHING IS WORKING :D

I've learned two things with all this...
First, never install an anti-virus on a virtual machine. Second, if you don't follow the first rule, don't use bit defender :p

Thanks for all the help guys!

oh...and i owe windows vista an apology, both my applications work fine with it :)