从远程服务器上运行的控制台应用程序捕获输出

Question

I am running console app on remote server using WMI (C# code)

I want to capture the output from the process.

Can someone please help me in doing so?

Scenario:

I have a webpage where user will provide server details. When user clicks on submit, I want to call app on remote server and pass parameters. App on remote server will perform the action and will return back the output (output can be number, string or bool).

Based on the output I need to prompt the user for more details or show error.

object[] processToRun = new[] { "\"Path to exe\" \"Param1\" \"Param2\"" };
ConnectionOptions connection = new ConnectionOptions();
connection.Username = "username";
connection.Password = "password";
ManagementScope wmiScope = new ManagementScope(String.Format("\\\\{0}\\root\\cimv2", "ServerName"), connection);
ManagementClass wmiProcess = new ManagementClass(wmiScope, new ManagementPath("Win32_Process"), new ObjectGetOptions());
wmiProcess.InvokeMethod("Create", processToRun);

I want to get the output of this.

Temporary Solution : But not working

As a temporary solution, i added code to create a log file on a shared path which is accessible by everyone. But again here I am facing issue, when i call exe on remote server to write to the shared path, i always get error saying 'Access denied'. But when i run the exe from the server itself, it works fine.

Answer 1

I've used PsExec in the past. This helped when a Firewall was blocking WMI requests. You should be able to start PsExec just like starting any other process from C# and just redirect standard output. I think you'll have some extra bytes and bits in the output stream from PsExec itself, but it should get you where you need to go after some trimming.

Answer 2

Assuming you want to collect the output from a second machine, one way forward is using sockets. Second machine will host a socket based listening service bound predefined port.

Once your console app has something to send back, it would open up a connection to the second machine and send the data.

If there are many clients and you are expecting frequent communication from them, you can go to a solution based on something like WCF or web services for which there are many off the shelf frameworks available

Based on the addition info you provided ,

When you create the remote console process, pass it a unique id (GUID) which ties it to the client request originated at the web server. Your console app will now capture input and send it to the server via a socket along with the guid.

The server can log each output into a database table using GUID to identify each session. Now the user who launch can query the output using the same GUID (remember the GUID is generated when the user create the request, so the user will have it always) and the server will get the output from the table and display

Additionally you can have a management task to clear output table of all results older than x number of days

Hope this helps