NT Performance Monitor Customizable COM Alerter


Home Page: http://www.OrbitDevelopment.com

Environment: Windows NT Server and Workstation 4.0

Problem

Companies do not have NT Monitorig systems to be able to determine, and notify engineering, when an NT Server resource gradually degrades, or become non-existing. There are tools available to do so, but they can become quite expensive.

If you are tired of your current monitoring system, like a phone call at 3:00am from your boss saying, "The server is down!", this tool is for you.

Goal

Find an easy low cost solution to resolve the problem. The solution must be cheap and effective, at the same time very customizable.

Solution

NT currently includes "NT Performance Monitor" (PerfMon). PerfMon allows users to see performance graphs on certain NT resources. It also allows users to set "Alerts" for the same NT resources that you can graphically monitor.

Adding an "Alert" allows you to "Run a Program on Alert". This is the basis for our solution. I have created a program called "PerfMonitorAlerter.exe". When setting an alert, you would use PerfMonitorAlerter as the program to run on any alert you set, for example:

PerfMonitorAlerter "Memory is running over 90 megabytes"

PerfMonitorAlerter process uses COM objects to process the alert message specified in its parameter text: "Memory is running over 90 megabytes".

Technology

When PerfMon triggers the alert, it will start PerfMonitorAlerter.exe program with the message ("Memory is running over 90 megabytes") as it fisrt parameter.

PerfMonitorAlerter.exe will check registry path store at

       HKEY_LOCAL_MACHINE\\SOFTWARE\\www.OrbitDevelopment.com\\PerfMonitorAlerter


for any COM CLSIDs that implement a common interface, IAlertInterface:

    import "oaidl.idl";
    import "ocidl.idl";
	[
		object,
		uuid(B6C11879-3003-11D3-9257-00104BF7F161),
	
		helpstring("IAlertInterface Interface"),
		pointer_default(unique)
	]
	interface IAlertInterface : IUnknown
	{
		[helpstring("method EmergencyAlert")] HRESULT EmergencyAlert([in] BSTR bAlertName);
	};

In this code example case, PerfMonitorAlerter.exe will:

  • Create an instance of object "{B6C1187A-3003-11D3-9257-00104BF7F161}".
  • Query Interface IAlertInterface
  • Call method pIAlertInterface ->EmergencyAlert("Memory is running over 90 megabytes" )
  • Release the object

        : : : : : : : : : : : : 
        : : : : : : : : : : : : 
		
        // Registry COM Iterator

        while ( TRUE )    
        {
            char  szSubkey[ 64 ];
            DWORD cbSubkeySize = 64;
            CLSID clsidAlerter;
            BSTR  bstrCLSIDFromReg = NULL;

            if ( ERROR_NO_MORE_ITEMS == ::RegEnumKeyEx( regKey, regEnumIndex++, szSubkey, &cbSubkeySize, 0, 0, 0, 0 ) )            
                break;

            // Convert to BSTR, the convert it CLSID
            bstrCLSIDFromReg = charToBstr(szSubkey);
            ::CLSIDFromString(bstrCLSIDFromReg, &clsidAlerter); 
            if (bstrCLSIDFromReg) ::SysFreeString(bstrCLSIDFromReg);

            // Create object to notify
            if ( (bstrServiceName != NULL) && S_OK == ::CoCreateInstance( clsidAlerter, NULL, CLSCTX_INPROC_SERVER, IID_IAlertInterface, (void**) &pIAlertInterface )) 
            {
                // Send Alert
                pIAlertInterface->EmergencyAlert(bstrServiceName);
            }

            // Cleanup/Release the object
            if (pIAlertInterface) pIAlertInterface->Release();
        }

        : : : : : : : : : : : : 
This architecture will allow create one, or many, COM object to be notified, without concern for implementation details. The COM objects implementing IAlertInterface may send emails, log entries, call a pager, etc..

Although this is not the answer to all NT monitoring problems, it is an easy, customizable and powerful approach using COM communication technology.

Documentation Note

For detailed documentation (Word document), please feel free to contact me

Downloads

Download PerfMonitorAlerter application - 14 kb
Download source - 8 Kb


Comments

  • There are no comments yet. Be the first to comment!

Leave a Comment
  • Your email address will not be published. All fields are required.

Top White Papers and Webcasts

  • Intelligent N+X Redundancy, Placement Affinities, & Future Proofing in the Virtualized Data Center Virtualization brought about the ability to simplify business continuity management in IT. Workload portability and data replication capabilities mean that physical infrastructure failures no longer need impact application services, and they can rapidly be recovered even in the event of complete site failure. However, Enterprises and Service Providers face new challenges ensuring they have enough compute …

  • Data integrity and ultra-high performance dictate the success and growth of many companies. One of these companies is BridgePay Network Solutions, a recently launched and rapidly growing financial services organization that allows merchants around the world to process millions of daily credit card transactions. Due to the nature of their business, their IT team needed to strike the perfect balance between meeting regulatory-mandated data security measures with the lowest possible levels of latency and …

Most Popular Programming Stories

More for Developers

RSS Feeds

Thanks for your registration, follow us on our social networks to keep up-to-date