StartupFailureisFatal Property question

Question

What is name of the registeredclass that we need to specify for the
startupFailureisFatal property. I have set it to the name of my virtual
startup class, the name of my startup class, left it alone as
registeredclass but I don't get the weblogic 4.5 server to shutdown when I
throw an exception from the startup class. I would appreciate it if someone
could give me an example of its use. Thanks.
My weblogic.property file
weblogic.system.startupClass.mezziaStartup=com.mezzia.planner.admin.MezziaSt
artup
weblogic.system.startupArgs.mezziaStartup=serverType=development,\
serverOwner=me
I have tried each of the following three settings but none of them cause the
server to shutdown.
weblogic.system.startupFailureisFatal.mezziaStartup=true
weblogic.system.startupFailureisFatal.registeredclass=true
weblogic.system.startupFailureisFatal.MezziaStartup=true
I get the following error but the server doesn't shutdown.
Wed Sep 06 16:37:37 EST 2000:<E> <WebLogicServer> Failed to invoke startup
class mezziaStartup=com.mezzia.planner.admin.MezziaStartup
I know it is probably something really simple but I can't figure out what I
am doing wrong. Thanks Again.
IRfan
---- BEA Documentation for Weblogic 4.5 ----
You can use the following property to require that specific startup classes
are correctly loaded before the server will start:
weblogic.system.startupFailureisFatal.registeredclass=boolean
The default is false, which will not affect the startup of WebLogic Server.
If set to true, the server will not start if the class or classes registered
to the indicated name throw an exception.

Answer 1

If an expression evaluates to multiple objects, then the Powershell pipeline will automatically create and return an array of those objects.
If the expression evaluates to a single object (sometimes referred to as being
scalar) then the pipeline will return one object of that type.
This can cause problems if the expression is normally expected to return multiple objects, and you write your script to always expect an array to be returned.  To mitigate that, in V3 they added a feature that will synthesize a Count property for the
object with a value of 1.  But this can only work as long as the object doesn't already have Count property.  If the object already has a Count property, then any reference to .count will return the value of the native property.
You see the same kind of situation with hash tables.  Hash table objects in Powershell have synthetic .keys and .values properties that return arrays of the hash table keys and values, respectively.  But Powershell syntax also allows you to reference
hash table keys using the same dot notation used for returning properties.  This produces a conflict if the hash table actually has keys that are named 'keys' or 'values'.  If it does then the actual hash table entry having those key names is returned,
not they key or value collection.  
[string](0..33|%{[char][int](46+("686552495351636652556262185355647068516270555358646562655775 0645570").substring(($_*2),2))})-replace " "