[snip]
>
>We have developed the following preliminary list of business goals:
>
1> o System Manager (SM) or Operator (OP) can identify all of the clients
>   that are actively using the server:
>   - can identify the client associated with any server process.
>   - can detect if a client is currently doing any "work".
>
2> o SM or OP can identify the cause of a problem:
>   - can identify the client that is the root cause of a performance
>     problem on the machine.
>
3> o SM or OP can easily resolve problems:
>   - can reach an end user (via phone, email, etc) client of the 3000.
>
4> o SM can easily avoid problems:
>   - can write home grown "security" scripts that define 3000 access rules
>     based in part on the connection type.
>   - can use a 3rd party security package that can identify the originator
>     of each connection.
>
5> o SM (and MIS management) can monitor usage of services and develop a
>   trend-history of usage for any service offered.
>
 
Our primary uses for this would be for security, either thru HP or a 3rd
party,
and to identify the client that is the root cause of a problem.  This is
probably
the most important point, since when a problem occurs, it's usually too late
to
ask the system to track the client down, because you are already hung up or
nearly so.  If the client info was readily available, it would go a long way
to
solving those pesky intermittent slow-downs, etc.
 
============================================================
M. J. (Mike) Lippold              [log in to unmask]
Manager, Systems and Telecommunications
Lockheed Engineering and Sciences Company