500 Internal Server error message when trying to access Oracle Enterprise Manager 10g Application Server Control

We have 4 servers running Oracle AS 10.1.2.0.2 on AIX 5.3.  Last week they were running fine but this week when we try to access the Enterprise Manager Application Server Control on three of them we get a HTTP/500 Internal Server error message and on the fourth we just get a cannot connect error (basically there's no HTTPD process listening on the port).  The HTTP/500 error message is the same on each one:

500 Internal Server Error

java.lang.NoClassDefFoundError: oracle/sysman/eml/app/Console
      at java.lang.Class.forName0(Native Method)
      at java.lang.Class.forName(Class.java(Compiled Code))
      at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.2)].server.http.HttpApplication.loadServlet(HttpApplication.java(Compiled Code))
      at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.2)].server.http.HttpApplication.findServlet(HttpApplication.java(Compiled Code))
      at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.2)].server.http.HttpApplication.getRequestDispatcher(HttpApplication.java(Compiled Code))
      at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.2)].server.http.HttpRequestHandler.processRequest(HttpRequestHandler.java(Compiled Code))
      at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.2)].server.http.HttpRequestHandler.run(HttpRequestHandler.java(Compiled Code))
      at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.2)].server.http.HttpRequestHandler.run(HttpRequestHandler.java(Compiled Code))
      at com.evermind[Oracle Application Server Containers for J2EE 10g (10.1.2.0.2)].util.ReleasableResourcePooledExecutor$MyWorker.run(ReleasableResourcePooledExecutor.java(Compiled Code))
      at java.lang.Thread.run(Thread.java:568)

running 'opmnctl status' on those boxes gives:

$ opmnctl status

Processes in Instance: docutrn_AS.doctm.sap.brm.pri
-------------------+--------------------+---------+---------
ias-component      | process-type       |     pid | status
-------------------+--------------------+---------+---------
HTTP_Server        | HTTP_Server        | 1212618 | Alive
LogLoader          | logloaderd         |     N/A | Down
dcm-daemon         | dcm-daemon         |  520380 | Alive
DSA                | DSA                |     N/A | Down
OC4J               | home               |  245904 | Alive

$

On the one which doesn't the same command gives the same result except for the dcm-daemon process isn't running.

We've tried Google and Metalink searches but nothing promising shows up.

The only thing that we know that happened between when we know it was last working and when we first discovered it wasn't is that the machine that was used as an X-Server (i.e. display) for the install was rebooted.  we do not know of any changes, patches or other things that might have caused this.

Stephen