2.3.1 Starting the Logical Server and Checking the Operations
The following figure illustrates starting the logical server and checking the operations:
-
The Management Server sends a request to the Administration Agent to start the logical server.
-
The administration agent executes the start command of the logical server to start the requested logical server.
-
The process ID of the logical server process is acquired in the Administration Agent.
-
The administration agent notifies the Management Server that the logical server is now running.
-
The administration agent monitors the logical server process and checks the operation of the logical server.
The administration agent uses the process ID of the logical server process, checks if the process exists, and if it is able
to verify the existence of the process, it checks the operation of the logical server.
-
The administration agent notifies the status of the logical server to the Management Server.
The following table describes the method of checking the process and the operation for each type of logical server. Note that
for the logical servers in case of which the operation checking method is not mentioned in the table, only the check for verifying
whether the process exists is carried out.
Table 2‒2: Methods for checking the operations of logical servers
Type of logical server
|
Method of starting the process#
|
Method of checking the operation of the logical server
|
Check that the process exists
|
Check the operation
|
Logical performance tracer
|
Indirect startup
|
Check that the process exists by using the process ID obtained by using a command provided by the performance tracer.
|
--
|
Logical Smart Agent
|
Direct startup
|
Check that the process exists by using the process ID of the command that started the process.
|
--
|
Logical Naming Service
|
Direct startup
|
Check that the process exists by using the process ID of the command that started the process.
|
Check that the root context can be obtained.
|
Logical CTM domain manager
|
Indirect startup
|
Check that the process exists by using the process ID obtained by using a command provided by the CTM domain manager.
|
--
|
Logical CTM
|
Naming service
|
Direct startup
|
Check that the process exists by using the process ID of the command that started the process.
|
Check that the root context can be obtained.
|
CTM daemon
|
Indirect startup
|
Check that the process exists by using the process ID obtained by using a command provided by the CTM daemon.
|
--
|
Logical J2EE server
|
Direct startup
|
Check that the process exists by using the process ID of the command that started the process.
|
Check that there is a response to the call by RMI.
|
Logical Web server
|
Indirect startup
|
Check that the process exists by using the process ID of the control process obtained from the httpd.pid file generated by the web server.
|
Check that a correct response can be received through the HTTP access of the URL for checking the Cosminexus HTTP Server operations.
The response is sent by the server process of the Web server.
|
Logical user server
|
Direct startup
|
Check that the process exists by using the process ID of the command that started the process.
|
When the isAlive command is defined in the logical user server definition file, check that the processing is executed by using the isAlive command. When not defined, it is assumed that the operation check was successful.
|
Indirect startup
|
Check that the process exists by using the process ID obtained by using the getProcessID command.
|
Use the isAlive command when it is defined in the logical user server definition file. When not defined, it is assumed that the operation
check was successful.
|
- Legend:
-
--: None
- #
-
In the case of direct setup, the command that is executed is also monitored. In the case of indirect setup, the services or
the processes invoked by the command are monitored. The command is also used to stop the services and processes that are invoked
using other commands.
To Page Top