8.2.1 adminagent.properties (Administration Agent property file)

{adminagent.adapter.allowedHosts

Specify a host name or an IP address that can access to the server.

To specify multiple IP addresses, demarcate using commas (,) and you do not insert any spaces. You can specify the IP address as a regular expression using meta characters. When an IP address is specified as a regular expression, the Administration Agent uses an available local IP address that matches the expression. If multiple IP addresses exist, all the IP addresses that match the expression will be granted the access permission.

Note that when you specify an IP address as a regular expression, you cannot specify multiple values using commas (,).

The values that can be specified are as follows:

• Host name

• IPv4 address

None

05-05

adminagent.adapter.bind_host

Specify the host name or the IP address that will be used for communicating with the Administration Agent. You can choose any of the IP addresses for communication, when using Administration Agent on a host that has multiple physical network interfaces or a host that has multiple logical IP addresses assigned for a single physical network interface.

If an invalid value is specified, the Administration Agent will terminate abnormally.

You can specify the IP address as a regular expression using meta characters. When the IP address is specified as a regular expression, the Administration Agent uses an available local IP address that matches the expression.^#1

The values that can be specified are as follows:

• Host name

• IPv4 address

None

06-00

adminagent.adapter.port

Specify port numbers used by the server functionality.

Specify the value using an integer from 1 to 65535.

20295

05-05

adminagent.cluster.localaddress.check

Specify whether to stop the standby node during node switching in the Application Server.

During node switching in the Application Server, if the logical server and Administration Agent in the standby node are not stopped, this key will stop the logical server and Administration Agent.

If you specify true:

The running logical server and Administration Agent in the standby node are stopped during the node switching in the Application Server.

The local address is checked and if the cluster IP address is not specified, the logical server will be stopped and then the Administration Agent will be stopped.

If you specify false:

The running logical server and Administration Agent in the standby node are not stopped during the node switching in the Application Server.

Specify one of the following values:

• true

• false

false

07-10

adminagent.finalization.stop_servers

Specify whether to stop the logical server when the Administration Agent terminates.

If you specify true:

All the logical servers will be stopped in the Administration Agent termination processing.

If you specify false:

The logical servers will not be stopped in the Administration Agent termination processing.

Specify one of the following values:

• true

• false

false

06-50

adminagent.forcestop.threaddump

Specify whether to output the thread dump when the J2EE server is terminated forcefully. J2EE servers are the only logical servers to which the above setting are applicable.

If you specify true:

Thread dump will be output. In such cases, the forced termination processing is same as the case when the cjstopsv command is executed with the -fd option.^#2

If you specify false:

Thread dump will not be output.

Specify one of the following values:

• true

• false

false

06-50

adminagent.forcestop.threaddump.interval

Specify an integer from 1 to 2147483647 (unit: seconds) for the interval to confirm the completion of thread dump output, when settings are specified to output thread dump during the forced termination of the J2EE server.

When the value outside the range of 1 to 2147483647 is specified, the default value will be set.

Specify the value using an integer from 1 to 2147483647.

1

06-50

adminagent.forcestop.threaddump.timeout

Specify an integer from -1 to 2147483647 (unit: seconds) for the output wait time for thread dump, crash dump, or core output when Do not monitor is set for Forced stop monitoring time in the settings for starting and stopping the J2EE server.

If you specify -1, the system waits until thread dump, crash dump, or core output finishes.

When you specify the value outside the range of -1 to 2147483647, the default value will be set.

If the specified time passes and the thread dump, crash dump, or core output is not yet finished, the J2EE server is forcibly stopped.^#2

Specify the value using an integer from -1 to 2147483647.

45

06-50

adminagent.hws.group

When using Cosminexus HTTP Server, specify the group name of the user starting Cosminexus HTTP Server.

When omitted, start with the group name of the Administration Agent.

This key is for UNIX.^#3

Group name

None

08-00

adminagent.hws.owner

When using Cosminexus HTTP Server, specify the name of the user starting Cosminexus HTTP Server.

When omitted, start with the user name of the administration agent.

Note that this key is used for UNIX. ^#3

User name

None

08-00

adminagent.hws.sys_cmd.abnormal_end.traceinfo

Specify whether to collect the Web server internal trace, when executing the command for error detection.

If you specify true:

The internal trace will be collected.

If you specify false:

The internal trace will not be collected.

Specify one of the following values:

• true

• false

true

07-00

adminagent.hws.watch.method

Specify the HTTP method for checking the operation of Cosminexus HTTP Server when 2 is specified in adminagent.hws.watch.level. For the specifiable method, you can specify either "HEAD" or "OPTIONS". ^#4

If you specify "HEAD":

Accesses the URL specified in the key adminagent.hws.watch.url and checks whether returns any response.

If you specify "OPTIONS":

Checks whether the Web server can receive HTTP methods.

If you specify a forward slash (/) as path in the ProxyPass directive, all URLs are redirected to the web container of the J2EE server. In this case, therefore, we recommend that you specify OPTIONS so that URL access does not occur.

Specify one of the following values:

• HEAD

• OPTIONS

HEAD^#5

06-50

adminagent.hws.watch.url

Specify the URL for confirming the operations of Cosminexus HTTP Server, when "HEAD" is specified in adminagent.hws.watch.method. The Administrator Agent will determine the status of Hitachi Web Server from the response of this URL, so specify a URL that is actually accessible. Note that you specify the URL as an absolute path starting with "http" or as a relative path from the root context. When building multiple Web servers on the same host, specify the URL as a relative path.

(Example of specification)

When you specify http://HostA:80/index.html:

adminagent.hws.watch.url=http://HostA:80/index.html

When you specify http://localhost:port-nmber/index.html (in a multiple server configuration):

adminagent.hws.watch.url=index.html

URL

Root context (http://localhost:port-number/)

06-00

adminagent.j2ee.process.console_event.enabled

Specify whether to display the console output information of the J2EE server to the Eclipse Plug-in using Management Server.

If you specify true:

The console output information will be output to the Eclipse Plug-in using Management Server.

If you specify false:

The console output information will not be output to the Eclipse Plug-in using Management Server.

The key adminagent.process.consolelog.enabled must be set to true.

Since this consumes resources, Hitachi recommends that the key be set to false in operations that do not use Eclipse Plug-ins using Management Server.

Specify one of the following values:

• true

• false

false

07-10

adminagent.j2ee.process.console_log.enabled

Specify whether to output the console output information for the J2EE server in the console log.

If you specify true:

The console output information will be output to the console log.

If you specify false:

The console output information will not be output to the console log.

The key adminagent.process.consolelog.enabled must be set to true.

Specify one of the following values:

• true

• false

false

07-10

adminagent.j2ee.sys_cmd.abnormal_end.javatrace

Specify whether to acquire J2EE server stack trace when executing the command for error detection.

If you specify true:

The stack trace will be acquired.

If you specify false:

The stack trace will not be acquired.

Specify one of the following values:

• true

• false

true

07-00

adminagent.j2ee.sys_cmd.abnormal_end.threaddump

Specify whether to collect the J2EE server thread dump when executing the command for error detection.

If you specify true:

The J2EE server thread dump will be collected.

If you specify false:

The J2EE server thread dump will not be collected.

Specify one of the following values:

• true

• false

true

06-50

adminagent.jp1event.event_server_name

Specify a value same as the address set for the ports key in the event server settings file (conf) of the JP1/Base event service that is being used. If multiple addresses are specified for the ports key, choose any one of the specified addresses. When using an event service that has the value '0.0.0.0' (default value) specified for the ports key address, either omit this key or specify the host name of the local machine or localhost.

Host name or IPv4 dotted notation

localhost

06-70

adminagent.jp1event_enabled

Specify whether JP1 events will be issued from the Administration Agent.

If you specify true:

JP1 events will be issued.

If you specify false:

JP1 events will not be issued.

Specify one of the following values:

• true

• false

false

06-70

adminagent.log.filenum

Specify an integer from 1 to 16 for the number of log files.

Specify the value using an integer from 1 to 16.

4

05-05

adminagent.log.filesize

Specify a numeric value from 65536 to 2147483647 for the maximum size of each log file (units: bytes).

Specify the value using an integer from 65536 to 2147483647.

65536

05-05

adminagent.log.level

Specify an integer from -1 to 1000 for the level of log output.

• -1: No log will be output.

• 0: Normal operation: Normal operation.

• 10: Normal operation (verbose): same as normal operation, waiting for recurrence.

• 20: Recurrence test: Acquires debug level information used in the system environment setup or test phases.

• 30: Error investigation: Suitable for acquiring detailed error information for errors that are difficult to investigate.

Specify the value using an integer from -1 to 1000.

0

05-05

adminagent.lsinfo_dir

Specify the directory where the logical server information file will be output. If the specified directory does not exist, create it.

When you restart the Administration Agent, the Administration Agent will use this information to once again manage the logical servers being managed before the Administration Agent stops and make them operable. This directory is created when starting the logical server and is deleted after the logical server has stopped.

Directory name (absolute path)

Manager-installation-directory/spool/lsinfo

06-50

adminagent.maintenance.log.filenum

Specify an integer from 1 to 16 for the number of maintenance log files.

Specify the value using an integer from 1 to 16.

4

06-70

adminagent.maintenance.log.filesize

Specify an integer from 65536 to 2147483647 for the maximum size of each maintenance log file (units: bytes).

Specify the value using an integer from 65536 to 2147483647.

16777216

06-70

adminagent.prftrace_dir

Specify a character string for the path name of the directory to which the trace based performance analysis file is temporarily output. If no directory is specified, the specified directory is created.

Note:

The trace based performance analysis file is temporarily output in the following cases:

• When the system error detection command is executed due to the abnormal termination of the logical server

• When the snapshot log is collected by specifying the sub-command collect of the Management Server management command (mngsvrutil)

• When the snapshot log is collected with the management portal

Therefore, allocate free space at the output destination based on the following formula: File size output at the specification destination (maximum value) = PRF-trace-size (value specified for PrfTraceFileSize) ×number-of-PRF-trace-files (value specified for PrfTraceCount)

Note that if multiple PRFs exist on the same host, the sum of size of all the PRF trace files is used.

Directory name (Absolute path)

Application-Server-installation-directory/manager/tmp/

09-50

adminagent.process.consolelog.enabled

Specify whether to acquire the console output information for the processes invoked by the Administration Agent. However, when using the Eclipse Plug-in that uses Management Server, you must specify true as the value for this key.

If you specify true:

The console output information will be acquired.

If you specify false:

The console output information will not be acquired.

Specify one of the following values:

• true

• false

true

07-00

adminagent.process.consolelog.event.queue_size

Specify the event queue size for the event notification to the Management Server from the console output information for the processes invoked by the Administration Agent.

If the queue becomes full, the events are destroyed from the oldest event in the queue.

Hitachi recommends that you specify a default value for this property; and therefore, do not change the settings.

Specify the value using an integer from 1024 to 2147483647.

1024

07-00

adminagent.process.consolelog.filenum

Specify an integer from 1 to 16 as the number of log files output for the console output information of the processes invoked by the Administration Agent.

If the shift mode is specified in the sinaviagent.log.style key, this number becomes the number of backup files.

Specify the value using an integer from 1 to 16.

4

07-00

adminagent.process.consolelog.filesize

Specify an integer from 65536 to 2147483647 as the maximum size (unit: bytes) of each log file output for the console output information of the processes invoked by the Administration Agent.

Specify the value using an integer from 65536 to 2147483647.

65536

07-00

adminagent.process.consolelog.time

Specifies the time for rotating the log files used to output the console output information of the processes started by the Administration Agent in HHMMSS format as a value from 000000 to 235959. However, if the log file size reaches the upper limit before the time set in this property is reached, the log file will be rotated at that point.

If no time is specified, the files will be rotated based only on the log size.

Specify a value from 000000 to 235959 in the HHMMSS format.

None

09-00

adminagent.process.consolelog.style

Specifies the method of rotating the log files that output the console output information of the processes started by the Administration Agent.

If you specify SHIFT:

The log file will be rotated in the shift mode.

If you specify WRAP:

The log file will be rotated in the wraparound mode.

Specify one of the following values:

• SHIFT

• WRAP

WRAP

09-00

adminagent.rmi.log.filenum

Specify a numeric value for the number of maintenance log files in the RMI processing performed by the Administration Agent.

Specify the value using an integer from 1 to 16.

8

07-60

adminagent.rmi.log.filesize

Specify a numeric value for the maximum size of each maintenance log file in the RMI processing performed by the Administration Agent.

Specify the value using an integer from 65536 to 2147483647.

16777216

07-60

adminagent.rmi.log.level

Specify a character string for the output level of the maintenance log in the RMI processing performed by the Administration Agent.

<OFF>

The output level will not be output.

<SEVERE>, <WARNING>, <INFO>, <CONFIG>, <FINE>, <FINER>, <FINEST>

SEVERE only outputs severe problems. The order described here results in more detailed log output.

Specify one of the following values:

• OFF

• SEVERE

• WARNING

• INFO

• CONFIG

• FINE

• FINER

• FINEST

FINER

07-60

adminagent.snapshotlog.listfile.2.num_snapshots

Specify a numeric value from -2147483648 to 2147483647 as the number of snapshot log files that will be acquired as the secondary data for each logical server. If exceeds the specified file count, the log files with the oldest creation time are deleted sequentially. Also, if you specify a value not more than 0, the log files are not deleted.

Specify the value using an integer from -2147483648 to 2147483647.

10

06-50

adminagent.snapshotlog.log_dir

Specify a character string for the directory to output the snapshot log. If the specified directory does not exist, the directory will be created.

Directory name (absolute path)

com.cosminexus.manager.log.dir-of-manager.cfg/snapshot/

06-00

adminagent.snapshotlog.num_snapshots

Specify a numeric value from -2147483648 to 2147483647 as the number of snapshot log files that will be acquired as the primary data for each logical server. If exceeds the specified file count, the log files with the oldest creation time are deleted sequentially. Also, if you specify a value not more than 0, the log files are not deleted.

Specify the value using an integer from -2147483648 to 2147483647.

10

06-00

adminagent.sys_cmd.abnormal_end.prftrace

Specify whether to collect the performance analysis trace files when executing the command for error detection.

If you specify true:

The performance analysis trace files will be collected.

If you specify false:

The performance analysis trace files will not be collected.

Specify one of the following values:

• true

• false

true

06-50

adminagent.userserver.process.console_event.enabled

Specify whether to display the console output information of the user server with the Eclipse Plug-in using Management Server.

If you specify true:

The console output information will be output to the Eclipse Plug-in using Management Server.

If you specify false:

The console output information will not be output to the Eclipse Plug-in using Management Server.

The key adminagent.process.consolelog.enabled must be set to true.

Since this consumes resources, Hitachi recommends that the key be set to false in operations that do not use Eclipse Plug-ins using Management Server.

Specify one of the following values:

• true

• false

false

07-10

adminagent.userserver.process.console_log.enabled

Specify whether the console output information for the user-defined server will be output to the console log.

If you specify true:

The console output information will be output to the console log.

If you specify false:

The console output information will not be output to the console log.

The key adminagent.process.consolelog.enabled must be set to true.

Specify one of the following values:

• true

• false

false

07-10

adminagent.watch.retry_timeout.enabled

Specify whether retry is performed for the occurrence of a timeout during the confirmation of the logical server operations.

If you specify true:

Retry performed for occurrence of the timeout.

If you specify false:

Retry is not performed for occurrence of the timeout.

Specify one of the following values:

• true

• false

• Default: false

• Initial value: true

06-70

adminagent.server-type^#6.usr_cmd.abnormal_end

Specify the command for error detection using the absolute path. Use "/" as a file separator. You can change the command according to the type of the logical server where the error has occurred. If you specify the value that cannot be specified in server-type^#6, the value will be ignored.

In Windows:

When you create a command for error detection using a batch file, you must add "%{ComSpec} /C" before the command name. "%{ComSpec}" is replaced with the absolute path of cmd.exe. If "%{ComSpec} /C" is not specified, the command might not execute normally. Also, when you include one-byte spaces in the command path, make sure that you enclose the absolute path of the command in double quotations ("").

Example: If executing d:/my bat/j2eeabnormalend.bat, when an error occurs in the J2EE server

adminagent.j2ee.usr_cmd.abnormal_end=%{ComSpec} /c "d:/my bat/j2eeabnormalend.bat"

In UNIX:

When you create a command for error detection using a shell script, you must add "/bin/sh" before the command name. If "/bin/sh" is not specified, the command might not execute normally.

Example: If executing /home/user1/j2eeabnormalend.sh, when an error occurs in the J2EE server

adminagent.j2ee.usr_cmd.abnormal_end=/bin/sh/home/user1/j2eeabnormalend.sh

Command name (absolute path)

None

06-50

adminagent.server-type^#6.watch.interval

Specify an integer from 1 to 86400 as the interval (unit: seconds) for confirming the operations of the logical server.

If you specify an invalid value, the default value will be set.

Note that when the server type is user server, the watch-interval value specified in the Logical user server definition file or Easy Setup definition file becomes valid. If you use the logical server termination functionality and specify the time to wait for before terminating or forcibly terminating the logical server, specify a value larger than the value specified for this property. If you specify a value smaller than the value specified for this property, termination or forced termination might fail.

Specify the value using an integer from 1 to 86400.

10

06-00

adminagent.server-type^#6.watch.level

Specify the logical server operation confirmation level. You can specify 1 or 2 as the value.

If you specify 1:

The logical server operations will be confirmed by checking whether the process exists.

If you specify 2:

The logical server operations will be confirmed by checking whether the process exists and the access to the logical server. The logical servers on which operations can be checked are naming services, J2EE servers, and Web servers. You cannot specify the operations of logical user servers. For other logical servers, even if you specify 2, what you can perform is only checking whether the process exists.

Specify 1 or 2.

• For Naming service, J2EE server, and Web server: 2

• For logical servers other than above: 1

06-00

adminagent.server-type^#6.watch.retry_count

Specify an integer from 0 to 86400 as the retry count when the operation confirmation processing fails during the confirmation of logical server operations. By specifying the retry count, you can permit temporary operation confirmation failure.

Specify the retry interval using adminagent.server-type.watch.interval. If you specify 0, an error is assumed without retrying when the operation confirmation processing fails.

When the operation confirmation processing fails and when tried retrying, KEOS21033-W or KEOS21034-W is output to the Administration Agent log.

When the operation confirmation processing fails and when retry is not executed or if the retry count exceeds the specified count, KEOS20511-E or KEOS21035-E will be output to the Administration Agent log and the error will be notified to the Management Server.

Specify the value using an integer from 0 to 86400.

1

06-70

adminagent.server-type^#6.watch.start_time

When starting the logical server, specify an integer from 0 to 86400 for the time (unit: seconds) starting from the execution of the start command up to starting the operation confirmation.

By acquiring the time required for actual startup from the logical server log for logical server startup (KDJE30028-I of cjmessage?.log for the J2EE server) and by specifying the value for this property which is few seconds shorter than this time, you can reduce the unnecessary log that is output before the logical server starts and report the complete startup of the logical server to the Management Server promptly.

Specify the value greater than the value specified in this property as the start monitoring time set up in the logical server start and stop functionality. If you specify the value lesser than the specified value, the logical server will fail to start.

Specify the value using an integer from 0 to 86400.

Depending on the server-type^#6, the default value differs as follows:

• prf: 0

• smartagent: 3

• naming: 10

• ctmdm: 0

• ctm: 0

• ots: 0

• tcs: 3

• sfo: 45

• j2ee: 45

• hws: 0

• userserver: 0

06-70

adminagent.server-type^#6.watch.timeout

Specify an integer from 1 to 9223372036854775 (unit: seconds) for the time until timeout occurs during the confirmation of the logical server operations.

Specify the value using an integer from 1 to 9223372036854775.

60

06-50