Job Management Partner 1/Automatic Job Management System 3 Administration Guide

[Contents][Glossary][Index][Back][Next]


7.2.1 Temporarily changing the start mode of JP1/AJS3

The following describes the procedure for temporarily changing the start mode of JP1/AJS3 - Manager or JP1/AJS3 - Agent.

To specify the start mode always used in JP1/AJS3 - Manager, use the jajs_config command to set the STARTMODE environment setting parameter.

For details about the environment setting parameters, see 2.2 Setting up the scheduler service environment in the Job Management Partner 1/Automatic Job Management System 3 Configuration Guide 2.

Organization of this subsection
(1) In Windows
(2) In UNIX
(3) Jobnet and job statuses for each start mode
(4) Job statuses on the manager host when an agent host is restarted

(1) In Windows

To temporarily change the start mode:

  1. In the Control Panel window, choose Services. Alternatively, choose Administrative Tools and then Services.
  2. Select the name of the JP1/AJS3 service you want to start.
  3. In Startup Parameter, specify the start mode.
    The following table lists the options you can specify.

    Table 7-1 Options that can be specified in Startup Parameter

    Operand Start mode
    -cold The service is cold-started.
    -warm The service is warm-started.
    -hot The service is hot-started.

  4. Click the Start button.

Note that if you have executed the net start service-name command, you cannot temporarily change the start mode.

Cautionary notes
  • If the temporarily specified service start mode differs from the service start mode that was specified during setup, the temporarily specified service start mode has priority.
  • If you do not specify an option in JP1/AJS3 - Manager, the service start mode specified at setup is used.
  • If you do not specify an option in JP1/AJS3 - Agent, -warm is assumed.

The status of jobnets and jobs depends on the specified service start mode. For details about the status of jobnets and jobs for each service start mode, see (3) Jobnet and job statuses for each start mode.

(2) In UNIX

To temporarily change the start mode:

  1. Execute the jajs_spmd command with an argument specified.
    The following table lists the options you can specify.

    Table 7-2 Options that can be specified in the jajs_spmd command

    Operand Operation
    -cold The service is cold-started.
    -warm The service is warm-started.
    -hot The service is hot-started.

Cautionary notes
  • If the temporarily specified service start mode differs from the service start mode that was specified during setup, the temporarily specified service start mode has priority.
  • If you do not specify an option in JP1/AJS3 - Manager, the service start mode specified at setup is used.
  • If you do not specify an option in JP1/AJS3 - Agent, -warm is assumed.

The status of jobnets and jobs depends on the specified service start mode. For details about the status of jobnets and jobs for each service start mode, see (3) Jobnet and job statuses for each start mode.

(3) Jobnet and job statuses for each start mode

When a JP1/AJS3 service on the manager host is restarted, the statuses of jobnets and jobs differ depending on the start mode of the JP1/AJS3 service. When a JP1/AJS3 service on the agent host is restarted, the status of jobs on the agent host differs depending on the start mode of the JP1/AJS3 service.

The following describes the status of jobnets and jobs when a JP1/AJS3 service on the manager host is restarted, and the status of jobs when a JP1/AJS3 service on the agent host is restarted.

(a) Statuses when a JP1/AJS3 service on the manager host is restarted

The following shows the statuses of jobnets and jobs for each start mode when a JP1/AJS3 service on the manager host is restarted. The Condition column in the table indicates any condition, such as the actual status after the JP1/AJS3 service stops or the status monitored by JP1/AJS3, that might cause the job and jobnet status to change.

n Statuses of jobnets and jobs when a hot-start is performed

The following table shows the statuses of jobnets and jobs when a JP1/AJS3 service on the manager host is hot-started.

Table 7-3 Jobnet and job statuses when a hot-start is performed (-hot)

Unit Status before JP1/AJS3 stops Condition Status after JP1/AJS3 starts
Jobnet Wait for prev. to end None Wait for prev. to end#1
Wait for start time (root job) None Wait for start time#2, #3
Wait for start time (nested job) None Wait for start time#1
Being held (root job) None Being held#2
Being held (nested job) None Being held#1
Now running None Now running#4
Running + Warning None Running + Warning#5
Running + Abend None Running + Abend
Now monitoring None Now monitoring
End status None No change
Remote jobnet Wait for prev. to end None Wait for prev. to end#1
Wait for start time (root job) None Wait for start time#2, #3
Wait for start time (nested job) None Wait for start time#1
Being held (root job) None Being held#2
Being held (nested job) None Being held#1
Now running (root job) None Ended abnormally
Now running (nested job) None Ended abnormally
End status None No change
Job (standard job, action job, custom job) Wait for prev. to end None Wait for prev. to end#1
Being held None Being held#1
Waiting to execute
  • The job is being queued (when the JP1/AJS3 database uses the standard configuration).
Wait for prev. to end#6

  • The job is being queued (when the JP1/AJS3 database uses the compatible ISAM configuration).
Now queuing

  • The JP1/AJS3 service stops before the job being executed on a remote host ends.
Now running

  • The JP1/AJS3 service stops before the job being executed on the local host ends.
Killed

  • The job ends.
End status#7

  • The job status can no longer be managed because the JP1/AJS3 service did not stop correctly due to a server failure.
  • An attempt to acquire the job status fails.
    - When the JP1/AJS3 service is restarted, the job status cannot be acquired due to insufficient memory or a communication error.
    - The job information has already been deleted because the number of days for saving job information expired.#8
Unknown end status
Now queuing
  • The job is being queued (when the JP1/AJS3 database uses the standard configuration).
Wait for prev. to end#6

  • The job is being queued (when the JP1/AJS3 database uses the compatible ISAM configuration).
Now queuing

  • The JP1/AJS3 service stops before the job being executed on a remote host ends.
Now running

  • The JP1/AJS3 service stops before the job being executed on the local host ends.
Killed

  • The job ends.
End status#7

  • The job status can no longer be managed because the JP1/AJS3 service did not stop correctly due to a server failure.
  • An attempt to acquire the job status fails.
    - When the JP1/AJS3 service is restarted, the job status cannot be acquired due to insufficient memory or a communication error.
    - The job information has already been deleted because the number of days for saving job information expired.#8
Unknown end status
Now running
  • The target host is a remote host.
Now running

  • The target host is the local host.
Killed

  • The job ends.
End status#7

  • The job status can no longer be managed because the JP1/AJS3 service did not stop correctly due to a server failure.
  • An attempt to acquire the job status fails.
    - When the JP1/AJS3 service is restarted, the job status cannot be acquired due to insufficient memory or a communication error.
    - The job information has already been deleted because the number of days for saving job information expired.#8
Unknown end status
End status None No change
Queueless job Wait for prev. to end None Wait for prev. to end#1
Being held None Being held#1
Waiting to execute
  • The job is waiting for another queueless job to end because the maximum number of concurrently executable jobs set for the queueless agent service has been exceeded.
Waiting to execute

  • The queueless job is running.
Now running

  • The queueless job ends.
End status#7, #9

  • When the JP1/AJS3 service is restarted, the job status cannot be acquired due to insufficient memory or a communication error.
  • Queueless job functionality is disabled when the JP1/AJS3 service restarts.
Unknown end status
Now running
  • The queueless job is running.
Now running

  • The queueless job ends.
End status#7, #9

  • When the JP1/AJS3 service is restarted, the job status cannot be acquired due to insufficient memory or a communication error.
  • Queueless job functionality has been uninstalled when the JP1/AJS3 service is started.
Unknown end status
End status None No change
Event job In a jobnet Wait for prev. to end None Wait for prev. to end#1
Being held None Being held#1
Now queuing, Now running
  • The target host is a remote host.
  • The target host is a remote host or the local host, and the system is restarted after a system failure.
Now queuing, Now running

  • The target host is the local host.
Ended abnormally#10
End status None No change
In start conditions Wait for prev. to end None Wait for prev. to end
Now queuing, Now running
  • The event job is running.
Now running

  • The event job ends.
End status#7
End status None No change
Jobnet connector Wait for prev. to end None Wait for prev. to end#1
Now running
  • The connection-destination jobnet ends normally.
Ended normally

  • The connection-destination jobnet ends with warning.
Ended with warning

  • The connection-destination jobnet is running.
Now running

  • The status of the connection-destination jobnet is Running + Warning.
Running + Warning

  • The connection-destination jobnet contains a unit that has ended abnormally.
Running + Abend
Running + Warning
  • The connection-destination jobnet ends with warning.
Ended with warning

  • The status of the connection-destination jobnet is Running + Warning.
Running + Warning

  • The connection-destination jobnet contains a unit that has ended abnormally.
Running + Abend
Running + Abend None Running + Abend
End status None No change

#1
If the preceding unit is handled as an abnormal end, the status is Not executed + Ended.

#2
If the hot start is performed after the jobnet timeout period expires, the status is Skipped so not exe.

#3
If the hot start is performed within the jobnet timeout period but later than the scheduled start time, the status is Now running.

#4
If the jobnet contains a unit that is handled as an abnormal end, the status is Running + Abend or Ended abnormally.
If the jobnet does not contain a unit that is handled as an abnormal end, but contains a unit whose status is Running + Warning or Ended with warning, the status is Running + Warning or Ended with warning.

#5
If the jobnet contains a unit that is handled as an abnormal end, the status is Running + Abend or Ended abnormally.

#6
When the JP1/AJS3 database uses the standard configuration, jobs to be submitted that have been queued by job execution control are canceled and returned to Wait for prev. to end status. These jobs, now in Wait for prev. to end status, are submitted and queued again by job execution control. Note, however, that if the previous unit ended abnormally, the jobs are not resubmitted, but are placed in Not executed + Ended status. For details, see 7.4 Resubmitting jobs when a JP1/AJS3 service is restarted.

#7
Jobs and jobnets can have the following end statuses:
  • Ended normally
  • Ended with warning
  • Ended abnormally
  • Killed
  • Failed to start

#8
This condition might occur when the JP1/AJS3 database uses the compatible ISAM configuration and the number of days for saving job information is set to 0. The job information necessary for managing the job status might have been deleted because the job information retention period (default: 10 minutes) expired before the JP1/AJS3 service was restarted.
If you set 0 for the number of days for saving job information, estimate the time required for the JP1/AJS3 service to start after it has stopped, and then specify a job information retention period greater than the estimated value. If an estimate is impossible, specify 1 or a greater value for the number of days for saving job information. For details about the number of days for saving job information, see 2.3 Setting up the job execution environment in the Job Management Partner 1/Automatic Job Management System 3 Configuration Guide 2.

#9
You need to create a status file for use with queueless jobs beforehand. If there is no status file, the job status is Unknown end status.

#10
If a failover occurs due to a system failure or if the option to continue execution of active event jobs is enabled, the status is Now running.

n Statuses of jobnets and jobs when a warm-start is performed

The following table shows the statuses of jobnets and jobs when the JP1/AJS3 service on the manager host is warm-started.

Table 7-4 Jobnet and job statuses when a warm-start is performed (-warm)

Unit Status before JP1/AJS3 stops Condition Status after JP1/AJS3 starts
Jobnet Wait for prev. to end
  • The root jobnet is in Wait for start time or Being held status.
No change

  • The root jobnet is in Now running status.#1
Not executed + Ended
Wait for start time (root job) None Wait for start time#2, #3
Wait for start time (nested job) None Not executed + Ended
Being held (root job) None Being held#2, #4
Being held (nested job) None Not executed + Ended
Now running None Interrupted#5
Running + Warning None Interrupted#5
Running + Abend None Interrupted#5
Now monitoring None Now monitoring
End status None No change
Remote jobnet Wait for prev. to end
  • The root jobnet is in Wait for start time or Being held status.
No change

  • The root jobnet is in Now running status.#1
Not executed + Ended
Wait for start time (root job) None Wait for start time#2, #3
Wait for start time (nested job) None Not executed + Ended
Being held (root job) None Being held#2, #4
Being held (nested job) None Not executed + Ended
Now running (root job) None Interrupted
Now running (nested job) None Interrupted
End status None No change
Job (standard job, action job, custom job) Wait for prev. to end
  • The root jobnet is in Wait for start time or Being held status.
No change

  • The root jobnet is in Now running status.#1
Not executed + Ended
Being held None Not executed + Ended
Waiting to execute
  • The job is being queued
Not executed + Ended

  • The job ends.
End status#6

  • The JP1/AJS3 service stops before the job enters the queuing status.
  • The job is running.
  • An attempt to acquire the job status fails.
    - When the JP1/AJS3 service is restarted, the job status cannot be acquired due to insufficient memory or a communication error.
    - The job information has already been deleted because the number of days for saving job information expired.#7
Unknown end status
Now queuing
  • The job is being queued.
Not executed + Ended

  • The job ends.
End status#6

  • The job is running.
  • The job information has already been deleted because the number of days for saving job information expired.#7
Unknown end status
Now running
  • The job is running.
  • The job information has already been deleted because the number of days for saving job information expired.#7
Unknown end status

  • The job ends.
End status#6
End status None No change
Queueless job Wait for prev. to end
  • The root jobnet is in Wait for start time or Being held status.
No change

  • The root jobnet is in Now running status.#1
Not executed + Ended
Being held None Not executed + Ended
Waiting to execute
  • The queueless job ends.
End status#6, #8

  • The job is waiting for another queueless job to end because the maximum number of concurrently executable jobs set for the queueless agent service has been exceeded.
Unknown end status
Now running
  • The queueless job ends.
End status#6, #8

  • The queueless job is running.
  • When the JP1/AJS3 service is restarted, the job status cannot be acquired due to insufficient memory or a communication error.
  • Queueless job functionality is disabled when the JP1/AJS3 service restarts.
Unknown end status
End status None No change
Event job In a jobnet Wait for prev. to end
  • The root jobnet is in Wait for start time or Being held status.
No change

  • The root jobnet is in Now running status.#1
Not executed + Ended
Being held None Not executed + Ended
Now queuing None Unknown end status
Now running None Unknown end status
End status None No change
In start conditions Wait for prev. to end
  • The root jobnet is in Wait for start time or Being held status.
No change

  • The root jobnet is in Now running status.#1
Not executed + Ended
Now queuing, Now running
  • The event job is running.
Now running

  • The event job ends.
End status#6
End status None No change
Jobnet connector. Wait for prev. to end
  • The root jobnet is in Wait for start time or Being held status.
No change

  • The root jobnet is in Now running status.#1
Not executed + Ended
Now running None Unknown end status
Running + Warning None Unknown end status
Running + Abend None Unknown end status
End status None No change

#1
Jobs and jobnets can have the following running statuses:
  • Now running
  • Running + Warning
  • Running + Abend

#2
If you specify plan in the OVERSCHEDULE environment setting parameter for the scheduler service, the status of jobnets scheduled to be executed on that day changes to Skipped so not exe. If you specify skip or select Execute from next time for Planned time passed when daemon starts in the Register for Execution dialog box when registering jobnets for execution, the status of jobnets scheduled to be executed before the scheduler service started changes to Skipped so not exe.
The jobnet status also becomes Skipped so not exe. if the scheduler service was started after the timeout period for the jobnet expired.

#3
If the warm start was performed later than the scheduled start time and the condition in #2 above does not apply, the jobnet status is Now running.

#4
To recalculate the schedule, the jobnet status changes to Wait for start time and then Being held.

#5
If the status of a job in the jobnet was Waiting to execute or Now running before the service stopped, its status changes to Interrupted after the execution results of all jobs are checked and status of the jobs changes to an end status.

#6
Jobs and jobnets can have the following end statuses:
  • Ended normally
  • Ended with warning
  • Ended abnormally
  • Killed
  • Failed to start

#7
This condition might occur when the JP1/AJS3 database uses the compatible ISAM configuration and the number of days for saving job information is set to 0. The job information necessary for managing the job status might have been deleted because the job information retention period (default: 10 minutes) expired before the JP1/AJS3 service was restated. If you set 0 for the number of days for saving job information, estimate the time required for the JP1/AJS3 service to start after it has stopped, and then specify a job information retention period greater than the estimated value. If an estimate is impossible, specify 1 or a greater value for the number of days for saving job information. For details about the number of days for saving job information, see 2.3 Setting up the job execution environment in the Job Management Partner 1/Automatic Job Management System 3 Configuration Guide 2.

#8
You need to create a status file for use with queueless jobs beforehand. If there is no status file, the job status is Unknown end status.

n Statuses of jobnets and jobs when a cold-start is performed

If you cold-start (-cold) a JP1/AJS3 service on the manager host, the status of all registered jobnets changes to Not registered, and the JP1/AJS3 service is started with all the previous execution results for jobnets and jobs deleted.

n Statuses of jobnets and jobs when the start mode is not specified

If you do not specify the start mode for the JP1/AJS3 service on the manager host, the JP1/AJS3 service will be started in the service start mode specified during setup. If a service mode is not specified during setup, operation is the same as when a warm start is specified.

n Statuses of jobnets and jobs when an event reset occurs

For details about the statuses of jobnets and jobs when an event reset occurs, see jpomanevreset in 2. Commands in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 1.

(b) Statuses when a JP1/AJS3 service on the agent host is restarted

The status of the jobs on the agent host when the JP1/AJS3 service on the agent host is restarted differs depending on the start mode and job type.

Note that if a start mode is not specified for the JP1/AJS3 service on the agent host, operation is the same as when a warm start is specified.

The following table describes how the job status differs according to the start mode for each job type.

Table 7-5 JP1/AJS3 start modes and the statuses of jobs on the agent host

Start mode PC job, Unix job, and QUEUE job Event job Event job specified in the start conditions
Warm start When the status of a job cannot be reported to the manager host, the status of the job on the manager host changes to Ended abnormally. However, if you use the -rs option in the jpqjobsub command to specify the job recovery status, the specified job status is set. The job is killed (the status of the job on the manager host changes to Ended abnormally).
However, if either of the following conditions exists, the status appropriate for the condition is set:
Of the event information detected by the agent host before JP1/AJS3 stopped, the event information that could not be reported to the manager host is reported to the manager host after the agent host is started (the status of the job on the manager host remains Now monitoring, and the job is executed when the agent host is restarted).
Hot start
Cold start The job is executed when the agent host is restarted.
Of the event information detected by the agent host before JP1/AJS3 stopped, the event information that could not be reported to the manager host is discarded (the status of the job on the manager host remains Now monitoring).

(4) Job statuses on the manager host when an agent host is restarted

When you restart a JP1/AJS3 service on the agent host in a manager and agent system configuration, the statuses of jobs on the manager host differ in the following cases:

Table 7-6 describes the statuses of jobs on the manager host when a stopped agent host is restarted. Table 7-7 describes the statuses of jobs on the manager host when an agent host that stopped due to a system failure is restarted.

Table 7-6 Statuses of jobs on the manager host when a stopped agent host is restarted

Unit type Status before the agent host stops Status after the agent host stops
Job#1 Wait for prev. to end Wait for prev. to end
Being held Being held
Now queuing Now queuing#2
Now running Killed#3
Other status No change
Queueless job Wait for prev. to end Not executed + Ended#4
Being held Being held#4
Waiting to execute Unknown end status#4
Now running Unknown end status#4
Other status No change
Event job In a jobnet Wait for prev. to end Wait for prev. to end
Being held Being held
Now queuing Now queuing
Now running Ended abnormally#5
Other status No change
In start conditions Wait for prev. to end Wait for prev. to end
Now queuing Now queuing
Now running Now running
Other status No change

#1
A job as used here is a standard job, action job, or custom job.

#2
If the agent host accepts a job execution request during shutdown processing, an attempt to start the job process might fail, in which case the job status changes to Failed to start. If the agent host stops before it sends a response to the job execution request from the manager host, the job execution request results in a timeout error. When the manager host then polls to check the status, the job status changes to Killed.

#3
For details about operation when a job is killed, see 4.5.10 Killing a job or jobnet in the manual Job Management Partner 1/Automatic Job Management System 3 Overview.

#4
After the agent host has started, some time might be required for the status to change.

#5
If either of the following conditions exists, the status appropriate for the condition is set.
  • If the option to continue execution of active event jobs is enabled, the status is Now running.
  • If either the scheduler service or a JP1/AJS3 service on the manager host has stopped:
    The status is set that is described in (3)(a) Statuses when a JP1/AJS3 service on the manager host is restarted.

    Table 7-7 Statuses of jobs on the manager host when an agent host that stopped due to a system failure is restarted

    Unit type Status before the agent host stops Status after the agent host stops
    Job#1 Wait for prev. to end Wait for prev. to end
    Being held Being held
    Now queuing Now queuing
    Now running Killed#2
    Other status No change
    Queueless job Wait for prev. to end Not executed + Ended#3
    Being held Being held#3
    Waiting to execute Unknown end status#3
    Now running Unknown end status#3
    Other status No change
    Event job In a jobnet Wait for prev. to end Wait for prev. to end
    Being held Being held
    Now queuing Now queuing
    Now running Now running
    Other status No change
    In start conditions Wait for prev. to end Wait for prev. to end
    Now queuing Now queuing
    Now running Now running
    Other status No change

#1
A job as used here is a standard job, action job, or custom job.

#2
For details about operation when a job is killed, see 4.5.10 Killing a job or jobnet in the manual Job Management Partner 1/Automatic Job Management System 3 Overview.

#3
It may take some time before the status is changed after the agent host has started.

[Contents][Back][Next]


[Trademarks]

Copyright (C) 2009, 2010, Hitachi, Ltd.
Copyright (C) 2009, 2010, Hitachi Solutions, Ltd.