Hitachi

JP1 Version 12 JP1/Automatic Job Management System 3 Configuration Guide


20.5.1 List of the environment setting parameters for the job execution environment

The job execution environment requires different settings in the manager environment (JP1/AJS3 - Manager) and the agent environment (JP1/AJS3 - Agent).

The table below lists the environment setting parameters to be used to set the job execution environment in the manager environment (JP1/AJS3 - Manager) and the agent environment (JP1/AJS3 - Agent).

The job execution control manager that is started is determined by the job type.

For which job execution manager type you need to specify an environment setting parameter, see the Applicable manager column.

Some environment setting parameters for scheduler services have two definition keys. One definition key is used to set a value common to all scheduler services (SCHEDULER is part of a key name). The other definition key is used to set a different value for each scheduler service (scheduler-service-name is part of a key name). If you specify both definition keys, the definition key for setting a different value for each scheduler service is valid.

For definition keys used to set a common value for the scheduler services, you will see the following text: For scheduler service (common). For definition keys used to set different values for the scheduler services, you will see the following text: For scheduler service (individual).

For definition keys for the job execution control manager for submit jobs, you will see the following text: For submit jobs.

To determine whether an environment setting parameter is to be set in Windows or UNIX, see the Applicable OS column. To determine whether an environment setting parameter is to be set in the manager environment (JP1/AJS3 - Manager) or the agent environment (JP1/AJS3 - Agent), see the Applicable product column.

Cautionary note on limiting resources in the job execution environment for UNIX

If you limit the resources for jobs executed from job execution control, the system might not be able to execute the jobs correctly, depending on the values specified in the environment setting parameters. For example, if you set 0 for the upper limit of the stack segment size, the system might not be able to execute jobs.

Therefore, make sure that you specify either limit values that have sufficient margin or do not specify limits. If you want to set a limit, first execute the ps command to check the segment size and CPU time, and then determine the values to be specified based on the data. Note, however, that even if you do not set limits, it does not mean there are no limits on resources. It only means that JP1/AJS3 does not explicitly set limits, and the default limits of the system (the one in which JP1/AJS3 was started) are in effect.

There are some environment setting parameters that are not subject to the OS limits.

If a value larger than the default limit is specified, the job execution user must have the appropriate permission (superuser). If the job execution user does not have the appropriate permission, the resource limit is not set when a job is executed and a warning message (KAVU3587-W) is output to the integrated trace log. That message reports that the job is executed with the default limit.

Limits placed on resources also apply to any child processes created by the job process (unless specific limits are set on a child process).

For AIX and Linux, although you use /etc/security/limits (In Linux, /etc/security/limits.conf) to define resource limits for the user who executes the job, those values do not take effect when a job is executed. In this case, you need to define resource limits for the user when JP1/AJS3 (root) is started.

For details, see the section corresponding to the type of job in the following manual:

  • For Unix jobs

    See 7.4 Notes on using Unix jobs in the JP1/Automatic Job Management System 3 System Design (Work Tasks) Guide.

  • For flexible jobs

    See 7.8 Notes on using flexible jobs in the JP1/Automatic Job Management System 3 System Design (Work Tasks) Guide.

  • For HTTP connection jobs

    See 7.9 Notes on using HTTP connection jobs in the JP1/Automatic Job Management System 3 System Design (Work Tasks) Guide.

Table 20‒4: Environment setting parameters (job execution environment setting)

No.

Definition key

Environment setting parameter

Definition

Applicable manager

Applicable OS

Applicable product

1

[{JP1_DEFAULT|logical-host-name}\JP1NBQMANAGER\Process]#

"WorkPath"=

Work directory name when a manager process is executed

C

Windows

UNIX

M

2

"IsHA"=

Definition for cluster environment setup for manager processes

C, S

Windows

UNIX

M

3

[{JP1_DEFAULT|logical-host-name}\JP1NBQMANAGER\Database]#

"DatabasePath"=

Name of the directory used to store the database

C

Windows

UNIX

M

4

"PreserveTerm"=

Number of days that job execution results are retained

C

Windows

UNIX

M

5

"StandardTime"=

Standard time for deleting job information

C

Windows

UNIX

M

6

"FlashOption"=

Data file update mode

C

Windows

UNIX

M

7

"DeleteFlashOption"=

Asynchronous option for job information deletion

C

Windows

UNIX

M

8

"PreserveMinute"=

Retention period for information when 0 is specified as the number of days for retaining job information

C

Windows

UNIX

M

9

"StartUpDelete"=

Option for suppressing job information deletion when the JP1/AJS3 service starts

C

Windows

UNIX

M

10

"ColdStartMode"=

Database initialization when the JP1/AJS3 service is cold-started

C

Windows

UNIX

M

11

[{JP1_DEFAULT|logical-host-name}\JP1NBQMANAGER\Queue]#

"MaximumQueue"=

Maximum number of definable queues

C

Windows

UNIX

M

12

  • For scheduler service (common):

    [{JP1_DEFAULT|logical-host-name}\JP1AJS2\SCHEDULER\QUEUE\MANAGER\Job]#

  • For scheduler service (individual):

    [{JP1_DEFAULT|logical-host-name}\JP1AJSMANAGER\scheduler-service-name\QUEUE\MANAGER\Job]#

  • For submit jobs:

    [{JP1_DEFAULT|logical-host-name}\JP1NBQMANAGER\Job]#

"MaximumContentJob"=

Maximum number of jobs in the system

C, S

Windows

UNIX

M

13

"AlartContentJob"=

Number of warning jobs in the system

C, S

Windows

UNIX

M

14

"AlarmContentJobEventInterval"=

Interval for issuing a JP1 event when the number of warning jobs in the system reaches the threshold

S

Windows

UNIX

M

15

"ReceiveFileSizeStatus"=

Behavior when the size of the job result file or transfer file has reached the upper limit

C, S

Windows

UNIX

M

16

"LimitReceiveFileSize"=

Upper limit on the size of the job result file or transfer file

C, S

Windows

UNIX

M

17

"ReceiveFileOption"=

Definition for preventing duplicated reception of the job result file

C, S

Windows

UNIX

M

18

"QueuingJobRecoveryTime"=

Length of time an agent waits for recovery from a failure occurring in a queuing job

C, S

Windows

UNIX

M

19

"ExecutingJobRecoveryTime"=

Length of time an agent waits for recovery from a failure occurring in a running job

C, S

Windows

UNIX

M

20

"MaximumExecJobLogOption"=

Setting for whether to output a message confirming that the maximum number of concurrently executable jobs has been reached

C, S

Windows

UNIX

M

21

"MaximumExecJobLogInterval"=

Interval for outputting a message confirming that the maximum number of concurrently executable jobs has been reached

C, S

Windows

UNIX

M

22

"PreventChangeJobOrder"=

Setting for preventing changes to the execution order of submit jobs

C

Windows

UNIX

M

23

"EnumJobSaveMemory"=

Setting for reducing the memory consumed by the job execution control manager when job information is enumerated

C

Windows

UNIX

M

24

"ExecutingJobChangeStatus"=

Setting for placing all jobs that are being executed in an end status when a communication error occurs

C, S

Windows

UNIX

M

25

  • For scheduler service (common):

    [{JP1_DEFAULT|logical-host-name}\JP1AJS2\SCHEDULER\QUEUE\MANAGER\Network]#

  • For scheduler service (individual):

    [{JP1_DEFAULT|logical-host-name}\JP1AJSMANAGER\scheduler-service-name\QUEUE\MANAGER\Network]#

  • For submit jobs:

    [{JP1_DEFAULT|logical-host-name}\JP1NBQMANAGER\Network]#

"ConnectTimeout"=

Timeout period for TCP/IP connections from the manager process for job execution control, to the status reporting process for job execution control

C, S

Windows

UNIX

M

26

"CommunicateRetryCount"=

Number of retries for TCP/IP connections from the manager process for job execution control, to the status reporting process for job execution control

C, S

Windows

UNIX

M

27

"CommunicateRetryInterval"=

Retry interval for TCP/IP connections from the manager process for job execution control, to the status reporting process for job execution control

C, S

Windows

UNIX

M

28

  • For scheduler service (common):

    [{JP1_DEFAULT|logical-host-name}\JP1AJS2\SCHEDULER\QUEUE\MANAGER\Agent]#

  • For scheduler service (individual):

    [{JP1_DEFAULT|logical-host-name}\JP1AJSMANAGER\scheduler-service-name\QUEUE\MANAGER\Agent] #

  • For submit jobs:

    [{JP1_DEFAULT|logical-host-name}\JP1NBQMANAGER\Agent] #

"MaximumAgent"=

Maximum number of definable agents

C

Windows

UNIX

M

29

"ObserveInterval"=

Agent monitoring interval

C, S

Windows

UNIX

M

30

"AutoCreate"=

Control flag for automatic agent definition

C, S

Windows

UNIX

M

31

"AutoCreateExecJobs"=

Maximum number of concurrently executable jobs when an agent is defined automatically

C, S

Windows

UNIX

M

32

"AutoCreateMaxJobs"=

Maximum number of jobs when an agent is defined automatically

C

Windows

UNIX

M

33

"AutoCreateWarnJobs"=

Warning value for the number of jobs when an agent is defined automatically

C

Windows

UNIX

M

34

"AutoCreatePriority"=

Agent priority when an agent is defined automatically

C

Windows

UNIX

M

35

"LeastRecentlyUsed"=

Definition of the method for determining the agent host to which jobs are distributed

C, S

Windows

UNIX

M

36

[{JP1_DEFAULT|logical-host-name}\JP1NBQMANAGER\Resource] #

"MaximumResource"=

Maximum number of definable execution-locked resources

C

Windows

UNIX

M

37

[{JP1_DEFAULT|logical-host-name}\JP1NBQAGENT\Process] #

"WorkPath"=

Name of the work directory when an agent process is executed

--

Windows

UNIX

M, A, ME

38

"IsHA"=

Definition for cluster environment step for agent processes

--

Windows

UNIX

M, A, ME

39

[{JP1_DEFAULT|logical-host-name}\JP1NBQAGENT\Job] #

"IsExecFindExecutable"=

Definition for executing a job with a long file name

--

Windows

M, A, ME

40

"LoadUserProfile"=

Definition for whether a user profile is loaded

--

Windows

M, A, ME

41

"IsAccessTokenCache"=

Setting for reusing an access token when a job is executed

--

Windows

M, A, ME

42

"ENABLE_LONG_EXT"=

Setting for whether to execute a job specified for an executable file whose extension is 4 bytes or more in an application associated with the executable-file extension

--

Windows

M, A, ME

43

"IsLimitCoreSize"=

Limit flag for the maximum core file size for each process

--

UNIX

M, A, ME

44

"LimitCoreSize"=

Maximum core file size for each process

--

UNIX

M, A, ME

45

"IsLimitDataSize"=

Limit flag for the maximum data segment size for each process

--

UNIX

M, A, ME

46

"LimitDataSize"=

Maximum data segment size for each process

--

UNIX

M, A, ME

47

"IsLimitFileSize"=

Limit flag for the maximum permanent file size for each process

--

UNIX

M, A, ME

48

"LimitFileSize"=

Maximum permanent file size for each process

--

UNIX

M, A, ME

49

"IsLimitMemorySize"=

Limit flag for the maximum memory size for each process

--

UNIX

M, A, ME

50

"LimitMemorySize"=

Maximum memory size for each process

--

UNIX

M, A, ME

51

"IsLimitStackSize"=

Limit flag for the maximum stack segment size for each process

--

UNIX

M, A, ME

52

"LimitStackSize"=

Maximum stack segment size for each process

--

UNIX

M, A, ME

53

"IsLimitCpuTime"=

Limit flag for the maximum CPU time for each process

--

UNIX

M, A, ME

54

"LimitCpuTime"=

Maximum CPU time for each process

--

UNIX

M, A, ME

55

"IsLimitWorkSize"=

Limit flag for the maximum working set size for each process

--

UNIX

M, A, ME

56

"LimitWorkSize"=

Maximum working set size for each process

--

UNIX

M, A, ME

57

"IsLimitNoFile"=

Limit flag for the maximum number of file descriptors for each process

--

UNIX

M, A, ME

58

"LimitNoFile"=

Maximum number of file descriptors for each process

--

UNIX

M, A, ME

59

"PipeComTimeout"=

Definition of the pipe communication timeout value for agents

--

UNIX

M, A, ME

60

"AfterForkPipeTimeout"=

Definition of the timeout value for the job process execution ready notification

--

UNIX

M, A, ME

61

"IsCacheCmdSearchPath"=

Flag for specifying whether to cache the value in the command search path

--

UNIX

M, A, ME

62

"IsUseWorkPath"=

Flag for specifying whether to change the work (current) path when the job is executed

--

UNIX

M, A, ME

63

"IsSigchldDefault"=

Definition for changing the setting of child processes started by JP1/AJS3 to the SIG_DFL setting whereby the SIGCHLD signal is received

--

UNIX

M, A, ME

64

"FileAccessPermissionMode"=

Setting of the method to check file access permissions when a job is executed on a UNIX host

--

UNIX

M, A, ME

65

[{JP1_DEFAULT|logical-host-name}\JP1NBQAGENT\Network] #

"NotfyJobStateInterval"=

Interval for resending the job result file

--

Windows

UNIX

M, A, ME

66

"NotfyJobStateCount"=

Number of times that the job result file is resent

--

Windows

UNIX

M, A, ME

67

"ConnectTimeout"=

Definition of the TCP/IP connection timeout value on the job execution control agent

--

Windows

UNIX

M, A, ME

68

"CommunicateRetryCount"=

Definition of the number of TCP/IP connection error retries on the job execution control agent

--

Windows

UNIX

M, A, ME

69

"CommunicateRetryInterval"=

Definition of the TCP/IP connection error retry interval on the job execution control agent

--

Windows

UNIX

M, A, ME

70

"LogonUserRetryCount"=

Number of retries for obtaining access tokens when Standard is specified for Exec. service

--

Windows

M, A, ME

71

"LogonUserRetryInterval"=

Retry interval for obtaining access tokens when Standard is specified for Exec. service

--

Windows

M, A, ME

72

[{JP1_DEFAULT|logical-host-name}\JP1NBQAGENT\Variable] #

"variable-name"=

Installation directory of an execution program indicated by the variable name on each applicable host

--

Windows

UNIX

M, A, ME

73

[{JP1_DEFAULT|logical-host-name}\JP1NBQCLIENT\PathEnv] #

"All Users"=

Search path for a transfer file

--

Windows

UNIX

M, A, ME

74

"JP1-user-name"=

Search path for a transfer file

--

Windows

UNIX

M, A, ME

75

[{JP1_DEFAULT|logical-host-name}\JP1NBQCLIENT\Network] #

"ConnectTimeout"=

Definition of the TCP/IP connection timeout value for the commands and schedulers used to execute jobs

--

Windows

UNIX

M, A, ME

76

"CommunicateRetryCount"=

Definition of the number TCP/IP connection error retries for the commands and schedulers used to execute jobs

--

Windows

UNIX

M, A, ME

77

"CommunicateRetryInterval"=

Definition of the TCP/IP connection error retry interval for the commands and schedulers used to execute jobs

--

Windows

UNIX

M, A, ME

78

"CommunicateTimeout"=

Connection timeout period for the manager process of job execution control

--

Windows

UNIX

M, A, ME

79

[{JP1_DEFAULT|logical-host-name}\JP1NBQCLIENT\Process] #

"MacroOptionReplaceMode"=

Setting for whether to replace the passing information in a macro variable with the NULL character string

--

Windows

UNIX

M

80

"IsHA"=

Definition for cluster environment setup for the commands and schedulers used to execute jobs

--

Windows

UNIX

M, A, ME

81

"PassingInfoUsePreRc"=

Settings for using the return value of the preceding job in the passing information setting jobs

--

Windows

UNIX

M

82

  • For scheduler service (common)

    [{JP1_DEFAULT|logical-host-name}\JP1AJS2\SCHEDULER\QUEUE\NOTIFY\Network] #

  • For scheduler service (individual)

    [{JP1_DEFAULT|logical-host-name}\JP1AJSMANAGER\scheduler-service-name\QUEUE\NOTIFY\Network] #

"ConnectTimeout"=

Definition of the TCP/IP connection timeout value in the status reporting process for job execution control

--

Windows

UNIX

M

83

"CommunicateRetryCount"=

Definition of the number TCP/IP connection error retries in the status reporting process for job execution control

--

Windows

UNIX

M

84

"CommunicateRetryInterval"=

Definition of the TCP/IP connection error retry interval in the status reporting process for job execution control

--

Windows

UNIX

M

85

[{JP1_DEFAULT|logical-host-name}\JP1NBQNOTIFY\Process] #

"IsHA"=

Definition for cluster environment setup for the status reporting process for job execution control

--

Windows

UNIX

M, A, ME

86

[{JP1_DEFAULT|logical-host-name}\JP1NBQAGENT\Job] #

"MergeEnvMode"=

Definition for whether an environment variable in a job definition is enabled when the name is the same as a system environment variable name

--

Windows

M, A, ME

87

"UACAdministratorsExec"=

Setting for executing jobs as a user with administrator permissions when UAC is enabled

--

Windows

M, A, ME

88

"GetPasswd_RetryCount"=

Number of retry attempts when an attempt to acquire user information from the OS password file has failed

--

UNIX

M, A, ME

89

"GetPasswd_RetryInterval"=

Retry interval when an attempt to acquire user information from the OS password file has failed

--

UNIX

M, A, ME

90

"IgnoreResultFileError"=

Operation if an error occurs during transfer processing of a job result file and Standard is specified for Exec. Service

--

Windows

UNIX

M, A, ME

91

"PreventResultFileLack"=

Setting for ensuring that data generated by jobs is output to result files

--

Windows

M, A, ME

92

"SendFileSizeStatus"=

Behavior when the total size of job result files reaches the upper limit

--

Windows

UNIX

M, A, ME

93

"LimitSendFileSize"=

Upper limit for the size of job result files

--

Windows

UNIX

M, A, ME

94

[JP1_DEFAULT\JP1NBQAGENT\Log]

"JobHistoryLogFileSize"=

Size of the job execution history log file

--

Windows

UNIX

ME

95

"JobHistoryLogFileCount"=

Number of the job execution history log files

--

Windows

UNIX

ME

Legend:

C: Job execution control manager for submit jobs

S: Job execution control manager for scheduler services

M: JP1/AJS3 - Manager

A: JP1/AJS3 - Agent

ME: JP1/AJS3 - Agent Minimal Edition

--: Not applicable

#

Specify JP1_DEFAULT for the physical host. For logical-host-name, specify the name of a logical host.

Parameters that are not described here are used for maintenance purposes. Do not change their values.