SCHEDULE
Automated Job Submission System
Guide and Reference Manual


Previous Contents Index

It is recommended that resources be listed in the RESOURCE list using the editor as opposed to this method.

-process=account:text

The ACCOUNT field will specify which account to charge the batch job to when submitted. This is an eight-character field. The privilege CMKRNL is required to use this option .

-process=characteristics:number

The CHARACTERISTICS field will specify the number of a characteristic to be associated with the job.

-process=cli:name

The SHELL field specifies the shell to be used to process the job. Selections include C shell, Korn shell and Bourne shell. .

-process=cputime:delta_time

CPUTIME will define a CPU time limit for the batch job. Time can be specified as delta time, or as 0 or NONE (no limit on the time). If the queue on which the job will execute has a defined CPUMAXIMUM value, the smaller of the PROCESS command and queue values is used.

-process=delete

-process=nodelete

The DELETE field will control whether or not command files are deleted after processing. The default setting is NODELETE.

-process=hold

-process=nohold

The HOLD field will place the job into a queue in the HOLD state. The default setting is NOHOLD.

-process=keep

-process=nokeep

The KEEP field will indicate that the log file produced by the execution of the job is to be kept and not deleted. The default setting is KEEP.

-process=log_file:file_spec

-process=nolog_file

The LOG_FILE field specifies the name of the job log file. By default the file is placed in the user's login directory. The default name is the job name and the default type is .LOG.

-process=name:text

The NAME field allows a different name (from that of the job itself) to be used when the job is entered into an execution queue.

-process=node:node_name

The NODE field indicates which node is to process the job. In a cluster, it is assumed that any node can process a job for execution on behalf of any cluster member.

-process=parameter:(p1,p2,...p8)

The PARAMETER datum will provide the values of up to eight optional parameters (p1,p2,...p8) for each command procedure in the job. The symbols are local to the specified command procedure. Characters other than alphanumeric characters must be included in quotation marks.

-process=printer:queue_name

-process=noprinter

The PRINTER datum will queue the job log file for printing when the job is completed. PRINTER allows the user to specify a particular print queue. The default print queue is SYS$PRINT.

-process=priority:queue_priority

The PRIORITY field will specify what queue priority level to use when entering the job into an execution queue.

-process=queue:queue_name

The QUEUE field will identify the batch queue on which the job will be entered. The default queue is SYS$BATCH.

-process=restart

-process=norestart

The RESTART field will mark the created job as being restartable. Thus, if the job is restarted during execution, the environment variable $RESTART is set to true. Then the user provided command procedure will check this value and continue execution as needed. The default setting is NORESTART.

-process=retain

-process=noretain

The RETAIN flag determines whether the job will be retained if delete_upon_completion is ON and the job fails. The default setting is NORETAIN.

-process=user:username

The USER field indicates on who's behalf the job is to be processted for execution. Normally this user name is the same as person who owns the job. To change this field requires CMKRNL privilege.

-process=variable:(var1=value, var2=value etc.)

The VARIABLE qualifier creates or modifies one or more existing job variables. If the job variable exists in the Notes cosection of the job it will be modified, if not it will be created.

-process=wsdefault:number

The WSDEFAULT field will define for a batch job a working set default, the default number of physical pages that the job can use. If the queue on which the job is executed has a nonzero working set default, the smaller of the specified job and queue values is used. If the default set is 0, the smaller of the specified job and user authorization file (UAF) values is used.

-process=wsextent:number

The WSEXTENT field will define for a batch job a working set extent, the maximum amount of physical memory that the job can use. The job uses the maximum amount of physical memory only when the system has excess free pages. If the queue on which the job is executed has a nonzero working set extent, the smaller of the specified job and queue values is used. If the default set is 0, the smaller of the specified job and user authorization file (UAF) values is used.

-process=wsquota:number

The WSQUOTA field will define for a batch job a working set quota, the amount of physical memory that the job is guaranteed. If the queue on which the job is executed has a nonzero working set quota, the smaller of the specified job and queue values is used. If the default set is 0, the smaller of the specified job and user authorization file (UAF) values is used.

more

Type a list.

The more command will type out the various list sections that can be associated with a job definition.


General Format

more [-options] job_name

Alternate

type -job [-options] job_name

option summary

general options:

-[no]log
-[no]confirm
-hostname = target_node
section identifiers:

commands (default)
-prerequisites
-initiates
-batchlog
-distribution
-notices
-holds_up
-activated_by
-resources [-check, -expanded, -full]
-runtimes[-repeat=n]
-endactions
-all
other features:

-env[=prefix_string,format] (csh)

Examples

#1

 
 
 
 Schedule> more incremental -prerequisites 
 
 /pclark/incremental/prerequisites 
 
  /pclark/inca 
  /pclark/incb 
 
      

#2

 
 
 Schedule> more /demo/a/a1 -env 
 
   sd == "@sys$login:sd" 
   sd_0001 = "$ wait 'p1'" 
   sd_0002 = "" 
   sd_count = "2" 
 
 
      


Options

-activated_by

The "ACTIVATED_BY list" is the list of jobs that activate the job being edited.

-all

The ALL option selects that all the lists are to be type out. That is the command, prerequisite, initiates, distribution, notices, holds_up and activated_by lists.

-batchlog

The batchlog qualifier types out the log file for the job. The location of the log file is designated by the log file location Use in conjunction with -tail=n to set the number of lines from the bottom of the log file to display, -head=n to set the number of lines from the top of the log file to display -maximum=n to set the maximum size of the log. The -file=n qualifier will designate the version of the log to be viewed, 0 equalling the curreent version and 1...n equalling each version previous.

-commands

The COMMANDS option is the list of script commands that will be submitted in the form of a command file.

-confirm

-noconfirm

CONFIRM requests that the user be prompted prior to any operation to verify that the selected entry is the correct one.

-node=target_node

The hostname option specifies what hostname (either local or remote) you want the operation performed on. -hostname may be substituted for -node.

-distribution

The "DISTRIBUTION list" is a list of PRINT commands to be issued. The idea is to provide a central mechanism for maintaining the how many, where to, and what printer information in a common location separate from the job command stream.

endactions

The ENDACTIONS listing is a list of resource adjustments that are to be run upon completion of the job.

-head

The head qualifier determines how many lines of the top of the log file will be generated using the -batchlog qualifier. The default is 20. It can also be used with -tail=n and -maximum=n .

-holds_up

The "HOLDS_UP list" is a list of jobs that are held up by the job being edited.

-initiates

The "INITIATES list" is the list of other jobs to start up when this one completes successfully.

-log

-nolog

The LOG qualifier causes each completed operation to issue a note to the user.

-maximum

The maximum qualifier determines the maximum number of lines a log file may be between the head and tail when using the -batchlog qualifier. Default is 2000. It can also be used with -head=n and -tail=n .

-notices

The "NOTICES list" is a free format text entry that is used by the NOTIFICATION procedure to augment the various messages that it sends.

-prerequisites

The "PREREQUISITES list" is the list of other jobs that must be successfully completed before this one is started.

-runtimes[-repeat=n]

The RUNTIMES qualifier types out future run times for the job. The default number of run times is 5. To modify the number of run times displayed use the -repeat=n qualifier.

-resources [-check, -expanded, -full]

Displays a list of generalized resources. The resource list is where the bulk of advanced job programming occurs. Here IF-THEN or SET statements are processed that can check or alter the value of a variable, supress a job run or cause a wait to occur.

The CHECK qualifier performs a syntax check on the resource list to ensure the correct IF-THEN and set statements can take effect. The EXPANDED qualifier used in conjunction with CHECK will display expanded information about the current status of the resource list. The FULL qualifier will give a complete listing of resource information

-env[=prefix_string] or [=(prefix_string,format)]

The environmental variable option requests that all job information be output to a file containing a list of enviroment variables. Every field in the control record is listed in the file as an environment variable. The environment variable name is constructed by prefixing the field name with the provide string. If none is provided a default value of SC is used.

The format defines the way that environmental variables are output. The default output specification is: "$export %s=\" %s\" "" (ksh), "$setenv %s=\" %s\" "" (csh) which produces the following format style.


 
   SC_ACCOUNT = "DEVELOPM" 
 

Alternate output styles may be created by changing the format field.

-tail

The tail qualifier determines how many lines of the bottom of the log file will be generated using the -batchlog qualifier. The default is 40. It can also be used with -head=n and -maximum=n .

mv (or mvjob)

Rename a job.

The mv command will change the name of a scheduled job definition. All references to a job are also revised to point to this new job definition. It is possible to rename across directories.


General Format

mv [-options] old_job_name new_job_name

Alternate

rename -job [-options] old_job_name new_job_name

option summary

general options:

-[no]log
-[no]confirm
-node = target_node

Examples

#1

 
 
 
 Schedule> mv test daily_incremental -log 
 Renamed /ajackson/test to /ajackson/daily_incremental 
 
 
      


Options

-confirm

-noconfirm

CONFIRM requests that the user be prompted prior to any operation to verify that the selected entry is the correct one.

-node=target_node

The hostname option specifies what hostname (either local or remote) you want the operation performed on. -hostname may be substituted for -node.

-log

-nolog

The LOG qualifier causes each completed operation to issue a note to the user.

rebjob

Rebuild the job-to-job connections.

rebjob examines job-to-job connections and verifies and corrects any that are broken or missing. The links that are examined are the PREREQUISITES list, INITIATES list, ACTIVATED_BY list, and HOLDS_UP list.

The INITIATES list and the ACTIVATED_BY list are the inverse of one another. When one list is edited the other list is updated automatically. The same relationship exists for the PREREQUISITES list and the HOLDS_UP list.

In general, this command is not needed since the job-to-job connections are rebuilt automatically when the lists are modified. However, if a job did not exist when it was entered into a list, then at the time when the job is created the REBUILD command should be used.


General Format

rebjob [-options] job_name

Alternate

rebuild -job [-options] job_name

option summary

-[no]log
-[no]confirm
-[no]node=target_node

Examples

#1

 
 Schedule> rebjob incremental -log 
      


Options

-confirm

-noconfirm

CONFIRM requests that the user be prompted prior to any operation to verify that the selected entry is the correct one.

-node=target_node

The hostname option specifies what hostname (either local or remote) you want the operation performed on. -hostname may be substituted for -node.

-log

-nolog

The LOG qualifier causes each completed operation to issue a note to the user.

rm (or rmjob)

Delete a job schedule.

The rm command permanently removes a job definition and all references to this job from the on-line database. This includes any mention to this job on the prerequisite or initiate list of any other job.


General Format

rmjob [-options] job_name

Alternate

delete -job [-options] job_name

option summary

-[no]log
-[no]confirm
-[no]node=target_node

Examples

#1

 
 
 
 Schedule> rmjob abc* -confirm 
 Delete /smith/abcd [Y/N] ? (Y): Y 
 Delete /smith/abce [Y/N] ? (Y): Y 
 Delete /smith/abcf [Y/N] ? (Y): N 
 Delete /smith/abcg [Y/N] ? (Y): Y 
 
 
      


Options

-confirm

-noconfirm

CONFIRM requests that the user be prompted prior to any operation to verify that the selected entry is the correct one.

-node=target_node

The hostname option specifies what hostname (either local or remote) you want the operation performed on. -hostname may be substituted for -node.

-log

-nolog

The LOG qualifier causes each completed operation to issue a note to the user.

sigjob

Signal a scheduling event.

The sigjob command sends a signal to the SCHEDULE SERVER. This command is automatically added to the temporary command files created when ever a job is submitted into an execution queue. Once at the beginning of job to signal that the job has started and once at the end to signal that the job has completed.3 The signal operation is optional. If a job terminates (via a crash or other abrupt mechanism) this event is picked up and becomes the completion signal.


General Format

sigjob [-options] job_name

Alternate

signal -job [-options] job_name

option summary

type of signal:

-completed (default)
-started

other options:

-status = number (default is 1)
-set_id = number (default of 0 implies all job sets)

general options:

-[no]log
-[no]confirm
-node = target_node

Examples

#1

 
 
 
 Schedule> sigjob first_of_three_jobs -started -log 
 signaled /bmossberg/first_of_three_jobs 
 
 
      


Options

-confirm

-noconfirm

CONFIRM requests that the user be prompted prior to any operation to verify that the selected entry is the correct one.

-node=target_node

The hostname option specifies what hostname (either local or remote) you want the operation performed on. -hostname may be substituted for -node.

-completed

The COMPLETED option sends a signal to the server that a job has successfully completed.

-log

-nolog

The LOG qualifier causes each completed operation to issue a note to the user.

-set_id

The SET_ID option points to a designated job set. If no number is specified the default implies all job sets.

-started

The STARTED option sends a signal to the server that a job is being executed.

-status

The STATUS option indicates the success (odd) or failure (even) of a job execution.

Note

3 This command replaces the earlier SCHEDULE COMPLETED command. This earlier command is still accepted and is a synonom for the SIGNAL command.


Previous Next Contents Index