GenericWorkflowJob¶
-
class
lsst.ctrl.bps.
GenericWorkflowJob
(name: str)¶ Bases:
object
Information about a job that may be needed by various workflow management services.
Attributes Summary
abort_on_value
Job exit value for signals to abort the entire workflow. abort_return_value
Exit value to use when aborting the entire workflow. arguments
Command line arguments for job. attrs
Key/value pairs of job attributes (for WMS that have attributes in addition to commands). category
WMS-facing label of job within single workflow (e.g., can be used for throttling jobs within a single workflow). cmdvals
Values for variables in cmdline when using lazy command line creation. compute_site
Key to look up site-specific information for running the job. concurrency_limit
Names of concurrency limits that the WMS plugin can appropriately translate to limit the number of this job across all running workflows. environment
Environment variable names and values to be explicitly set inside job. executable
Executable for job. label
Primary user-facing label for job. mail_to
Comma separated list of email addresses for emailing job status. memory_multiplier
Memory growth rate between retries. name
Name of job. number_of_retries
Number of times to automatically retry a failed job. post_cmdline
Command line to be executed after job executes. pre_cmdline
Command line to be executed prior to executing job. priority
Initial priority of job in WMS-format. profile
Nested dictionary of WMS-specific key/value pairs with primary key being WMS key (e.g., pegasus, condor, panda). queue
Name of queue to use. request_cpus
Max number of cpus that the job is expected to need. request_disk
Max amount of job scratch disk (in MB) that the job is expected to need. request_memory
Max memory (in MB) that the job is expected to need. request_walltime
Max amount of time (in seconds) that the job is expected to need. retry_unless_exit
Exit code for job that means to not automatically retry. tags
Other key/value pairs for job that user may want to use as a filter. when_to_mail
WMS-specific terminology for when to email job status. Attributes Documentation
-
abort_on_value
¶ Job exit value for signals to abort the entire workflow.
-
abort_return_value
¶ Exit value to use when aborting the entire workflow.
-
arguments
¶ Command line arguments for job.
-
attrs
¶ Key/value pairs of job attributes (for WMS that have attributes in addition to commands).
-
category
¶ WMS-facing label of job within single workflow (e.g., can be used for throttling jobs within a single workflow).
-
cmdvals
¶ Values for variables in cmdline when using lazy command line creation.
-
compute_site
¶ Key to look up site-specific information for running the job.
-
concurrency_limit
¶ Names of concurrency limits that the WMS plugin can appropriately translate to limit the number of this job across all running workflows.
-
environment
¶ Environment variable names and values to be explicitly set inside job.
-
executable
¶ Executable for job.
-
label
¶ Primary user-facing label for job. Does not need to be unique and may be used for summary reports.
-
mail_to
¶ Comma separated list of email addresses for emailing job status.
-
memory_multiplier
¶ Memory growth rate between retries.
-
name
¶ Name of job. Must be unique within workflow.
-
number_of_retries
¶ Number of times to automatically retry a failed job.
-
post_cmdline
¶ Command line to be executed after job executes.
Should be executed regardless of exit status.
-
pre_cmdline
¶ Command line to be executed prior to executing job.
-
priority
¶ Initial priority of job in WMS-format.
-
profile
¶ Nested dictionary of WMS-specific key/value pairs with primary key being WMS key (e.g., pegasus, condor, panda).
-
queue
¶ Name of queue to use. Different WMS can translate this concept differently.
-
request_cpus
¶ Max number of cpus that the job is expected to need.
-
request_disk
¶ Max amount of job scratch disk (in MB) that the job is expected to need.
-
request_memory
¶ Max memory (in MB) that the job is expected to need.
-
request_walltime
¶ Max amount of time (in seconds) that the job is expected to need.
-
retry_unless_exit
¶ Exit code for job that means to not automatically retry.
Other key/value pairs for job that user may want to use as a filter.
-
when_to_mail
¶ WMS-specific terminology for when to email job status.
-