JobStep

Note

This supersedes the pyslurm.jobstep class, which will be removed in a future release

pyslurm.JobStep

A Slurm Jobstep

Parameters:

Name	Type	Description	Default
job_id	Union[Job, int]	The Job this Step belongs to.	0
step_id	Union[int, str]	Step-ID for this JobStep object.	0

Other Parameters:

Name	Type	Description
time_limit	int	Time limit in Minutes for this step.

Attributes:

Name	Type	Description
stats	JobStepStatistics	Real-time statistics of a Step. Before you can access the stats data for a Step, you have to call the load_stats method of a Step instance or the Jobs collection.
pids	dict[str, list]	Current Process-IDs of the Step, organized by node name. Before you can access the pids data, you have to call the load_stats method of a Srep instance or the Jobs collection.
id	Union[str, int]	The id for this step.
job_id	int	The id for the Job this step belongs to.
name	str	Name of the step.
user_id	int	User ID who owns this step.
user_name	str	Name of the User who owns this step.
time_limit	int	Time limit in Minutes for this step.
network	str	Network specification for the step.
cpu_frequency_min	Union[str, int]	Minimum CPU-Frequency requested.
cpu_frequency_max	Union[str, int]	Maximum CPU-Frequency requested.
cpu_frequency_governor	Union[str, int]	CPU-Frequency Governor requested.
reserved_ports	str	Reserved ports for the step.
cluster	str	Name of the cluster this step runs on.
srun_host	str	Name of the host srun was executed on.
srun_process_id	int	Process ID of the srun command.
container	str	Path to the container OCI.
allocated_nodes	str	Nodes the Job is using.
start_time	int	Time this step started, as unix timestamp.
run_time	int	Seconds this step has been running for.
run_time_remaining	int	The amount of seconds the step has still left until hitting the time_limit.
elapsed_cpu_time	int	Amount of CPU-Time used by the step so far. This is the result of multiplying the run_time with the amount of cpus allocated.
partition	str	Name of the partition this step runs in.
state	str	State the step is in.
cpus	int	Number of CPUs this step uses in total.
ntasks	int	Number of tasks this step uses.
distribution	dict	Task distribution specification for the step.
command	str	Command that was specified with srun.
slurm_protocol_version	int	Slurm protocol version in use.

cancel() method descriptor

Cancel a Job step.

Implements the slurm_kill_job_step RPC.

Raises:

Type	Description
RPCError	When cancelling the Job was not successful.

Examples:

>>> import pyslurm
>>> pyslurm.JobStep(9999, 1).cancel()

load(job_id, step_id) staticmethod

Load information for a specific job step.

Implements the slurm_get_job_steps RPC.

Parameters:

Name	Type	Description	Default
job_id	Union[Job, int]	ID of the Job the Step belongs to.	required
step_id	Union[int, str]	Step-ID for the Step to be loaded.	required

Returns:

Type	Description
JobStep	Returns a new JobStep instance

Raises:

Type	Description
RPCError	When retrieving Step information from the slurmctld was not successful.

Examples:

>>> import pyslurm
>>> jobstep = pyslurm.JobStep.load(9999, 1)

load_stats() method descriptor

Load realtime stats for this Step.

Calling this function returns the live statistics of the step, and additionally populates the stats and pids attribute of the instance.

Returns:

Type	Description
JobStepStatistics	The statistics of the Step.

Raises:

Type	Description
RPCError	When retrieving the stats for the Step failed.

Examples:

>>> import pyslurm
>>> step = pyslurm.JobStep.load(9999, 1)
>>> stats = step.load_stats()
>>>
>>> # Print the CPU Time Used
>>> print(stats.total_cpu_time)
>>>
>>> # Print the Process-IDs for the Step, organized by hostname
>>> print(step.pids)

modify(changes) method descriptor

Modify a job step.

Implements the slurm_update_step RPC.

Parameters:

Name	Type	Description	Default
changes	JobStep	Another JobStep object that contains all the changes to apply. Check the Other Parameters of the JobStep class to see which properties can be modified.	required

Raises:

Type	Description
RPCError	When updating the JobStep was not successful.

Examples:

>>> import pyslurm
>>>
>>> # Setting the new time-limit to 20 days
>>> changes = pyslurm.JobStep(time_limit="20-00:00:00")
>>> pyslurm.JobStep(9999, 1).modify(changes)

send_signal(signal) method descriptor

Send a signal to a running Job step.

Implements the slurm_signal_job_step RPC.

Parameters:

Name	Type	Description	Default
signal	Union[str, int]	Any valid signal which will be sent to the Job. Can be either a str like SIGUSR1, or simply an int.	required

Raises:

Type	Description
RPCError	When sending the signal was not successful.

Examples:

Specifying the signal as a string:

>>> import pyslurm
>>> pyslurm.JobStep(9999, 1).send_signal("SIGUSR1")

or passing in a numeric signal:

>>> pyslurm.JobStep(9999, 1).send_signal(9)

to_dict() method descriptor

JobStep information formatted as a dictionary.

Returns:

Type	Description
dict	JobStep information as dict

pyslurm.JobSteps

Bases: builtins.dict

A dict of pyslurm.JobStep objects for a given Job.

Raises:

Type	Description
RPCError	When getting the Job steps from the slurmctld failed.

load(job) staticmethod

Load the Job Steps from the system.

Parameters:

Name	Type	Description	Default
job	Union[Job, int]	The Job for which the Steps should be loaded.	required

Returns:

Type	Description
JobSteps	JobSteps of the Job

Examples:

>>> import pyslurm
>>> steps = pyslurm.JobSteps.load(1)
>>> print(steps)
pyslurm.JobSteps({'batch': pyslurm.JobStep('batch')})
>>> print(steps[1])
pyslurm.JobStep('batch')

load_all() staticmethod

Loads all the steps in the system.

Returns:

Type	Description
dict	A dict where every JobID (key) is mapped with an instance of its JobSteps (value).