ICC Home Privacy and Legal Notice LC User Documents Banner

UCRL-WEB-201386

SLURM Reference Manual


SCONTROL (Manage Configurations)

ROLE.
SCONTROL is the SLURM utility that manages SLURM's own configuration, including the properties that it assigns to nodes, node partitions, and other SLURM-controlled system features. Most SCONTROL options and commands are intended for, and can only be successfully executed by, a system administrator (a privileged or root user). Some SCONTROL commands report useful configuration information or manage job checkpoints, however, and any user can benefit from invoking them appropriately. The rest of this section discusses only those few general-user SCONTROL commands.

EXECUTE LINE.
Run SCONTROL by typing

scontrol command argument

where the general-user commands and their specific arguments are summarized below (most commands are for system administrators only). Without any command, SCONTROL prompts for input.

GENERAL-USER COMMANDS.
The subset of SCONTROL commands that any user can invoke includes these:

checkpoint action jobid[.stepid]
requests one of several allowed checkpoint activities on those LC AIX machines that also use SLURM (instead of IBM's LoadLeveler). You must (previously) set environment variable CHECKPOINT to YES and specify a location and name for future checkpoint files. See the "Checkpointing with SLURM and POE" section of the LCRM Reference Manual for details, especially for LCRM batch jobs. Here:
action
specifies what to do after the requested checkpoint occurs, where the three most useful alternatives are:
create
requests a checkpoint and continues the job(step) after it occurs.
vacate
requests a checkpoint and terminates the job(step) after it occurs.
restart
resumes execution of a previously checkpointed job(step).
jobid[.stepid]
specifies the range for the checkpointing activity, which can be all existing steps for a specified jobid alone (e.g., 4812), or the individual job step indicated by a jobid.stepid combination (e.g., 4812.4).
exit
(or QUIT) terminates SCONTROL during interactive sessions.
show entity id
displays the current state of the SLURM-managed item that you specify, where
entity
can be any of these alternative literal strings:
config [see "Scheduler Types" above]
daemons
job
node
partition
step
id
specifies which individual entity to report (for example, by providing a node name (e.g., mcr123), a partition name (e.g., pdebug), or a job ID number (e.g., 1428)).


Navigation Links: [ Document List ] [ HPC Home ] [ Previous ] [ Next ]