SACCT(1) Slurm components SACCT(1) NAME sacct - displays accounting data for all jobs and job steps in the SLURM job accounting log or SLURM database SYNOPSIS sacct [OPTIONS...] DESCRIPTION Accounting information for jobs invoked with SLURM are either logged in the job accounting log file or saved to the SLURM database. The sacct command displays job accounting data stored in the job accounting log file or SLURM database in a variety of forms for your analysis. The sacct command displays infor- mation on jobs, job steps, status, and exitcodes by default. You can tailor the output with the use of the --format= option to specify the fields to be shown. For the root user, the sacct command displays job accounting data for all users, although there are options to filter the output to report only the jobs from a specified user or group. For the non-root user, the sacct command limits the display of job accounting data to jobs that were launched with their own user identifier (UID) by default. Data for other users can be displayed with the --all, --user, or --uid options. Note: If the AccountingStorageType is set to "accounting_storage/filetxt", space charac- ters embedded within account names, job names, and step names will be replaced by underscores. If account names with embedded spaces are needed, it is recommended that a database type of accounting storage be configured. Note: The content's of SLURM's database are maintained in lower case. This may result in some sacct output differing from that of other SLURM commands. Note: Much of the data reported by sacct has been generated by the wait3() and getrusage() system calls. Some systems gather and report incomplete information for these calls; sacct reports values of 0 for this missing data. See your systems getrusage (3) man page for information about which data are actually available on your system. If --dump is specified, the field selection options (--brief, --format, ...) have no effect. Elapsed time fields are presented as 2 fields, integral seconds and integral microseconds If --dump is not specified, elapsed time fields are presented as [[days-]hours:]minutes:seconds.hundredths. The default input file is the file named in the AccountingStorageLoc parameter in slurm.conf. OPTIONS -a, --allusers Displays the current user's jobs. Displays all users jobs when run by root. -A account_list , --accounts=account_list Displays jobs when a comma separated list of accounts are given as the argument. -b, --brief Displays a brief listing, which includes the following data: jobid status exitcode This option has no effect when the ---dump option is also specified. -c, --completion Use job completion instead of job accounting. The JobCompType parameter in the slurm.conf file must be defined to a non-none option. -d, --dump Dumps the raw data records. The section titled "INTERPRETING THE --dump OPTION OUTPUT" describes the data output when this option is used. --duplicates If SLURM job ids are reset, some job numbers will probably appear more than once in the accounting log file but refer to different jobs. Such jobs can be dis- tinguished by the "submit" time stamp in the data records. When data for specific jobs are requested with the --jobs option, sacct returns the most recent job with that number. This behavior can be overridden by speci- fying --duplicates, in which case all records that match the selection criteria will be returned. -e, --helpformat Print a list of fields that can be specified with the --format option. Fields available: AllocCPUS Account AssocID AveCPU AvePages AveRSS AveVMSize BlockID Cluster Comment CPUTime CPUTimeRAW DerivedExitCode Elapsed Eligible End ExitCode GID Group JobID JobName Layout MaxPages MaxPagesNode MaxPagesTask MaxRSS MaxRSSNode MaxRSSTask MaxVMSize MaxVMSizeNode MaxVMSizeTask MinCPU MinCPUNode MinCPUTask NCPUS NNodes NodeList NTasks Priority Partition QOS QOSRAW ReqCPUS Reserved ResvCPU ResvCPURAW Start State Submit Suspended SystemCPU Timelimit TotalCPU UID User UserCPU WCKey WCKeyID The section titled "Job Accounting Fields" describes these fields. -E end_time, --endtime=end_time Select jobs eligible before time. If states are given with the -s option return jobs in this state before this period. Valid time formats are... HH:MM[:SS] [AM|PM] MMDD[YY] or MM/DD[/YY] or MM.DD[.YY] MM/DD[/YY]-HH:MM[:SS] YYYY-MM-DD[THH:MM[:SS]] -f file, --file=file Causes the sacct command to read job accounting data from the named file instead of the current SLURM job accounting log file. Only applicable when running the filetxt plugin. -g gid_list, --gid=gid_list --group=group_list Displays the statistics only for the jobs started with the GID or the GROUP specified by the gid_list or thegroup_list operand, which is a comma-separated list. Space characters are not allowed. Default is no restrictions.. -h, --help Displays a general help message. -j job(.step) , --jobs=job(.step) Displays information about the specified job(.step) or list of job(.step)s. The job(.step) parameter is a comma-separated list of jobs. Space characters are not permitted in this list. NOTE: A step id of 'batch' will display the information about the batch step. The batch step information is only available after the batch job is complete unlike regular steps which are available when they start. The default is to display information on all jobs. -k, --timelimit-min Only send data about jobs with this timelimit. If used with timelimit_max this will be the minimum timelimit of the range. Default is no restriction. -K, --timelimit-max Ignored by itself, but if timelimit_min is set this will be the maximum time- limit of the range. Default is no restriction. -l, --long Equivalent to specifying: --format=jobid,jobname,partition,maxvmsize,maxvmsizenode,maxvmsizetask,avevm- size,maxrss,maxrssnode,maxrsstask,averss,maxpages,maxpagesnode,maxpages- task,avepages,mincpu,mincpunode,mincputask,avecpu,ntasks,allocc- pus,elapsed,state,exitcode -L, --allclusters Display jobs ran on all clusters. By default, only jobs ran on the cluster from where sacct is called are displayed. -M cluster_list, --clusters=cluster_list Displays the statistics only for the jobs started on the clusters specified by the cluster_list operand, which is a comma-separated list of clusters. Space characters are not allowed in the cluster_list. Use -1 for all clusters. The default is current cluster you are executing the sacct command on. -n, --noheader No heading will be added to the output. The default action is to display a header. This option has no effect when used with the --dump option. -N node_list, --nodelist=node_list Display jobs that ran on any of these node(s). node_list can be a ranged string. -o, --format Comma separated list of fields. (use "--helpformat" for a list of available fields). NOTE: When using the format option for listing various fields you can put a %NUMBER afterwards to specify how many characters should be printed. e.g. format=name%30 will print 30 characters of field name right justified. A %-30 will print 30 characters left justified. -O, --formatted_dump Dumps accounting records in an easy-to-read format. This option is provided for debugging. -p, --parsable output will be '|' delimited with a '|' at the end -P, --parsable2 output will be '|' delimited without a '|' at the end -q, --qos Only send data about jobs using these qos. Default is all. -r, --partition Comma separated list of partitions to select jobs and job steps from. The default is all partitions. -s state_list , --state=state_list Selects jobs based on their state during the time period given. Unless other- wise specified, the start and end time will be the current time when the --state option is specified and only currently running jobs can be displayed. A start and/or end time must be specified to view information about jobs not currently running. The following state designators are valid and multiple state names may be specified using comma separators. Either the short or long form of the state name may be used (e.g. CA or CANCELLED) and the the the name is case insensitive (e.g. ca and CA both work). CA CANCELLED Job was explicitly cancelled by the user or system adminis- trator. The job may or may not have been initiated. CD COMPLETED Job has terminated all processes on all nodes. CF CONFIGURING Job has been allocated resources, but are waiting for them to become ready for use (e.g. booting). CG COMPLETING Job is in the process of completing. Some processes on some nodes may still be active. F FAILED Job terminated with non-zero exit code or other failure con- dition. NF NODE_FAIL Job terminated due to failure of one or more allocated nodes. PD PENDING Job is awaiting resource allocation. PR PREEMPTED Job terminated due to preemption. R RUNNING Job currently has an allocation. RS RESIZING Job is about to change size. S SUSPENDED Job has an allocation, but execution has been suspended. TO TIMEOUT Job terminated upon reaching its time limit. The state_list operand is a comma-separated list of these state designators. Space characters are not allowed in the state_list NOTE: When specifying states and no start time is given the default starttime is 'now'. . -S, --starttime Select jobs eligible after the specified time. Default is midnight of current day. If states are given with the -s option then return jobs in this state at this time, 'now' is also used as the default time. Valid time formats are... HH:MM[:SS] [AM|PM] MMDD[YY] or MM/DD[/YY] or MM.DD[.YY] MM/DD[/YY]-HH:MM[:SS] YYYY-MM-DD[THH:MM[:SS]] -T, --truncate Truncate time. So if a job started before --starttime the start time would be truncated to --starttime. The same for end time and --endtime. -u uid_list, --uid=uid_list, --user=user_list Use this comma separated list of uids or user names to select jobs to display. By default, the running user's uid is used. --usage Display a command usage summary. -v, --verbose Primarily for debugging purposes, report the state of various variables during processing. -V, --version Print version. -W wckey_list, --wckeys=wckey_list Displays the statistics only for the jobs started on the wckeys specified by the wckey_list operand, which is a comma-separated list of wckey names. Space char- acters are not allowed in the wckey_list. Default is all wckeys. -x associd_list, --associations=assoc_list Displays the statistics only for the jobs running under the association ids specified by the assoc_list operand, which is a comma-separated list of associa- tion ids. Space characters are not allowed in the assoc_list. Default is all associations. -X, --allocations Only show cumulative statistics for each job, not the intermediate steps. Job Accounting Fields The following describes each job accounting field: alloccpus Count of allocated processors. account Account the job ran under. associd Reference to the association of user, account and cluster. AveCPU Average (system + user) CPU time of all tasks in job. AvePages Average number of page faults of all tasks in job. AveRSS Average resident set size of all tasks in job. AveVMSize Average Virtual Memory size of all tasks in job. blockid Block ID, applicable to BlueGene computers only. cluster Cluster name. Comment The job's comment string when the AccountingStoreJobComment parameter in the slurm.conf file is set (or defaults) to YES. The Comment string can be modified by invoking sacctmgr modify job or the specialized sjobexit- mod command. cputime Formatted number of cpu seconds a process was allocated. cputimeraw How much cpu time process was allocated in second format, not formatted like above. DerivedExitCode The highest exit code returned by the job's job steps (srun invocations). Following the colon is the signal that caused the process to terminate if it was terminated by a signal. The DerivedExitCode can be modified by invoking sacctmgr modify job or the specialized sjobexitmod command. elapsed The jobs elapsed time. The format of this fields output is as follows: [DD-[hh:]]mm:ss as defined by the following: DD days hh hours mm minutes ss seconds eligible When the job became eligible to run. end Termination time of the job. Format output is as follows: MM/DD-hh:mm:ss as defined by the following: MM month DD day hh hours mm minutes ss seconds exitcode The exit code returned by the job script or salloc, typically as set by the exit() function. Following the colon is the signal that caused the process to terminate if it was terminated by a signal. gid The group identifier of the user who ran the job. group The group name of the user who ran the job. JobID The number of the job or job step. It is in the form: job.jobstep. jobname The name of the job or job step. The slurm_accounting.log file is a space delimited file. Because of this if a space is used in the jobname an underscore is substituted for the space before the record is written to the accounting file. So when the jobname is displayed by sacct the job- name that had a space in it will now have an underscore in place of the space. layout What the layout of a step was when it was running. This can be used to give you an idea of which node ran which rank in your job. MaxPages Maximum number of page faults of all tasks in job. MaxPagesNode The node on which the maxpages occurred. MaxPagesTask The task ID where the maxpages occurred. MaxRSS Maximum resident set size of all tasks in job. MaxRSSNode The node on which the maxrss occurred. MaxRSSTask The task ID where the maxrss occurred. MaxVMSize Maximum Virtual Memory size of all tasks in job. MaxVMSizeNode The node on which the maxvmsize occurred. MaxVMSizeTask The task ID where the maxvmsize occurred. MinCPU Minimum (system + user) CPU time of all tasks in job. MinCPUNode The node on which the mincpu occurred. MinCPUTask The task ID where the mincpu occurred. ncpus Total number of CPUs allocated to the job. nodelist List of nodes in job/step. nnodes Number of nodes in a job or step. NTasks Total number of tasks in a job or step. priority Slurm priority. partition Identifies the partition on which the job ran. qos Name of Quality of Service. qosraw Id of Quality of Service. reqcpus Required CPUs. reserved How much wall clock time was used as reserved time for this job. This is derived from how long a job was waiting from eligible time to when it actually started. resvcpu Formatted time for how long (cpu secs) a job was reserved for. resvcpuraw Reserved CPUs in second format, not formatted. start Initiation time of the job in the same format as end. state Displays the job status, or state. Output can be RUNNING, RESIZING, SUSPENDED, COMPLETED, CANCELLED, FAILED, TIMEOUT, PREEMPTED or NODE_FAIL. If more information is available on the job state than will fit into the current field width (for example, the uid that CANCELLED a job) the state will be followed by a "+". You can increase the size of the displayed state using the "%NUMBER" format modi- fier described earlier. submit The time and date stamp (in Universal Time Coordinated, UTC) the job was submitted. The format of the output is identical to that of the end field. NOTE: If a job is requeued, the submit time is reset. To obtain the original submit time it is necessary to use the -D or --duplicate option to display all duplicate entries for a job. suspended How long the job was suspended for. SystemCPU The amount of system CPU time used by the job or job step. The format of the output is identical to that of the elapsed field. NOTE: SystemCPU provides a measure of the task's parent process and does not include CPU time of child processes. timelimit What the timelimit was/is for the job. TotalCPU The sum of the SystemCPU and UserCPU time used by the job or job step. The total CPU time of the job may exceed the job's elapsed time for jobs that include multiple job steps. The format of the output is identical to that of the elapsed field. NOTE: TotalCPU provides a measure of the task's parent process and does not include CPU time of child processes. uid The user identifier of the user who ran the job. user The user name of the user who ran the job. UserCPU The amount of user CPU time used by the job or job step. The format of the output is identical to that of the elapsed field. NOTE: UserCPU provides a measure of the task's parent process and does not include CPU time of child processes. wckey Workload Characterization Key. Arbitrary string for grouping orthog- onal accounts together. wckeyid Reference to the wckey. INTERPRETING THE -DUMP OPTION OUTPUT The sacct command's --dump option displays data in a horizontal list of fields depending on the record type. There are three record types: JOB_START, JOB_STEP, and JOB_TERMI- NATED. There is a subsection that describes the output for each record type. When the data output is a job accounting field, as described in the section titled "Job Accounting Fields", only the name of the job accounting field is listed. Otherwise, addi- tional information is provided. Note: The output for the JOB_STEP and JOB_TERMINATED record types present a pair of fields for the following data: Total CPU time, Total User CPU time, and Total System CPU time. The first field of each pair is the time in seconds expressed as an integer. The second field of each pair is the fractional number of sec- onds multiplied by one million. Thus, a pair of fields output as "1 024315" means that the time is 1.024315 seconds. The least significant digits in the second field are truncated in formatted displays. Output for the JOB_START Record Type The following describes the horizontal fields output by the sacct --dump option for the JOB_START record type. Field # Field 1 job 2 partition 3 submitted 4 The jobs start time; this value is the number of non-leap seconds since the Epoch (00:00:00 UTC, January 1, 1970) 5 uid.gid 6 (Reserved) 7 JOB_START (literal string) 8 Job Record Version (1) 9 The number of fields in the record (16) 10 uid 11 gid 12 The job name 13 Batch Flag (0=no batch) 14 Relative SLURM priority 15 ncpus 16 nodes Output for the JOB_STEP Record Type The following describes the horizontal fields output by the sacct --dump option for the JOB_STEP record type. Field # Field 1 job 2 partition 3 submitted 4 The jobs start time; this value is the number of non-leap seconds since the Epoch (00:00:00 UTC, January 1, 1970) 5 uid.gid 6 (Reserved) 7 JOB_STEP (literal string) 8 Job Record Version (1) 9 The number of fields in the record (38) 10 jobid 11 end 12 Completion Status; the mnemonics, which may appear in uppercase or lower- case, are as follows: CA Cancelled CD Completed successfully F Failed NF Job terminated from node failure R Running S Suspended TO Timed out 13 exitcode 14 ntasks 15 ncpus 16 elapsed time in seconds expressed as an integer 17 Integer portion of the Total CPU time in seconds for all processes 18 Fractional portion of the Total CPU time for all processes expressed in microseconds 19 Integer portion of the Total User CPU time in seconds for all processes 20 Fractional portion of the Total User CPU time for all processes expressed in microseconds 21 Integer portion of the Total System CPU time in seconds for all processes 22 Fractional portion of the Total System CPU time for all processes expressed in microseconds 23 rss 24 ixrss 25 idrss 26 isrss 27 minflt 28 majflt 29 nswap 30 inblocks 31 outblocks 32 msgsnd 33 msgrcv 34 nsignals 35 nvcsw 36 nivcsw 37 vsize Output for the JOB_TERMINATED Record Type The following describes the horizontal fields output by the sacct --dump option for the JOB_TERMINATED (literal string) record type. Field # Field 1 job 2 partition 3 submitted 4 The jobs start time; this value is the number of non-leap seconds since the Epoch (00:00:00 UTC, January 1, 1970) 5 uid.gid 6 (Reserved) 7 JOB_TERMINATED (literal string) 8 Job Record Version (1) 9 The number of fields in the record (38) Although thirty-eight fields are displayed by the sacct command for the JOB_TERMINATED record, only fields 1 through 12 are recorded in the actual data file. The sacct command aggregates the remainder. 10 The total elapsed time in seconds for the job. 11 end 12 Completion Status; the mnemonics, which may appear in uppercase or lower- case, are as follows: CA Cancelled CD Completed successfully F Failed NF Job terminated from node failure R Running TO Timed out 13 exitcode 14 ntasks 15 ncpus 16 elapsed time in seconds expressed as an integer 17 Integer portion of the Total CPU time in seconds for all processes 18 Fractional portion of the Total CPU time for all processes expressed in microseconds 19 Integer portion of the Total User CPU time in seconds for all processes 20 Fractional portion of the Total User CPU time for all processes expressed in microseconds 21 Integer portion of the Total System CPU time in seconds for all processes 22 Fractional portion of the Total System CPU time for all processes expressed in microseconds 23 rss 24 ixrss 25 idrss 26 isrss 27 minflt 28 majflt 29 nswap 30 inblocks 31 outblocks 32 msgsnd 33 msgrcv 34 nsignals 35 nvcsw 36 nivcsw 37 vsize EXAMPLES This example illustrates the default invocation of the sacct command: # sacct Jobid Jobname Partition Account AllocCPUS State ExitCode ---------- ---------- ---------- ---------- ---------- ---------- -------- 2 script01 srun acct1 1 RUNNING 0 3 script02 srun acct1 1 RUNNING 0 4 endscript srun acct1 1 RUNNING 0 4.0 srun acct1 1 COMPLETED 0 This example shows the same job accounting information with the brief option. # sacct --brief Jobid State ExitCode ---------- ---------- -------- 2 RUNNING 0 3 RUNNING 0 4 RUNNING 0 4.0 COMPLETED 0 # sacct --allocations Jobid Jobname Partition Account AllocCPUS State ExitCode ---------- ---------- ---------- ---------- ------- ---------- -------- 3 sja_init andy acct1 1 COMPLETED 0 4 sjaload andy acct1 2 COMPLETED 0 5 sja_scr1 andy acct1 1 COMPLETED 0 6 sja_scr2 andy acct1 18 COMPLETED 2 7 sja_scr3 andy acct1 18 COMPLETED 0 8 sja_scr5 andy acct1 2 COMPLETED 0 9 sja_scr7 andy acct1 90 COMPLETED 1 10 endscript andy acct1 186 COMPLETED 0 This example demonstrates the ability to customize the output of the sacct command. The fields are displayed in the order designated on the command line. # sacct --format=jobid,elapsed,ncpus,ntasks,state Jobid Elapsed Ncpus Ntasks State ---------- ---------- ---------- -------- ---------- 3 00:01:30 2 1 COMPLETED 3.0 00:01:30 2 1 COMPLETED 4 00:00:00 2 2 COMPLETED 4.0 00:00:01 2 2 COMPLETED 5 00:01:23 2 1 COMPLETED 5.0 00:01:31 2 1 COMPLETED COPYING Copyright (C) 2005-2007 Copyright Hewlett-Packard Development Company L.P. Copyright (C) 2008-2009 Lawrence Livermore National Security. Produced at Lawrence Liver- more National Laboratory (cf, DISCLAIMER). CODE-OCEC-09-009. All rights reserved. This file is part of SLURM, a resource management program. For details, see . SLURM is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. SLURM is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. FILES /etc/slurm.conf Entries to this file enable job accounting and designate the job accounting log file that collects system job accounting. /var/log/slurm_accounting.log The default job accounting log file. By default, this file is set to read and write permission for root only. SEE ALSO sstat(1), ps (1), srun(1), squeue(1), getrusage (2), time (2) sacct 2.2 March 2010 SACCT(1)