table of contents
- NAME
- SYNOPSIS
- DESCRIPTION
- OPTIONS
- COMMANDS
- ENTITIES
- GENERAL SPECIFICATIONS FOR ASSOCIATION BASED ENTITIES
- SPECIFICATIONS FOR ACCOUNTS
- LIST/SHOW ACCOUNT FORMAT OPTIONS
- SPECIFICATIONS FOR ASSOCIATIONS
- LIST/SHOW ASSOCIATION FORMAT OPTIONS
- SPECIFICATIONS FOR CLUSTERS
- LIST/SHOW CLUSTER FORMAT OPTIONS
- SPECIFICATIONS FOR COORDINATOR
- SPECIFICATIONS FOR EVENTS
- LIST/SHOW EVENT FORMAT OPTIONS
- SPECIFICATIONS FOR JOB
- LIST/SHOW JOB FORMAT OPTIONS
- SPECIFICATIONS FOR QOS
- LIST/SHOW QOS FORMAT OPTIONS
- SPECIFICATIONS FOR RESOURCE
- LIST/SHOW RESOURCE FORMAT OPTIONS
- SPECIFICATIONS FOR TRANSACTIONS
- LIST/SHOW TRANSACTIONS FORMAT OPTIONS
- SPECIFICATIONS FOR USERS
- LIST/SHOW USER FORMAT OPTIONS
- LIST/SHOW WCKey
- GLOBAL FORMAT OPTION
- FLAT FILE DUMP AND LOAD
- ARCHIVE FUNCTIONALITY
- EXAMPLES
- COPYING
- SEE ALSO
SACCTMGR(1) | Slurm components | SACCTMGR(1) |
NAME¶
sacctmgr - Used to view and modify Slurm account information.SYNOPSIS¶
sacctmgr [OPTIONS...] [COMMAND...]DESCRIPTION¶
sacctmgr is used to view or modify Slurm account information. The account information is maintained within a database with the interface being provided by slurmdbd (Slurm Database daemon). This database can serve as a central storehouse of user and computer information for multiple computers at a single site. Slurm account information is recorded based upon four parameters that form what is referred to as an association. These parameters are user, cluster, partition, and account. user is the login name. cluster is the name of a Slurm managed cluster as specified by the ClusterName parameter in the slurm.conf configuration file. partition is the name of a Slurm partition on that cluster. account is the bank account for a job. The intended mode of operation is to initiate the sacctmgr command, add, delete, modify, and/or list association records then commit the changes and exit.- Note:
- The content's of SLURM's database are maintained in lower case. This may result in some sacctmgr output differing from that of other SLURM commands.
OPTIONS¶
- -h, --help
- Print a help message describing the usage of sacctmgr. This is equivalent to the help command.
- -i, --immediate
- commit changes immediately.
- -n, --noheader
- No header will be added to the beginning of the output.
- -p, --parsable
- Output will be '|' delimited with a '|' at the end.
- -P, --parsable2
- Output will be '|' delimited without a '|' at the end.
- -Q, --quiet
- Print no messages other than error messages. This is equivalent to the quiet command.
- -r, --readonly
- Makes it so the running sacctmgr cannot modify accounting information. The readonly option is for use within interactive mode.
- -s, --associations
- Use with show or list to display associations with the entity. This is equivalent to the associations command.
- -v, --verbose
- Enable detailed logging. This is equivalent to the verbose command.
- -V , --version
- Display version number. This is equivalent to the version command.
COMMANDS¶
- add <ENTITY> <SPECS>
- Add an entity. Identical to the create command.
- associations
- Use with show or list to display associations with the entity.
- create <ENTITY> <SPECS>
- Add an entity. Identical to the add command.
- delete <ENTITY> where <SPECS>
- Delete the specified entities.
- dump <ENTITY> <File=FILENAME>
- Dump cluster data to the specified file.
- exit
- Terminate sacctmgr interactive mode. Identical to the quit command.
- help
- Display a description of sacctmgr options and commands.
- list <ENTITY> [<SPECS>]
- Display information about the specified entity. By default, all entries are displayed, you can narrow results by specifying SPECS in your query. Identical to the show command.
- load <FILENAME>
- Load cluster data to the specified file.
- modify <ENTITY> where <SPECS> set <SPECS>
- Modify an entity.
- problem
- Use with show or list to display entity problems.
- quiet
- Print no messages other than error messages.
- quit
- Terminate the execution of sacctmgr interactive mode. Identical to the exit command.
- show <ENTITY> [<SPECS>]
- Display information about the specified entity. By default, all entries are displayed, you can narrow results by specifying SPECS in your query. Identical to the list command.
- verbose
- Enable detailed logging. This includes time-stamps on data structures, record counts, etc. This is an independent command with no options meant for use in interactive mode.
- version
- Display the version number of sacctmgr.
- !!
- Repeat the last command.
ENTITIES¶
- account
- A bank account, typically specified at job submit time using the --account= option. These may be arranged in a hierarchical fashion, for example accounts chemistry and physics may be children of the account science. The hierarchy may have an arbitrary depth.
- association
- The entity used to group information consisting of four parameters: account, cluster, partition (optional), and user. Used only with the list or show command. Add, modify, and delete should be done to a user, account or cluster entity. This will in-turn update the underlying associations.
- cluster
- The ClusterName parameter in the slurm.conf configuration file, used to differentiate accounts from on different machines.
- configuration
- Used only with the list or show command to report current system configuration.
- coordinator
- A special privileged user usually an account manager or such that can add users or sub accounts to the account they are coordinator over. This should be a trusted person since they can change limits on account and user associations inside their realm.
- event
- Events like downed or draining nodes on clusters.
- job
- Job - but only two specific fields of the job: Derived Exit Code and the Comment String
- qos
- Quality of Service.
- Resource
- Resources - software resources for the system.
- transaction
- List of transactions that have occurred during a given time period.
- user
- The login name.
- wckeys
- Workload Characterization Key. An arbitrary string for grouping orthogonal accounts.
GENERAL SPECIFICATIONS FOR ASSOCIATION BASED ENTITIES¶
NOTE: The group limits (GrpJobs, GrpNodes, etc.) are tested when a job is being considered for being allocated resources. If starting a job would cause any of its group limit to be exceeded, that job will not be considered for scheduling even if that job might preempt other jobs which would release sufficient group resources for the pending job to be initiated.- DefaultQOS=<default qos>
- The default QOS this association and its children should have. This is overridden if set directly on a user. To clear a previously set value use the modify command with a new value of -1.
- Fairshare=<fairshare number | parent>
- Number used in conjunction with other accounts to determine job priority. Can also be the string parent, this means that the parent association is used for fairshare. To clear a previously set value use the modify command with a new value of -1.
- GraceTime=<preemption grace time in seconds>
- Specifies, in units of seconds, the preemption grace time to be extended to a job which has been selected for preemption. The default value is zero, no preemption grace time is allowed on this QOS.
- GrpCPUMins=<max cpu minutes>
- The total number of cpu minutes that can possibly be used by past, present and future jobs running from this association and its children. To clear a previously set value use the modify command with a new value of -1.
- GrpCPURunMins=<max cpu run minutes>
- Used to limit the combined total number of CPU minutes used by all jobs running with this association and its children. This takes into consideration time limit of running jobs and consumes it, if the limit is reached no new jobs are started until other jobs finish to allow time to free up.
- GrpCPUs=<max cpus>
- Maximum number of CPUs running jobs are able to be allocated in aggregate for this association and all associations which are children of this association. To clear a previously set value use the modify command with a new value of -1.
- GrpJobs=<max jobs>
- Maximum number of running jobs in aggregate for this association and all associations which are children of this association. To clear a previously set value use the modify command with a new value of -1.
- GrpMemory=<max memory (MB) >
- Maximum amount of memory running jobs are able to be allocated in aggregate for this association and all associations which are children of this association. To clear a previously set value use the modify command with a new value of -1.
- GrpNodes=<max nodes>
- Maximum number of nodes running jobs are able to be allocated in aggregate for this association and all associations which are children of this association. To clear a previously set value use the modify command with a new value of -1.
- GrpSubmitJobs=<max jobs>
- Maximum number of jobs which can be in a pending or running state at any time in aggregate for this association and all associations which are children of this association. To clear a previously set value use the modify command with a new value of -1.
- GrpWall=<max wall>
- Maximum wall clock time running jobs are able to be allocated in aggregate for this association and all associations which are children of this association. To clear a previously set value use the modify command with a new value of -1.
- MaxCPUMins=<max cpu minutes>
- Maximum number of CPU minutes each job is able to use in this association. This is overridden if set directly on a user. Default is the cluster's limit. To clear a previously set value use the modify command with a new value of -1.
- MaxCPUs=<max cpus>
- Maximum number of CPUs each job is able to use in this association. This is overridden if set directly on a user. Default is the cluster's limit. To clear a previously set value use the modify command with a new value of -1.
- MaxJobs=<max jobs>
- Maximum number of jobs each user is allowed to run at one time in this association. This is overridden if set directly on a user. Default is the cluster's limit. To clear a previously set value use the modify command with a new value of -1.
- MaxNodes=<max nodes>
- Maximum number of nodes each job is able to use in this association. This is overridden if set directly on a user. Default is the cluster's limit. To clear a previously set value use the modify command with a new value of -1. This is a c-node limit on BlueGene systems.
- MaxSubmitJobs=<max jobs>
- Maximum number of jobs which can this association can have in a pending or running state at any time. Default is the cluster's limit. To clear a previously set value use the modify command with a new value of -1.
- MaxWall=<max wall>
- Maximum wall clock time each job is able to use in this association. This is overridden if set directly on a user. Default is the cluster's limit. <max wall> format is <min> or <min>:<sec> or <hr>:<min>:<sec> or <days>-<hr>:<min>:<sec> or <days>-<hr>. The value is recorded in minutes with rounding as needed. To clear a previously set value use the modify command with a new value of -1.
- QosLevel<operator><comma separated list of qos names>
- Specify the default Quality of Service's that jobs are able to run at for this association. To get a list of valid QOS's use 'sacctmgr list qos'. This value will override its parents value and push down to its children as the new default. Setting a QosLevel to '' (two single quotes with nothing between them) restores its default setting. You can also use the operator += and -= to add or remove certain QOS's from a QOS list. Valid <operator> values include:
- =
- Set QosLevel to the specified value. Note: the QOS that can be used at a given account in the hierarchy are inherited by the children of that account. By assigning QOS with the = sign only the assigned QOS can be used by the account and its childern.
- +=
- Add the specified <qos> value to the current QosLevel. The account will have access to this QOS and the other previously assigned to it.
- -=
- Remove the specified <qos> value from the current QosLevel.
- See the EXAMPLES section below.
SPECIFICATIONS FOR ACCOUNTS¶
- Cluster=<cluster>
- Specific cluster to add account to. Default is all in system.
- Description=<description>
- An arbitrary string describing an account.
- Name=<name>
- The name of a bank account. Note the name must be unique and can not be represent different bank accounts at different points in the account hierarchy.
- Organization=<org>
- Organization to which the account belongs.
- Parent=<parent>
- Parent account of this account. Default is the root account, a top level account.
- RawUsage=<value>
- This allows an administrator to reset the raw usage accrued to an account. The only value currently supported is 0 (zero). This is a settable specification only - it cannot be used as a filter to list accounts.
- WithAssoc
- Display all associations for this account.
- WithCoord
- Display all coordinators for this account.
- WithDeleted
- Display information with previously deleted data.
LIST/SHOW ACCOUNT FORMAT OPTIONS¶
- Account
- The name of a bank account.
- Description
- An arbitrary string describing an account.
- Organization
- Organization to which the account belongs.
- Coordinators
- List of users that are a coordinator of the account. (Only filled in when using the WithCoordinator option.)
SPECIFICATIONS FOR ASSOCIATIONS¶
- Clusters=<comma separated list of cluster names>
- List the associations of the cluster(s).
- Accounts=<comma separated list of account names>
- List the associations of the account(s).
- Users=<comma separated list of user names>
- List the associations of the user(s).
- Partition=<comma separated list of partition names>
- List the associations of the partition(s).
- OnlyDefaults
- Display only associations that are default associations
- Tree
- Display account names in a hierarchical fashion.
- WithDeleted
- Display information with previously deleted data.
- WithSubAccounts
- Display information with subaccounts. Only really valuable when used with the account= option. This will display all the subaccount associations along with the accounts listed in the option.
- WOLimits
- Display information without limit information. This is for a smaller default format of Cluster,Account,User,Partition
- WOPInfo
- Display information without parent information. (i.e. parent id, and parent account name.) This option also invokes WOPLIMITS.
- WOPLimits
- Display information without hierarchical parent limits. (i.e. will only display limits where they are set instead of propagating them from the parent.)
LIST/SHOW ASSOCIATION FORMAT OPTIONS¶
- Account
- The name of a bank account in the association.
- Cluster
- The name of a cluster in the association.
- DefaultQOS
- The QOS the association will use by default if it as access to it in the QOS list mentioned below.
- Fairshare
- Number used in conjunction with other accounts to determine job priority. Can also be the string parent, this means that the parent association is used for fairshare.
- GrpCPUMins
- The total number of cpu minutes that can possibly be used by past, present and future jobs running from this association and its children.
- GrpCPURunMins
- Used to limit the combined total number of CPU minutes used by all jobs running with this association and its children. This takes into consideration time limit of running jobs and consumes it, if the limit is reached no new jobs are started until other jobs finish to allow time to free up.
- GrpCPUs
- Maximum number of CPUs running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- GrpJobs
- Maximum number of running jobs in aggregate for this association and all associations which are children of this association.
- GrpNodes
- Maximum number of nodes running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- GrpSubmitJobs
- Maximum number of jobs which can be in a pending or running state at any time in aggregate for this association and all associations which are children of this association.
- GrpWall
- Maximum wall clock time running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- ID
- The id of the association.
- LFT
- Associations are kept in a hierarchy: this is the left most spot in the hierarchy. When used with the RGT variable, all associations with a LFT inside this LFT and before the RGT are children of this association.
- MaxCPUMins
- Maximum number of CPU minutes each job is able to use.
- MaxCPUs
- Maximum number of CPUs each job is able to use.
- MaxJobs
- Maximum number of jobs each user is allowed to run at one time.
- MaxNodes
- Maximum number of nodes each job is able to use.
- MaxSubmitJobs
- Maximum number of jobs pending or running state at any time.
- MaxWall
- Maximum wall clock time each job is able to use.
- Qos
- Valid QOS´ for this association.
- ParentID
- The association id of the parent of this association.
- ParentName
- The account name of the parent of this association.
- Partition
- The name of a partition in the association.
- RawQOS
- The numeric values of valid QOS´ for this association.
- RGT
- Associations are kept in a hierarchy: this is the right most spot in the hierarchy. When used with the LFT variable, all associations with a LFT inside this RGT and after the LFT are children of this association.
- User
- The name of a user in the association.
SPECIFICATIONS FOR CLUSTERS¶
- Classification=<classification>
- Type of machine, current classifications are capability and capacity.
- Flags=<flag list>
- Comma separated list of Attributes for a particular cluster. Current Flags include AIX, BGL, BGP, BGQ, Bluegene, CrayXT, FrontEnd, MultipleSlurmd, SunConstellation, and XCPU
- Name=<name>
- The name of a cluster. This should be equal to the ClusterName parameter in the slurm.conf configuration file for some Slurm-managed cluster.
- RPC=<rpc list>
- Comma separated list of numeric RPC values.
- WOLimits
- Display information without limit information. This is for a smaller default format of Cluster,ControlHost,ControlPort,RPC
LIST/SHOW CLUSTER FORMAT OPTIONS¶
- Classification
- Type of machine, i.e. capability or capacity.
- Cluster
- The name of the cluster.
- ControlHost
- When a slurmctld registers with the database the ip address of the controller is placed here.
- ControlPort
- When a slurmctld registers with the database the port the controller is listening on is placed here.
- CPUCount
- The current count of cpus on the cluster.
- Flags
- Attributes possessed by the cluster.
- NodeCount
- The current count of nodes associated with the cluster.
- NodeNames
- The current Nodes associated with the cluster.
- PluginIDSelect
- The numeric value of the select plugin the cluster is using.
- RPC
- When a slurmctld registers with the database the rpc version the controller is running is placed here.
SPECIFICATIONS FOR COORDINATOR¶
- Account=<comma separated list of account names>
- Account name to add this user as a coordinator to.
- Names=<comma separated list of user names>
- Names of coordinators.
SPECIFICATIONS FOR EVENTS¶
- All_Clusters
- Get information on all cluster shortcut.
- All_Time
- Get time period for all time shortcut.
- Clusters=<comma separated list of cluster names>
- List the events of the cluster(s). Default is the cluster where the command was run.
- End=<OPT>
- Period ending of events. Default is now.
Valid time formats are...
HH:MM[:SS] [AM|PM]
- Event=<OPT>
- Specific events to look for, valid options are Cluster or Node, default is both.
- MaxCPUs=<OPT>
- Max number of cpus affected by an event.
- MinCPUs=<OPT>
- Min number of cpus affected by an event.
- Nodes=<comma separated list of node names>
- Node names affected by an event.
- Reason=<comma separated list of reasons>
- Reason an event happened.
- Start=<OPT>
- Period start of events. Default is 00:00:00 of previous day, unless states
are given with the States= spec events. If this is the case the default
behavior is to return events currently in the states specified.
Valid time formats are...
HH:MM[:SS] [AM|PM]
- States=<comma separated list of states>
- State of a node in a node event. If this is set, the event type is set automatically to Node.
- User=<comma separated list of users>
- Query against users who set the event. If this is set, the event type is set automatically to Node since only user slurm can perform a cluster event.
LIST/SHOW EVENT FORMAT OPTIONS¶
- Cluster
- The name of the cluster event happened on.
- ClusterNodes
- The hostlist of nodes on a cluster in a cluster event.
- CPUs
- Number of cpus involved with the event.
- Duration
- Time period the event was around for.
- End
- Period when event ended.
- Event
- Name of the event.
- EventRaw
- Numeric value of the name of the event.
- NodeName
- The node affected by the event. In a cluster event, this is blank.
- Reason
- The reason an event happened.
- Start
- Period when event started.
- State
- On a node event this is the formatted state of the node during the event.
- StateRaw
- On a node event this is the numeric value of the state of the node during the event.
- User
- On a node event this is the user who caused the event to happen.
SPECIFICATIONS FOR JOB¶
- DerivedExitCode
- The derived exit code can be modified after a job completes based on the user's judgement of whether the job succeeded or failed. The user can only modify the derived exit code of their own job.
- Comment
- The job's comment string when the AccountingStoreJobComment parameter in the slurm.conf file is set (or defaults) to YES. The user can only modify the comment string of their own job.
- The DerivedExitCode and Comment fields are the only fields
- of a job record in the database that can be modified after job completion.
LIST/SHOW JOB FORMAT OPTIONS¶
The sacct command is the exclusive command to display job records from the SLURM database.SPECIFICATIONS FOR QOS¶
NOTE: The group limits (GrpJobs, GrpNodes, etc.) are tested when a job is being considered for being allocated resources. If starting a job would cause any of its group limit to be exceeded, that job will not be considered for scheduling even if that job might preempt other jobs which would release sufficient group resources for the pending job to be initiated.- Flags
- Used by the slurmctld to override or enforce certain characteristics.
- DenyOnLimit
- If set jobs using this QOS will be rejected at submission time if they do not conform to the QOS 'Max' limits. By default jobs that go over these limits will pend until they conform.
- EnforceUsageThreshold
- If set, and the QOS also has a UsageThreshold, any jobs submitted with this QOS that fall below the UsageThreshold will be held until their Fairshare Usage goes above the Threshold.
- NoReserve
- If this flag is set and backfill scheduling is used, jobs using this QOS will not reserve resources in the backfill schedule's map of resources allocated through time. This flag is intended for use with a QOS that may be preempted by jobs associated with all other QOS (e.g use with a "standby" QOS). If the allocated is used with a QOS which can not be preempted by all other QOS, it could result in starvation of larger jobs.
- PartitionMaxNodes
- If set jobs using this QOS will be able to override the requested partition's MaxNodes limit.
- PartitionMinNodes
- If set jobs using this QOS will be able to override the requested partition's MinNodes limit.
- PartitionTimeLimit
- If set jobs using this QOS will be able to override the requested partition's TimeLimit.
- RequiresReservaton
- If set jobs using this QOS must designate a reservation when submitting a job. This option can be useful in restricting usage of a QOS that may have greater preemptive capability or additional resources to be allowed only within a reservation.
- GraceTime
- Preemption grace time to be extended to a job which has been selected for preemption.
- GrpCPUMins
- The total number of cpu minutes that can possibly be used by past, present and future jobs running from this QOS.
- GrpCPURunMins Used to limit the combined total number of CPU
- minutes used by all jobs running with this QOS. This takes into consideration time limit of running jobs and consumes it, if the limit is reached no new jobs are started until other jobs finish to allow time to free up.
- GrpCPUs
- Maximum number of CPUs running jobs are able to be allocated in aggregate for this QOS.
- GrpJobs
- Maximum number of running jobs in aggregate for this QOS.
- GrpNodes
- Maximum number of nodes running jobs are able to be allocated in aggregate for this QOS.
- GrpSubmitJobs
- Maximum number of jobs which can be in a pending or running state at any time in aggregate for this QOS.
- GrpWall
- Maximum wall clock time running jobs are able to be allocated in aggregate for this QOS. If this limit is reached submission requests will be denied and the running jobs will be killed.
- ID
- The id of the QOS.
- MaxCPUMins
- Maximum number of CPU minutes each job is able to use.
- MaxCPUs
- Maximum number of CPUs each job is able to use.
- MaxCpusPerUser
- Maximum number of CPUs each user is able to use.
- MaxJobs
- Maximum number of jobs each user is allowed to run at one time.
- MaxNodes
- Maximum number of nodes each job is able to use.
- MaxNodesPerUser
- Maximum number of nodes each user is able to use.
- MaxSubmitJobs
- Maximum number of jobs pending or running state at any time per user.
- MaxWall
- Maximum wall clock time each job is able to use.
- Name
- Name of the QOS.
- Preempt
- Other QOS´ this QOS can preempt.
- PreemptMode
- Mechanism used to preempt jobs of this QOS if the clusters PreemptType is configured to preempt/qos. The default preemption mechanism is specified by the cluster-wide PreemptMode configuration parameter. Possible values are "Cluster" (meaning use cluster default), "Cancel", "Checkpoint" and "Requeue". This option is not compatible with PreemptMode=OFF or PreemptMode=SUSPEND (i.e. preempted jobs must be removed from the resources).
- Priority
- What priority will be added to a job´s priority when using this QOS.
- RawUsage=<value>
- This allows an administrator to reset the raw usage accrued to a QOS. The only value currently supported is 0 (zero). This is a settable specification only - it cannot be used as a filter to list accounts.
- UsageFactor
- Usage factor when running with this QOS
- UsageThreshold
- A float representing the lowest fairshare of an association allowable to run a job. If an association falls below this threshold and has pending jobs or submits new jobs those jobs will be held until the usage goes back above the threshold. Use sshare to see current shares on the system.
- WithDeleted
- Display information with previously deleted data.
LIST/SHOW QOS FORMAT OPTIONS¶
- Description
- An arbitrary string describing a QOS.
- GraceTime
- Preemption grace time to be extended to a job which has been selected for preemption in the format of hh:mm:ss. The default value is zero, no preemption grace time is allowed on this partition. NOTE: This value is only meaningful for QOS PreemptMode=CANCEL.
- GrpCPUMins
- The total number of cpu minutes that can possibly be used by past, present and future jobs running from this QOS. To clear a previously set value use the modify command with a new value of -1. NOTE: This limit only applies when using the Priority Multifactor plugin. The time is decayed using the value of PriorityDecayHalfLife or PriorityUsageResetPeriod as set in the slurm.conf. When this limit is reached all associated jobs running will be killed and all future jobs submitted with this QOS will be delayed until they are able to run inside the limit.
- GrpCPUs
- Maximum number of CPUs running jobs are able to be allocated in aggregate for this QOS. To clear a previously set value use the modify command with a new value of -1. (NOTE: This limit is not currently enforced in SLURM. You can still set this, but have to wait for future versions of SLURM before it is enforced.)
- GrpJobs
- Maximum number of running jobs in aggregate for this QOS. To clear a previously set value use the modify command with a new value of -1.
- GrpNodes
- Maximum number of nodes running jobs are able to be allocated in aggregate for this QOS. To clear a previously set value use the modify command with a new value of -1.
- GrpSubmitJobs
- Maximum number of jobs which can be in a pending or running state at any time in aggregate for this QOS. To clear a previously set value use the modify command with a new value of -1.
- GrpWall
- Maximum wall clock time running jobs are able to be allocated in aggregate for this QOS. To clear a previously set value use the modify command with a new value of -1. NOTE: This limit only applies when using the Priority Multifactor plugin. The time is decayed using the value of PriorityDecayHalfLife or PriorityUsageResetPeriod as set in the slurm.conf. When this limit is reached all associated jobs running will be killed and all future jobs submitted with this QOS will be delayed until they are able to run inside the limit.
- MaxCPUMins
- Maximum number of CPU minutes each job is able to use. To clear a previously set value use the modify command with a new value of -1.
- MaxCPUs
- Maximum number of CPUs each job is able to use. To clear a previously set value use the modify command with a new value of -1. (NOTE: This limit is not currently enforced in SLURM. You can still set this, but have to wait for future versions of SLURM before it is enforced.)
- MaxCpusPerUser
- Maximum number of CPUs each user is able to use. To clear a previously set value use the modify command with a new value of -1.
- MaxJobs
- Maximum number of jobs each user is allowed to run at one time. To clear a previously set value use the modify command with a new value of -1.
- MaxNodes
- Maximum number of nodes each job is able to use. To clear a previously set value use the modify command with a new value of -1.
- MaxNodesPerUser
- Maximum number of nodes each user is able to use. To clear a previously set value use the modify command with a new value of -1.
- MaxSubmitJobs
- Maximum number of jobs pending or running state at any time per user. To clear a previously set value use the modify command with a new value of -1.
- MaxWall
- Maximum wall clock time each job is able to use. <max wall> format is <min> or <min>:<sec> or <hr>:<min>:<sec> or <days>-<hr>:<min>:<sec> or <days>-<hr>. The value is recorded in minutes with rounding as needed. To clear a previously set value use the modify command with a new value of -1.
- Name
- Name of the QOS. Needed for creation.
- Preempt
- Other QOS´ this QOS can preempt. Setting a Preempt to '' (two single quotes with nothing between them) restores its default setting. You can also use the operator += and -= to add or remove certain QOS's from a QOS list.
- PreemptMode
- Mechanism used to preempt jobs of this QOS if the clusters PreemptType is configured to preempt/qos. The default preemption mechanism is specified by the cluster-wide PreemptMode configuration parameter. Possible values are "Cluster" (meaning use cluster default), "Cancel", "Checkpoint" and "Requeue". This option is not compatible with PreemptMode=OFF or PreemptMode=SUSPEND (i.e. preempted jobs must be removed from the resources).
- Priority
- What priority will be added to a job´s priority when using this QOS. To clear a previously set value use the modify command with a new value of -1.
- UsageFactor
- Usage factor when running with this QOS. This is a float that is factored into the time of running jobs. e.g. if the usagefactor of a QOS was 2 for every cpu second a job ran it would count for 2. Also if the usagefactor was .5 every second would only could for half the time. Setting this value to 0 will make it so any job running will not add time to fairshare or association/qos limits. To clear a previously set value use the modify command with a new value of -1.
SPECIFICATIONS FOR RESOURCE¶
Clusters=<name list> Comma separated list of cluster names on which specified resources are to be available. If no names are designated the all clusters already allowed to use this resource will be altered.- Count=<OPT>
- Number of software resources of a specific name configured on the system being controlled by a resource manager.
- Description=
- A brief description of the resource.
- Flags=<OPT>
- Flags that identify specific attributes of the system resource. At this time no flags have been defined.
- Manager=<OPT>
- The name of a software resource manager controlling this system resource.
- Names=<OPT>
- Comma separated list of the name of a resource configured on the system being controlled by a resource manager. If this resource is seen on the slurmctld it's name will be name@server to distinguish it from local resources defined in a slurm.conf.
- PercentAllowed=<percent allowed>
- Percentage of a specific resource that can be used on specified cluster.
- Server=<OPT>
- The name of the server serving up the resource. Default is 'slurmdb'.
- Type=<OPT>
- The type of the resource represented by this record. Currently the only valid type is License.
LIST/SHOW RESOURCE FORMAT OPTIONS¶
- Cluster
- Name of cluster resource is given to.
- Count
- The count of a specific resource configured on the system globally.
- CountAllowed
- The count of a specific resource configured on the system to a cluster.
- CountUsed
- The used count of a specific resource configured on the system globally.
- Description
- Description of the resource.
- Flags
- Flags that identify specific attributes associated with the resource. At
this time no flags have been defined.
.TP Manager Name of the resource manager that controls this resource.
- Name
- Name of this resource.
- PercentAllowed
- The percent of a specific resource given to a cluster.
- PercentUsed
- The used percentage of a specific resource globally.
.TP Server Server serving up the resource.
- Type
- Type of resource this record represents.
SPECIFICATIONS FOR TRANSACTIONS¶
- Accounts=<comma separated list of account names>
- Only print out the transactions affecting specified accounts.
- Action=<Specific action the list will display>
- Actor=<Specific name the list will display>
- Only display transactions done by a certain person.
- Clusters=<comma separated list of cluster names>
- Only print out the transactions affecting specified clusters.
- End=<Date and time of last transaction to return>
- Return all transactions before this Date and time. Default is now.
- Start=<Date and time of first transaction to return>
- Return all transactions after this Date and time. Default is epoch.
Valid time formats for End and Start are...
HH:MM[:SS] [AM|PM]
- Users=<comma separated list of user names>
- Only print out the transactions affecting specified users.
- WithAssoc
- Get information about which associations were affected by the transactions.
LIST/SHOW TRANSACTIONS FORMAT OPTIONS¶
- Action
- Actor
- Info
- TimeStamp
- Where
SPECIFICATIONS FOR USERS¶
- Account=<account>
- Account name to add this user to.
- AdminLevel=<level>
- Admin level of user. Valid levels are None, Operator, and Admin.
- Cluster=<cluster>
- Specific cluster to add user to the account on. Default is all in system.
- DefaultAccount=<account>
- Identify the default bank account name to be used for a job if none is specified at submission time.
- DefaultWCKey=<defaultwckey>
- Identify the default Workload Characterization Key.
- Name=<name>
- Name of user.
- Partition=<name>
- Partition name.
- RawUsage=<value>
- This allows an administrator to reset the raw usage accrued to a user. The only value currently supported is 0 (zero). This is a settable specification only - it cannot be used as a filter to list users.
- WCKeys=<wckeys>
- Workload Characterization Key values.
- WithAssoc
- Display all associations for this user.
- WithCoord
- Display all accounts a user is coordinator for.
- WithDeleted
- Display information with previously deleted data.
LIST/SHOW USER FORMAT OPTIONS¶
- AdminLevel
- Admin level of user.
- DefaultAccount
- The user's default account.
- Coordinators
- List of users that are a coordinator of the account. (Only filled in when using the WithCoordinator option.)
- User
- The name of a user.
LIST/SHOW WCKey¶
- WCKey
- Workload Characterization Key.
- Cluster
- Specific cluster for the WCKey.
- User
- The name of a user for the WCKey.
GLOBAL FORMAT OPTION¶
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.FLAT FILE DUMP AND LOAD¶
sacctmgr has the capability to load and dump SLURM association data to and from a file. This method can easily add a new cluster or copy an existing clusters associations into a new cluster with similar accounts. Each file contains SLURM association data for a single cluster. Comments can be put into the file with the # character. Each line of information must begin with one of the four titles; Cluster, Parent, Account or User. Following the title is a space, dash, space, entity value, then specifications. Specifications are colon separated. If any variable such as Organization has a space in it, surround the name with single or double quotes. To create a file of associations one can run > sacctmgr dump tux file=tux.cfg- GrpCPUMins=
- The total number of cpu minutes that can possibly be used by past, present and future jobs running from this association and its children.
- GrpCPURunMins=
- Used to limit the combined total number of CPU minutes used by all jobs running with this association and its children. This takes into consideration time limit of running jobs and consumes it, if the limit is reached no new jobs are started until other jobs finish to allow time to free up.
- GrpCPUs=
- Maximum number of CPUs running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- GrpJobs=
- Maximum number of running jobs in aggregate for this association and all associations which are children of this association.
- GrpNodes=
- Maximum number of nodes running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- GrpSubmitJobs=
- Maximum number of jobs which can be in a pending or running state at any time in aggregate for this association and all associations which are children of this association.
- GrpWall=
- Maximum wall clock time running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- FairShare=
- Number used in conjunction with other associations to determine job priority.
- MaxJobs=
- Maximum number of jobs the children of this association can run.
- MaxNodesPerJob=
- Maximum number of nodes per job the children of this association can run.
- MaxWallDurationPerJob=
- Maximum time (not related to job size) children of this accounts jobs can run.
- QOS=
- Comma separated list of Quality of Service names (Defined in sacctmgr).
- Followed by Accounts you want in this fashion...
-
Parent - root (Defined by default)
- Any of the options after a ':' can be left out and they can be in any order.
- If you want to add any sub accounts just list the Parent THAT HAS ALREADY BEEN CREATED before the account line in this fashion...
- All account options are
- Description=
- A brief description of the account.
- GrpCPUMins=
- Maximum number of CPU hours running jobs are able to be allocated in aggregate for this association and all associations which are children of this association. GrpCPURunMins= Used to limit the combined total number of CPU minutes used by all jobs running with this association and its children. This takes into consideration time limit of running jobs and consumes it, if the limit is reached no new jobs are started until other jobs finish to allow time to free up.
- GrpCPUs=
- Maximum number of CPUs running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- GrpJobs=
- Maximum number of running jobs in aggregate for this association and all associations which are children of this association.
- GrpNodes=
- Maximum number of nodes running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- GrpSubmitJobs=
- Maximum number of jobs which can be in a pending or running state at any time in aggregate for this association and all associations which are children of this association.
- GrpWall=
- Maximum wall clock time running jobs are able to be allocated in aggregate for this association and all associations which are children of this association.
- FairShare=
- Number used in conjunction with other associations to determine job priority.
- MaxJobs=
- Maximum number of jobs the children of this association can run.
- MaxNodesPerJob=
- Maximum number of nodes per job the children of this association can run.
- MaxWallDurationPerJob=
- Maximum time (not related to job size) children of this accounts jobs can run.
- Organization=
- Name of organization that owns this account.
- QOS(=,+=,-=)
- Comma separated list of Quality of Service names (Defined in sacctmgr).
- To add users to a account add a line like this after a Parent - line
- Parent - test
- All user options are
- AdminLevel=
- Type of admin this user is (Administrator, Operator)
- Coordinator=
- Comma separated list of accounts this user is coordinator over
- DefaultAccount=
- system wide default account name
- FairShare=
- Number used in conjunction with other associations to determine job priority.
- MaxJobs=
- Maximum number of jobs this user can run.
- MaxNodesPerJob=
- Maximum number of nodes per job this user can run.
- MaxWallDurationPerJob=
- Maximum time (not related to job size) this user can run.
- QOS(=,+=,-=)
- Comma separated list of Quality of Service names (Defined in sacctmgr).
ARCHIVE FUNCTIONALITY¶
Sacctmgr has the capability to archive to a flatfile and or load that data if needed later. The archiving is usually done by the slurmdbd and it is highly recommended you only do it through sacctmgr if you completely understand what you are doing. For slurmdbd options see "man slurmdbd" for more information. Loading data into the database can be done from these files to either view old data or regenerate rolled up data. These are the options for both dump and load of archive information. archive dump- Directory=
- Directory to store the archive data.
- Events
- Archive Events. If not specified and PurgeEventAfter is set all event data removed will be lost permanently.
- Jobs
- Archive Jobs. If not specified and PurgeJobAfter is set all job data removed will be lost permanently.
- PurgeEventAfter=
- Purge cluster event records older than time stated in months. If you want to purge on a shorter time period you can include hours, or days behind the numeric value to get those more frequent purges. (e.g. a value of '12hours' would purge everything older than 12 hours.)
- PurgeJobAfter=
- Purge job records older than time stated in months. If you want to purge on a shorter time period you can include hours, or days behind the numeric value to get those more frequent purges. (e.g. a value of '12hours' would purge everything older than 12 hours.)
- PurgeStepAfter=
- Purge step records older than time stated in months. If you want to purge on a shorter time period you can include hours, or days behind the numeric value to get those more frequent purges. (e.g. a value of '12hours' would purge everything older than 12 hours.)
- PurgeSuspendAfter=
- Purge job suspend records older than time stated in months. If you want to purge on a shorter time period you can include hours, or days behind the numeric value to get those more frequent purges. (e.g. a value of '12hours' would purge everything older than 12 hours.)
- Script=
- Run this script instead of the generic form of archive to flat files.
- Steps
- Archive Steps. If not specified and PurgeStepAfter is set all step data removed will be lost permanently.
- Suspend
- Archive Suspend Data. If not specified and PurgeSuspendAfter is set all suspend data removed will be lost permanently.
- File=
- File to load into database.
- Insert=
- SQL to insert directly into the database. This should be used very cautiously since this is writing your sql into the database.
EXAMPLES¶
NOTE: There is an order to set up accounting associations. You must define clusters before you add accounts and you must add accounts before you can add users.maxjobs=2 maxwall=30:00
Name
---------
normal
expedite
Cluster Account QOS
-------- ---------- -----
zebra root normal
zebra root normal
zebra g normal
zebra g1 normal
Cluster Account QOS
-------- -------- -------
zebra root normal
zebra root normal
zebra g normal
zebra g1 expedite,normal
Cluster Account QOS
--------- -------- -----
zebra root normal
zebra root normal
zebra g expedite
zebra g1 expedite,normal
Cluster Account QOS
--------- -------- -----
zebra root normal
zebra root normal
zebra g expedite
zebra banana expedite
zebra g1 expedite,normal
COPYING¶
Copyright (C) 2008-2010 Lawrence Livermore National Security. Produced at Lawrence Livermore National Laboratory (cf, DISCLAIMER).SEE ALSO¶
slurm.conf(5), slurmdbd(8)April 2009 | sacctmgr 2.0 |