From 7a7b8a2d95e32fb9a46a164e88850407cbda952e Mon Sep 17 00:00:00 2001 From: Brian Christiansen <brian@schedmd.com> Date: Thu, 26 Jan 2017 10:11:13 -0700 Subject: [PATCH] Revert "Remove sacctmgr federation docs" This reverts commit d0d6415895d5eed8f9adde3e7a4b0a6a91524428. --- doc/man/man1/sacctmgr.1 | 129 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 129 insertions(+) diff --git a/doc/man/man1/sacctmgr.1 b/doc/man/man1/sacctmgr.1 index 83726eeff79..a95dce512f3 100644 --- a/doc/man/man1/sacctmgr.1 +++ b/doc/man/man1/sacctmgr.1 @@ -222,6 +222,10 @@ account and user associations inside their realm. \fIevent\fR Events like downed or draining nodes on clusters. +.TP +\fIfederation\fP +A group of clusters that work together to schedule jobs. + .TP \fIjob\fR Used to modify specific fields of a job: Derived Exit Code and @@ -700,6 +704,37 @@ The name of a user in the association. \fIClassification\fP=<classification> Type of machine, current classifications are capability and capacity. +.TP +\fIFederation\fP=<federation> +The federation that this cluster should be a member of. A cluster can only be a +member of one federation at a time. + +.TP +\fIFedState\fP=<state> +The state of the cluster in the federation. +.br +Valid states are: +.RS +.TP +\fIACTIVE\fP +Cluster will actively accept and schedule federated jobs. + +.TP +\fIINACTIVE\fP +Cluster will not schedule or accept any jobs. + +.TP +\fIDRAIN\fP +Cluster will not accept any new jobs and will let existing federated jobs +complete. + +.TP +\fIDRAIN+REMOVE\fP +Cluster will not accept any new jobs and will remove itself from the federation +once all federated jobs have completed. When removed from the federation, the +cluster will accept jobs as a non\-federated cluster. +.RE + .TP \fIFlags\fP=<flag list> Comma separated list of Attributes for a particular cluster. Current @@ -715,6 +750,18 @@ configuration file for some Slurm\-managed cluster. \fIRPC\fP=<rpc list> Comma separated list of numeric RPC values. +.TP +\fIWeight\fP=# +Weight of the cluster in a federation. When submitting sibling jobs, clusters +that can run a job immediately will be prioritized and chosen according to their +weight. Weights can be used to favor jobs starting on specific clusters. Default +is 1. + +.TP +\fIWithFed\fP +Appends federation related columns to default format options +(e.g. Federation,ID,Weight,State). + .TP \fIWOLimits\fP Display information without limit information. This is for a smaller @@ -744,10 +791,27 @@ controller is placed here. When a slurmctld registers with the database the port the controller is listening on is placed here. +.TP +\fIFederation\fP +The name of the federation this cluster is a member of (if any). + +.TP +\fIFedState\fP +The state of the cluster in the federation (if a member of one). + +.TP +\fIFedStateRaw\fP +Numeric value of the name of the FedState. + .TP \fIFlags\fP Attributes possessed by the cluster. +.TP +\fIID\fP +The ID assigned to the cluster when a member of a federation. This ID uniquely +identifies the cluster and its jobs in the federation. + .TP \fINodeCount\fP The current count of nodes associated with the cluster. @@ -770,6 +834,10 @@ is running is placed here. Trackable RESources (BB (Burst buffer), CPU, Energy, GRES, License, Memory, and Node) this cluster is accounting for. +.TP +\fIWeight\fP +The weight of the cluster when a member of a federation. + .P NOTE: You can also view the information about the root association for the cluster. The Association format fields are described @@ -923,6 +991,67 @@ Number of TRES involved with the event. On a node event this is the user who caused the event to happen. +.SH "SPECIFICATIONS FOR FEDERATION" + +.TP +\fIClusters\fP[+|\-]=<comma separated list of cluster names> +List of clusters to add/remove to a federation. A blank value (e.g. clusters=) +will remove all federations for the federation. NOTE: a cluster can only be a +member of one federation. + +.TP +\fIName\fP=<name> +The name of the federation. + +.TP +\fIFlags\fP[+|\-]=<flags> +Used by the slurmctld to provide different scheduling behavior in the +federation. +.br +Valid options are: +.RS +.TP +\fILLC\fP +At submission time, favor starting sibling jobs that can run immediately on +clusters that are least loaded (ie. spread jobs across clusters). By default, +Slurm will pack jobs onto clusters in the federation. +.RE + +.TP +\fITree\fP +Display federations in a hierarchical fashion. + +.SH "LIST/SHOW FEDERATION FORMAT OPTIONS" + +.TP +\fIFederation\fP +The name of the federation. + +.TP +\fIFlags\fP +Flags for the federation. + +.TP +\fICluster\fP +Name of the cluster that is a member of the federation. + +.TP +\fIFedState\fP +The state of the cluster in the federation. + +.TP +\fIFedStateRaw\fP +Numeric value of the name of the FedState. + +.TP +\fIIndex\fP +The index of the cluster in the federation. + +.TP +\fIWeight\fP +The weight of the cluster in the federation. + + .SH "SPECIFICATIONS FOR JOB" .TP -- GitLab