31
31
## For example, to connect a computing cluster to the grid and for it to
32
32
## accept grid jobs you should have at least the [common], [grid-manager],
33
## [gridftpd], [gridftpd/jobs], [infosys], [cluster], [queue/name]
33
## [gridftpd], [gridftpd/jobs], [infosys], [cluster], [queue/queue_name]
34
34
## [infosys/cluster/registration/name] blocks configured.
36
36
## As another example we give the necessary blocks for an Index Service:
66
66
#hostname="myhost.org"
69
## This file shows all possible lrms options.
70
## Please refer to ARC documentation for possible combinations of these
69
## ARC supports various LRMS flavours, as listed in this section. For detailed
70
## description of options please refer to ARC CE sysadmin guide:
71
## http://www.nordugrid.org/documents/arc-ce-sysadm-guide.pdf
72
72
## ONLY ONE LRMS IS ALLOWED. MULTIPLE lrms ENTRIES WILL TRIGGER UNEXPECTED
75
## lrms [queue] - sets the type of the Local Resource Management System
76
## (queue system). Chose one from the following options: fork, sge,
77
## condor, pbs, lsf, ll, slurm. PBS has many flavours, we support
78
## OpenPBS, PBSPro, ScalablePBS and Torque (the official name for
79
## ScalablePBS). No need to specify the flavour or the version number
80
## of the PBS, simply write 'pbs'. <lrms> MUST be set here! The
81
## optional "queue" parameter specifies the default grid queue of the
82
## LRMS. The GM submits jobs to this queue if there is no queue set in
75
## lrms sets the type of the Local Resource Management System (queue system),
76
## and optionally - the default queue name, separated with a blank space:
77
## "lrmstype queue_name". For lrmstype, the following systems are supported
78
## and can be chosen (one per server):
79
## fork - simple forking of jobs to the same node as the server
80
## sge - (Sun/Oracle) Grid Engine
86
## PBS has many flavours, ARC currenly supports OpenPBS, PBSPro, ScalablePBS
87
## and Torque (the official name for ScalablePBS). There is no need to specify
88
## the flavour or the version number of the PBS, simply write 'pbs'. Similarly,
89
## there is no need to specify (Sun/Oracle) Grid Engine versions and flavours.
90
## "lrmstype" MUST be set here, it is a MANDATORY parameter!
91
## The optional "queue" parameter specifies the default Grid queue of the
92
## LRMS. Jobs will be submitted to this queue if they do not specify queue
93
## name in job description. Queue name must match one of the [queue/queue_name]
94
## block labels, see below.
88
99
## the path to the qstat,pbsnodes,qmgr etc PBS binaries,
1074
1085
## version is required for GLUE2 output.
1075
1086
#infosys_compat="disable"
1088
## infoproviders_timeout - this only applies to new infoproviders.
1089
## it changes A-REX behaviour with respect to a single infoprovider run.
1090
## Increase this value if you have many jobs in the controldir and
1091
## infoproviders need more time to process.
1092
## The value is in seconds.
1093
## Default is 600 seconds.
1094
# infoproviders_timeout = "600"
1077
1096
## debug - sets the debug level/verbosity of the startup script {0 or 1}.
1078
1097
## Default is 0.
1231
1250
#timelimit="30"
1232
1251
#sizelimit="10"
1253
## LDAP checkpoint enable/disable
1254
## This option was introduced to solve bug #2032, to reduce the number
1255
## of log files produced by BDII. It is usually not needed, but if BDII produces
1256
## large logs and huge number of files, try to enable it.
1257
# slapd_cron_checkpoint="enable"
1235
1260
####################################################################
1617
1642
####################################################################
1619
### [queue/queuename] block ########################################
1644
### [queue/queue_name] block #######################################
1621
1646
## Each grid-enabled queue should have a separate queue block.
1622
1647
## The queuename should be used as a label in the block name.
1623
## A queue can represent a PBS queue, a SGE pool, a Condor pool or a machine
1624
## with a 'fork' LRMS.
1648
## A queue can represent a PBS/LSF/SGE/SLURM/LL queue, a SGE pool, a Condor
1649
## pool or a single machine in case 'fork' type of LRMS is specified in
1650
## the [common] block.
1625
1651
## Queues don't need to be registerd (there is no queue registration block),
1626
## once you configured your cluster to register to a Index Service
1652
## once you configured your cluster to register to an Index Service
1627
1653
## the queue entries (configured with this block) automatically will be there.
1628
## Please consult the Infosys manual for detailed information
1629
## on queue attributes.
1654
## Please consult the ARC Information System manual for detailed information
1655
## on queue attributes:
1656
## http://www.nordugrid.org/documents/arc_infosys.pdf
1631
1658
## use the queue_name for labeling the block. The special name 'fork' should be
1632
## used for labeling the queue block of the fork LRMS.
1659
## used for labeling the queue block in case you specified 'fork' type of
1660
## LRMS in the [common] block.
1633
1661
#[queue/gridlong]
1635
## The name of the grid-enabled queue, it is also used as a queue block label
1636
## [queue/queuename].
1637
## Use "fork" for the fork LRMS. <MUST be specified>
1663
## name sets the name of the grid-enabled queue. It MUST match the queue_name
1664
## label of the corresponding queue block, see above.
1665
## Use "fork" if you specified 'fork' type of LRMS in the [common] block.
1666
## Queue name MUST be specified, even if the queue block is already correctly
1640
1670
## fork_job_limit, the allowed number of concurrent jobs in a fork system,
1641
1671
## relevant only for a fork queue. Default is 1. The special value 'cpunumber'
1673
1703
## queue_node_string for that queue). This parameter should contain the "node
1674
1704
## property" string of the queue-assigned nodes. Setting the queue_node_string
1675
1705
## changes how the queue-totalcpus, user freecpus are determined for this queue.
1676
## Read the pbs_config.txt for more info. You shouldn't use this option unless
1677
## you make sure that your PBS config makes use of the above configuration.
1706
## Essentially, queue_node_string value is used to construct nodes= string in
1707
## PBS script, such as nodes=count:queue_node_string where count is taken from
1708
## the job description (1 if not specified). You shouldn't use this option unless
1709
## you are sure that your PBS configuration makes use of the above configuration.
1710
## Read NorduGrid PBS instructions for more information:
1711
## http://www.nordugrid.org/documents/pbs-config.html
1678
1712
#queue_node_string="gridlong_nodes"
1713
#queue_node_string="ppn=4:ib"
1680
1715
## sge_jobopts - additional SGE options to be used when submitting jobs to SGE
1681
1716
## from this queue. If in doubt, leave it commented out