/usr/share/arc/LSF.pm is in nordugrid-arc-arex 5.4.2-1build1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 | package LSF;
use strict;
our @ISA = ('Exporter');
our @EXPORT_OK = ('cluster_info',
'queue_info',
'jobs_info',
'users_info');
use LogUtils ( 'start_logging', 'error', 'warning', 'debug' );
##########################################
# Saved private variables
##########################################
our $lsf_profile_path;
our $lsf_profile;
our $lshosts_command;
our $bhosts_command;
our $bqueues_command;
our $bqueuesl_command;
our $bjobs_command;
our $lsid_command;
our $total_cpus="0";
##########################################
# Private subs
##########################################
sub lsf_env($$){
my ($path)=shift;
$lsf_profile_path=shift;
$lsf_profile=`source $lsf_profile_path`;
$lshosts_command="$path/lshosts -w";
$bhosts_command = "$path/bhosts -w";
$bqueues_command = "$path/bqueues -w";
$bqueuesl_command = "$path/bqueues -l";
$bjobs_command = "$path/bjobs -W -w";
$lsid_command="$path/lsid";
}
sub totalcpus {
my %lsfnodes;
my $ncpus=0;
if ( $total_cpus eq "0"){
read_lsfnodes(\%lsfnodes);
#calculate totals
foreach my $node (keys %lsfnodes){
if( ($lsfnodes{$node}{"node_status"} eq "ok") || ($lsfnodes{$node}{"node_status"} eq "closed_Full") || ($lsfnodes{$node}{"node_status"} eq "closed_Excl") || ($lsfnodes{$node}{"node_status"} eq "closed_Busy") || ($lsfnodes{$node}{"node_status"} eq "closed_Adm") ){
$ncpus += $lsfnodes{$node}{"node_ncpus"};
}
}
return $ncpus;
}
return $total_cpus;
}
sub read_lsfnodes ($){
my ($hashref) =shift;
my ($node_count) = 0;
my ($cpu_count) = 0;
unless (open LSFHOSTSOUTPUT, "$lshosts_command |") {
debug("Error in executing lshosts command: $lshosts_command");
die "Error in executing lshosts: $lshosts_command\n";
}
while (my $line= <LSFHOSTSOUTPUT>) {
if (! ($line =~ '^HOST_NAME')) {
chomp($line);
my ($nodeid,$OStype,$model,$cpuf,$ncpus,$maxmem,$maxswp)= split(" ", $line);
${$hashref}{$nodeid}{"node_hostname"} = $nodeid;
${$hashref}{$nodeid}{"node_os_type"} = $OStype;
${$hashref}{$nodeid}{"node_model"} = $model;
${$hashref}{$nodeid}{"node_cpuf"} = $cpuf;
${$hashref}{$nodeid}{"node_maxmem"} = $maxmem;
${$hashref}{$nodeid}{"node_max_swap"} = $maxswp;
if($ncpus != "-") {
${$hashref}{$nodeid}{"node_ncpus"} = $ncpus;
}
else {
${$hashref}{$nodeid}{"node_ncpus"} = 1;
}
}
}
close LSFHOSTSOUTPUT;
unless (open LSFBHOSTSOUTPUT, "$bhosts_command |") {
debug("Error in executing bhosts command: $bhosts_command");
die "Error in executing bhosts: $bhosts_command\n";
}
while (my $line= <LSFBHOSTSOUTPUT>) {
if (! ( ($line =~ '^HOST_NAME') || ($line =~ '^My cluster') || ($line =~ '^My master') ) ) {
chomp($line);
# HOST_NAME STATUS JL/U MAX NJOBS RUN SSUSP USUSP RSV
my ($nodeid,$status,$lju,$max,$njobs,$run,$ssusp,$ususp,$rsv) = split(" ", $line);
${$hashref}{$nodeid}{"node_used_slots"} = $njobs;
${$hashref}{$nodeid}{"node_running"} = $run;
${$hashref}{$nodeid}{"node_suspended"} = $ssusp + $ususp;
${$hashref}{$nodeid}{"node_reserved"} = $rsv;
${$hashref}{$nodeid}{"node_status"} = $status;
}
}
close LSFBHOSTSOUTPUT;
}
sub type_and_version {
my (@lsf_version) = `$lsid_command -V 2>&1`;
my ($type) ="LSF";
my ($version)="0.0";
if($lsf_version[0] =~ /^Platform/) {
my @s = split(/ +/,$lsf_version[0]);
$type=$s[1];
$version=$s[2];
$version=~s/,$//;
}
my (@result) = [$type,$version];
return ($type,$version);
}
sub queue_info_user ($$$) {
my ($path) = shift;
my ($qname) = shift;
my ($user) = shift;
my (%lrms_queue);
#calculate running cpus and queues available
unless ($user eq ""){
$user = "-u " . $user;
}
unless (open BQOUTPUT, "$bqueues_command $user $qname|") {
debug("Error in executing bqueues command: $bqueues_command $user $qname");
die "Error in executing bqueues: $bqueues_command \n";
}
while (my $line= <BQOUTPUT>) {
if (! ($line =~ '^QUEUE')) {
chomp($line);
my ($q_name,$q_priority,$q_status,$q_mjobs,$q_mslots,$q_mslots_proc,$q_mjob_slots_host,$q_num_jobs,$q_job_pending,$q_job_running,$q_job_suspended) = split(" ", $line);
$lrms_queue{totalcpus} = "$q_mjobs";
$lrms_queue{maxrunning} = "$q_mjobs";
$lrms_queue{maxqueuable} = "$q_mjobs";
if ($q_mjobs eq "-") {
$lrms_queue{totalcpus} = totalcpus();
$lrms_queue{maxrunning} = $lrms_queue{totalcpus};
$lrms_queue{maxqueuable} = $lrms_queue{totalcpus};
}
$lrms_queue{maxuserrun} = "$q_mslots";
if ($q_mslots eq "-"){
$lrms_queue{maxuserrun} = $lrms_queue{totalcpus};
}
$lrms_queue{running}= $q_job_running;
$lrms_queue{status} = $q_status;
$lrms_queue{queued} = $q_job_pending;
}
}
close BQOUTPUT;
$lrms_queue{defaultcput} = "";
$lrms_queue{defaultwallt} = "";
$lrms_queue{maxcputime} = "";
$lrms_queue{maxwalltime} = "";
unless (open BQOUTPUT, "$bqueuesl_command $user $qname|") {
debug("Error in executing bqueues command: $bqueuesl_command $user $qname");
die "Error in executing bqueues: $bqueuesl_command \n";
}
my $lastline ="";
while (my $line= <BQOUTPUT>) {
if ( ($line =~ '^ CPULIMIT')) {
my $line2=<BQOUTPUT>;
chomp($line2);
my (@mcput)= split(" ", $line2);
#change from float to int.
$mcput[0]=~ s/(\d+).*/$1/;
if ($lastline =~ '^DEFAULT'){
$lrms_queue{defaultcput} = "$mcput[0]";
} else {
$lrms_queue{maxcputime} = "$mcput[0]";
if ($lrms_queue{maxwalltime} == "") {
$lrms_queue{maxwalltime} = "$mcput[0]";
}
}
}
if ( ($line =~ '^ RUNLIMIT')) {
my $line2=<BQOUTPUT>;
chomp($line2);
my (@mcput)= split(" ", $line2);
#change from float to int.
$mcput[0]=~ s/(\d+).*/$1/;
if ($lastline =~ '^DEFAULT'){
$lrms_queue{defaultwallt} = "$mcput[0]";
} else {
$lrms_queue{maxwalltime} = "$mcput[0]";
}
}
$lastline = $line;
}
close BQOUTPUT;
$lrms_queue{mincputime} = "0";
$lrms_queue{minwalltime} = "0";
return %lrms_queue;
}
sub get_jobinfo($){
my $id = shift;
my %job;
unless (open BJOUTPUT, "$bjobs_command $id|") {
debug("Error in executing bjobs command: $bjobs_command $id");
die "Error in executing bjobs: $bjobs_command \n";
}
while (my $line= <BJOUTPUT>) {
if (! ($line =~ '^JOBID')) {
chomp($line);
my ($j_id,$j_user,$j_stat,$j_queue,$j_fromh,$j_exech,$j_name,$j_submittime,$j_projname,$j_cput,$j_mem,$j_swap,$j_pids,$j_start,$j_finish) = split(" ", $line);
$job{id} = $j_id;
# Report one node per job. Needs improoving for multi-CPU jobs.
$job{nodes} = [ $j_exech ];
$job{cput} = $j_cput;
$job{mem} = $j_mem;
$job{start} = $j_start;
$job{finish} = $j_finish;
if ($j_stat eq "RUN"){
$job{status} = "R";
}
if ($j_stat eq "PEND"){
$job{status} = "Q";
}
if ($j_stat eq "PSUSP" || $j_stat eq "USUSP" || $j_stat eq "SSUSP"){
$job{status} = "S";
}
if ($j_stat eq "DONE" || $j_stat eq "EXIT"){
$job{status} = "E";
}
if ($j_stat eq "UNKWN" || $j_stat eq "WAIT" || $j_stat eq "ZOMBI"){
$job{status} = "O";
}
}
}
close BJOUTPUT;
return %job;
}
############################################
# Public subs
#############################################
sub cluster_info ($) {
my ($config) = shift;
lsf_env($$config{lsf_bin_path},$$config{lsf_profile_path});
#init
my %lrms_cluster;
my %lsfnodes;
$lrms_cluster{totalcpus} = 0;
$lrms_cluster{usedcpus} = 0;
$lrms_cluster{queuedcpus} = 0;
$lrms_cluster{runningjobs} = 0;
$lrms_cluster{queuedjobs} = 0;
my @cpudist;
$lrms_cluster{cpudistribution} = "";
$lrms_cluster{queue} = [];
#lookup batch type and version
($lrms_cluster{lrms_type},$lrms_cluster{lrms_version}) = type_and_version();
# cputime limit for parallel/multi-cpu jobs is treated as job-total
# OBS: Assuming LSB_JOB_CPULIMIT=y !
$lrms_cluster{has_total_cputime_limit} = 1;
#get info on nodes in cluster
read_lsfnodes(\%lsfnodes);
#calculate totals
foreach my $node (keys %lsfnodes){
if( ($lsfnodes{$node}{"node_status"} eq "ok") || ($lsfnodes{$node}{"node_status"} eq "closed_Full") || ($lsfnodes{$node}{"node_status"} eq "closed_Excl") || ($lsfnodes{$node}{"node_status"} eq "closed_Busy") || ($lsfnodes{$node}{"node_status"} eq "closed_Adm") ){
my $ncpus = $lsfnodes{$node}{"node_ncpus"};
# we use lshosts output, maybe we should use bhosts?
$lrms_cluster{totalcpus} += $ncpus;
$lrms_cluster{usedcpus} += $lsfnodes{$node}{"node_used_slots"};
$cpudist[$ncpus] += 1;
}
}
#write cpu distribution string of the form: 1cpu:15 4cpu:4
for (my $i=0; $i<=$#cpudist; $i++) {
next unless ($cpudist[$i]);
$lrms_cluster{cpudistribution} .= " " . $i . "cpu:" . $cpudist[$i];
}
#calculate queued cpus and queues available
unless (open BQOUTPUT, "$bqueues_command|") {
debug("Error in executing bqueues command: $bqueues_command ");
die "Error in executing bqueues: $bqueues_command \n";
}
my @queues;
while (my $line= <BQOUTPUT>) {
if (! ($line =~ '^QUEUE')) {
chomp($line);
my ($q_name,$q_priority,$q_status,$q_mjobs,$q_mslots,$q_mslots_proc,$q_mjob_slots_host,$q_num_jobs,$q_job_pending,$q_job_running,$q_job_suspended) = split(" ", $line);
#TODO: total number of jobs in queue is not equal to queued cpus.
$lrms_cluster{queuedcpus}+=$q_num_jobs;
$lrms_cluster{runningjobs}+=$q_job_running;
$lrms_cluster{queuedjobs}+=$q_job_pending;
push @queues, $q_name;
}
}
close BQOUTPUT;
@{$lrms_cluster{queue}} = @queues;
return %lrms_cluster;
}
sub queue_info($$){
my ($config) = shift;
my ($qname) = shift;
lsf_env($$config{lsf_bin_path},$$config{lsf_profile_path});
return queue_info_user($$config{lsf_bin_path},$qname,"");
}
#LSF time is on format: 000:00:00.28
#output should be an integer in minutes, rounded up.
sub translate_time_lsf_to_minutes ($@) {
my ($cputime) = shift;
my ($days,$hours,$rest) = split(/:/,$cputime);
my ($minutes, $seconds)=split(/\./,$rest);
if ( $seconds > 0){
$minutes++;
}
$minutes=$days*24*60+$hours*60+$minutes;
return $minutes;
}
sub jobs_info ($$@) {
my ($config) = shift;
my ($qname) = shift;
my ($jids) = shift;
lsf_env($$config{lsf_bin_path},$$config{lsf_profile_path});
my (%lrms_jobs);
my (%job);
my (@s);
foreach my $id (@$jids){
%job = get_jobinfo($id);
$lrms_jobs{$id}{status}=$job{status};
$lrms_jobs{$id}{nodes}=$job{nodes};
$lrms_jobs{$id}{mem}=$job{mem};
$lrms_jobs{$id}{cputime}=translate_time_lsf_to_minutes($job{cput});
$lrms_jobs{$id}{walltime}="";
$lrms_jobs{$id}{reqwalltime}="";
$lrms_jobs{$id}{reqcputime}="";
$lrms_jobs{$id}{comment}=["job started: $job{start}"];
$lrms_jobs{$id}{rank}="";
#TODO fix to support parallel jobs
$lrms_jobs{$id}{cpus}=1;
}
return %lrms_jobs;
}
sub users_info($$@) {
my ($config) = shift;
my ($qname) = shift;
my ($accts) = shift;
lsf_env($$config{lsf_bin_path},$$config{lsf_profile_path});
my (%lrms_users);
my (%queue);
foreach my $u ( @{$accts} ) {
%queue = queue_info_user( $$config{lsf_bin_path}, $qname, $u );
$lrms_users{$u}{freecpus} = $queue{maxrunning}-$queue{runnning};
$lrms_users{$u}{queuelength} = "$queue{queued}";
}
return %lrms_users;
}
1;
|