Queen.pm 24.5 KB
Newer Older
Jessica Severin's avatar
Jessica Severin committed
1 2 3 4 5 6 7 8
#
# You may distribute this module under the same terms as perl itself
#
# POD documentation - main docs before the code

=pod 

=head1 NAME
9

10
  Bio::EnsEMBL::Hive::Queen
Jessica Severin's avatar
Jessica Severin committed
11 12

=head1 SYNOPSIS
13

14
  The Queen of the Hive based job control system
Jessica Severin's avatar
Jessica Severin committed
15 16

=head1 DESCRIPTION
17

18 19 20 21 22 23 24 25 26 27 28 29
  The Queen of the Hive based job control system is responsible to 'birthing' the
  correct number of workers of the right type so that they can find jobs to do.
  It will also free up jobs of Workers that died unexpectantly so that other workers
  can claim them to do.

  Hive based processing is a concept based on a more controlled version
  of an autonomous agent type system.  Each worker is not told what to do
  (like a centralized control system - like the current pipeline system)
  but rather queries a central database for jobs (give me jobs).

  Each worker is linked to an analysis_id, registers its self on creation
  into the Hive, creates a RunnableDB instance of the Analysis->module,
30
  gets $worker->batch_size() jobs from the analysis_job table, does its
31
  work, creates the next layer of analysis_job entries by interfacing to
32 33 34
  the DataflowRuleAdaptor to determine the analyses it needs to pass its
  output data to and creates jobs on the next analysis database.
  It repeats this cycle until it has lived its lifetime or until there are no
35 36 37 38 39 40 41 42 43 44
  more jobs left.
  The lifetime limit is just a safety limit to prevent these from 'infecting'
  a system.

  The Queens job is to simply birth Workers of the correct analysis_id to get the
  work down.  The only other thing the Queen does is free up jobs that were
  claimed by Workers that died unexpectantly so that other workers can take
  over the work.

  The Beekeeper is in charge of interfacing between the Queen and a compute resource
45
  or 'compute farm'.  Its job is to query Queens if they need any workers and to
46 47 48
  send the requested number of workers to open machines via the runWorker.pl script.
  It is also responsible for interfacing with the Queen to identify worker which died
  unexpectantly.
Jessica Severin's avatar
Jessica Severin committed
49 50

=head1 CONTACT
51

52 53
  Contact Jessica Severin on EnsEMBL::Hive implemetation/design detail: jessica@ebi.ac.uk
  Contact Ewan Birney on EnsEMBL in general: birney@sanger.ac.uk
Jessica Severin's avatar
Jessica Severin committed
54 55

=head1 APPENDIX
56

57 58
  The rest of the documentation details each of the object methods. 
  Internal methods are usually preceded with a _
Jessica Severin's avatar
Jessica Severin committed
59 60 61 62 63 64

=cut

package Bio::EnsEMBL::Hive::Queen;

use strict;
65
use POSIX;
66 67 68
use Bio::EnsEMBL::Utils::Argument;
use Bio::EnsEMBL::Utils::Exception;

Jessica Severin's avatar
Jessica Severin committed
69 70 71
use Bio::EnsEMBL::Hive::Worker;
use Bio::EnsEMBL::DBSQL::BaseAdaptor;
use Sys::Hostname;
72
use Bio::EnsEMBL::Hive::DBSQL::AnalysisCtrlRuleAdaptor;
Jessica Severin's avatar
Jessica Severin committed
73

74

Jessica Severin's avatar
Jessica Severin committed
75 76
our @ISA = qw(Bio::EnsEMBL::DBSQL::BaseAdaptor);

77
############################
Jessica Severin's avatar
Jessica Severin committed
78
#
79
# PUBLIC API for Workers
Jessica Severin's avatar
Jessica Severin committed
80
#
81
############################
Jessica Severin's avatar
Jessica Severin committed
82 83 84

=head2 create_new_worker

85
  Arg [1]    : $analysis_id (optional)
Jessica Severin's avatar
Jessica Severin committed
86
  Example    :
87 88 89 90 91 92
  Description: If analysis_id is specified it will try to create a worker based
               on that analysis.  If not specified the queen will analyze the hive
               and pick the analysis that has the most amount of work to be done.
               It creates an entry in the hive table, and returns a Worker object 
               based on that insert.  This guarantees that each worker registered
               in this queens hive is properly registered.
Jessica Severin's avatar
Jessica Severin committed
93 94 95 96 97 98 99
  Returntype : Bio::EnsEMBL::Hive::Worker
  Exceptions :
  Caller     :

=cut

sub create_new_worker {
100 101
  my ($self, @args) = @_;

102 103
  my ($analysis_id, $beekeeper ,$pid) =
     rearrange([qw(analysis_id beekeeper process_id) ], @args);
Jessica Severin's avatar
Jessica Severin committed
104

105
  my $analStatsDBA = $self->db->get_AnalysisStatsAdaptor;
106
  return undef unless($analStatsDBA);
107

108
  return undef if($self->get_hive_current_load() >= 1.5);
109
  
110 111 112 113
  my $analysisStats;
  if($analysis_id) {
    $analysisStats = $analStatsDBA->fetch_by_analysis_id($analysis_id);
  } else {
114
    $analysisStats = $self->_pick_best_analysis_for_new_worker;
115
  }
116 117
  
  return undef unless($analysisStats);
118
  $analStatsDBA->decrement_needed_workers($analysisStats->analysis_id);
119 120 121 122 123 124 125 126 127
  $analysisStats->print_stats;
  
  if($analysisStats->status eq 'BLOCKED') {
    print("Analysis is BLOCKED, can't create workers\n");
    return undef;
  }
  if($analysisStats->status eq 'DONE') {
    print("Analysis is DONE, don't need to create workers\n");
    return undef;
Jessica Severin's avatar
Jessica Severin committed
128 129 130
  }

  my $host = hostname;
131
  $pid = $$ unless($pid);
132
  $beekeeper = '' unless($beekeeper);
Jessica Severin's avatar
Jessica Severin committed
133

134 135 136
  my $sql = q{INSERT INTO hive 
              (born, last_check_in, process_id, analysis_id, beekeeper, host)
              VALUES (NOW(), NOW(), ?,?,?,?)};
Jessica Severin's avatar
Jessica Severin committed
137 138

  my $sth = $self->prepare($sql);
139
  $sth->execute($pid, $analysisStats->analysis_id, $beekeeper, $host);
Jessica Severin's avatar
Jessica Severin committed
140 141 142 143 144 145
  my $hive_id = $sth->{'mysql_insertid'};
  $sth->finish;

  my $worker = $self->_fetch_by_hive_id($hive_id);
  $worker=undef unless($worker and $worker->analysis);

146
  if($worker and $analysisStats) {
147
    $analysisStats->update_status('WORKING');
Jessica Severin's avatar
Jessica Severin committed
148 149 150 151 152 153 154 155 156
  }
  return $worker;
}


sub register_worker_death {
  my ($self, $worker) = @_;

  return unless($worker);
157 158 159 160

  # if called without a defined cause_of_death, assume catastrophic failure
  $worker->cause_of_death('FATALITY') unless(defined($worker->cause_of_death));
  
Jessica Severin's avatar
Jessica Severin committed
161 162 163 164 165 166 167 168
  my $sql = "UPDATE hive SET died=now(), last_check_in=now()";
  $sql .= " ,work_done='" . $worker->work_done . "'";
  $sql .= " ,cause_of_death='". $worker->cause_of_death ."'";
  $sql .= " WHERE hive_id='" . $worker->hive_id ."'";

  my $sth = $self->prepare($sql);
  $sth->execute();
  $sth->finish;
169 170

  if($worker->cause_of_death eq "NO_WORK") {
171
    $self->db->get_AnalysisStatsAdaptor->update_status($worker->analysis->dbID, "ALL_CLAIMED");
172 173 174
  }
  if($worker->cause_of_death eq "FATALITY") {
    #print("FATAL DEATH Arrrrgggghhhhhhhh (hive_id=",$worker->hive_id,")\n");
175
    $self->db->get_AnalysisJobAdaptor->reset_dead_jobs_for_worker($worker);
176
  }
177
  
178
  # re-sync the analysis_stats when a worker dies as part of dynamic sync system
179
  if($self->safe_synchronize_AnalysisStats($worker->analysis->stats)->status ne 'DONE') {
180 181 182 183 184
    # since I'm dying I should make sure there is someone to take my place after I'm gone ...
    # above synch still sees me as a 'living worker' so I need to compensate for that
    $self->db->get_AnalysisStatsAdaptor->increment_needed_workers($worker->analysis->dbID);
  }

Jessica Severin's avatar
Jessica Severin committed
185 186 187
}


188
sub worker_check_in {
Jessica Severin's avatar
Jessica Severin committed
189 190 191 192
  my ($self, $worker) = @_;

  return unless($worker);
  my $sql = "UPDATE hive SET last_check_in=now()";
193
  $sql .= " ,work_done='" . $worker->work_done . "'";
Jessica Severin's avatar
Jessica Severin committed
194 195 196 197 198
  $sql .= " WHERE hive_id='" . $worker->hive_id ."'";

  my $sth = $self->prepare($sql);
  $sth->execute();
  $sth->finish;
199
  
200
  $self->safe_synchronize_AnalysisStats($worker->analysis->stats);
Jessica Severin's avatar
Jessica Severin committed
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
=head2 worker_grab_jobs

  Arg [1]           : Bio::EnsEMBL::Hive::Worker object $worker
  Example: 
    my $jobs  = $queen->worker_grab_jobs();
  Description: 
    For the specified worker, it will search available jobs, 
    and using the workers requested batch_size, claim/fetch that
    number of jobs, and then return them.
  Returntype : 
    reference to array of Bio::EnsEMBL::Hive::AnalysisJob objects
  Exceptions :
  Caller     :

=cut

sub worker_grab_jobs {
  my $self            = shift;
  my $worker          = shift;
  
  my $jobDBA = $self->db->get_AnalysisJobAdaptor;
  my $claim = $jobDBA->claim_jobs_for_worker($worker);
  my $jobs = $jobDBA->fetch_by_claim_analysis($claim, $worker->analysis->dbID);
  return $jobs;
}


=head2 grab_job_by_dbID

  Arg [1]: int $analysis_job_id
  Example: 
    my $job = $queen->grab_job_by_dbID($analysis_job_id);
  Description: 
    For the specified analysis_job_id it will fetch just that job, 
    reclaim it and return it.  Specifying a specific job bypasses 
    the safety checks, thus multiple workers could be running the 
    same job simultaneously (use only for debugging).
  Returntype : 
    Bio::EnsEMBL::Hive::AnalysisJob objects
  Exceptions :
  Caller     :

=cut

sub grab_job_by_dbID {
  my $self            = shift;
  my $analysis_job_id = shift;

  return undef unless($analysis_job_id);
    
  my $jobDBA = $self->db->get_AnalysisJobAdaptor;
  printf("fetching job for id ", $analysis_job_id, "\n");
  my $job = $jobDBA->fetch_by_dbID($analysis_job_id);
  return undef unless($job);
  
  $job->hive_id(0);
  $jobDBA->reclaim_job($job);
  $self->db->get_AnalysisStatsAdaptor->update_status($job->analysis_id, 'LOADING');
  return $job;
}


sub worker_register_job_done {
  my $self = shift;
  my $worker = shift;
  my $job = shift;
  
  return unless($job);
  return unless($job->dbID and $job->adaptor and $job->hive_id);
  return unless($worker and $worker->analysis and $worker->analysis->dbID);
  
  # create_next_jobs
  my $rules = $self->db->get_DataflowRuleAdaptor->fetch_from_analysis_job($job);
  foreach my $rule (@{$rules}) {
    Bio::EnsEMBL::Hive::DBSQL::AnalysisJobAdaptor->CreateNewJob (
        -input_id       => $job->input_id,
        -analysis       => $rule->to_analysis,
        -input_job_id   => $job->dbID,
    );
  }
 
  $job->update_status('DONE');

}


######################################
#
# Public API interface for beekeeper
#
######################################

296 297 298 299 300 301 302 303 304 305
sub fetch_overdue_workers {
  my ($self,$overdue_secs) = @_;

  $overdue_secs = 3600 unless(defined($overdue_secs));

  my $constraint = "h.cause_of_death='' ".
                   "AND (UNIX_TIMESTAMP()-UNIX_TIMESTAMP(h.last_check_in))>$overdue_secs";
  return $self->_generic_fetch($constraint);
}

Jessica Severin's avatar
Jessica Severin committed
306

307 308 309 310 311 312 313 314 315 316 317 318 319
=head2 synchronize_hive

  Example    : $queen->synchronize_hive();
  Description: Runs through all analyses in the system and synchronizes
              the analysis_stats summary with the states in the analysis_job 
              and hive tables.  Then follows by checking all the blocking rules
              and blocks/unblocks analyses as needed.
  Exceptions : none
  Caller     : general

=cut

sub synchronize_hive {
320 321
  my $self = shift;

322
  my $start_time = time();
323

324 325 326 327 328 329
  my $allAnalysis = $self->db->get_AnalysisAdaptor->fetch_all;
  print("analyze ", scalar(@$allAnalysis), "\n");
  foreach my $analysis (@$allAnalysis) {
    my $stats = $analysis->stats;
    $self->synchronize_AnalysisStats($stats);
  }
330
  foreach my $analysis (@$allAnalysis) {
331 332 333 334
    $self->check_blocking_control_rules_for_AnalysisStats($analysis->stats);
  }
  print((time() - $start_time), " secs to synchronize_hive\n");
}
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
=head2 safe_synchronize_AnalysisStats

  Arg [1]    : Bio::EnsEMBL::Hive::AnalysisStats object
  Example    : $self->synchronize($analysisStats);
  Description: Prewrapper around synchronize_AnalysisStats that does
               checks and grabs sync_lock before proceeding with sync.
               Used by distributed worker sync system to avoid contention.
  Exceptions : none
  Caller     : general

=cut

sub safe_synchronize_AnalysisStats {
  my $self = shift;
  my $stats = shift;

  return $stats unless($stats);
  return $stats unless($stats->analysis_id);
  return $stats if($stats->status eq 'SYNCHING');
  return $stats if($stats->status eq 'DONE');
  return $stats if($stats->sync_lock);
  return $stats if(($stats->status eq 'WORKING') and
                   ($stats->seconds_since_last_update < 5*60));

  # OK try to claim the sync_lock
  my $sql = "UPDATE analysis_stats SET status='SYNCHING', sync_lock=1 ".
            "WHERE sync_lock=0 and analysis_id=" . $stats->analysis_id;
  print("$sql\n");
  my $row_count = $self->dbc->do($sql);  
  return $stats unless($row_count == 1);
  printf("got sync_lock on analysis_stats(%d)\n", $stats->analysis_id);
  
  #OK have the lock, go and do the sync
  $self->synchronize_AnalysisStats($stats);
  
  return $stats;
}


376
=head2 synchronize_AnalysisStats
377

378 379 380 381 382 383 384 385
  Arg [1]    : Bio::EnsEMBL::Hive::AnalysisStats object
  Example    : $self->synchronize($analysisStats);
  Description: Queries the analysis_job and hive tables to get summary counts
               and rebuilds the AnalysisStats object.  Then updates the
               analysis_stats table with the new summary info
  Returntype : newly synced Bio::EnsEMBL::Hive::AnalysisStats object
  Exceptions : none
  Caller     : general
386

387
=cut
388

389
sub synchronize_AnalysisStats {
390
  my $self = shift;
391
  my $analysisStats = shift;
392

393
  return $analysisStats unless($analysisStats);
394
  return $analysisStats unless($analysisStats->analysis_id);
395
  
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413
  $analysisStats->total_job_count(0);
  $analysisStats->unclaimed_job_count(0);
  $analysisStats->done_job_count(0);
  $analysisStats->failed_job_count(0);
  $analysisStats->num_required_workers(0);

  my $sql = "SELECT status, count(*) FROM analysis_job ".
            "WHERE analysis_id=? GROUP BY status";
  my $sth = $self->prepare($sql);
  $sth->execute($analysisStats->analysis_id);

  while (my ($status, $count)=$sth->fetchrow_array()) {

    my $total = $analysisStats->total_job_count();
    $analysisStats->total_job_count($total + $count);

    if($status eq 'READY') {
      $analysisStats->unclaimed_job_count($count);
414 415 416
      my $numWorkers = POSIX::ceil($count * $analysisStats->avg_msec_per_job / 300000); 
        # guess num needed workers by total jobs / (num jobs a worker could do in 5 minutes)
      $numWorkers=$count if($numWorkers==0);
417 418 419 420 421 422 423 424 425
      if($analysisStats->hive_capacity>0 and $numWorkers > $analysisStats->hive_capacity) {
        $numWorkers=$analysisStats->hive_capacity;
      }
      $analysisStats->num_required_workers($numWorkers);
    }
    if($status eq 'DONE') { $analysisStats->done_job_count($count); }
    if($status eq 'FAILED') { $analysisStats->failed_job_count($count); }
  }
  $sth->finish;
426 427

  $analysisStats->determine_status();
428

429 430 431 432
  #
  # adjust_stats_for_living_workers
  #
  
433 434 435 436 437 438 439 440 441 442 443 444 445 446
  if($analysisStats->hive_capacity > 0) {
    my $sql = "SELECT count(*) FROM hive WHERE cause_of_death='' and analysis_id=?";
    $sth = $self->prepare($sql);
    $sth->execute($analysisStats->analysis_id);
    my($liveCount)=$sth->fetchrow_array();
    $sth->finish;

    my $numWorkers = $analysisStats->num_required_workers;

    my $capacityAdjust = ($numWorkers + $liveCount) - $analysisStats->hive_capacity;
    $numWorkers -= $capacityAdjust if($capacityAdjust > 0);
    $numWorkers=0 if($numWorkers<0);

    $analysisStats->num_required_workers($numWorkers);
447 448 449 450
  }
  
  $analysisStats->update;
  
451 452
  $self->check_blocking_control_rules_for_AnalysisStats($analysisStats);
  
453
  return $analysisStats;
454
}
455 456


457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479
sub check_blocking_control_rules_for_AnalysisStats
{
  my $self = shift;
  my $stats = shift;
  
  return unless($stats);

  #print("check ctrl on analysis ");  $stats->print_stats;
  my $ctrlRules = $self->db->get_AnalysisCtrlRuleAdaptor->
                  fetch_by_ctrled_analysis_id($stats->analysis_id);
  my $allRulesDone = 1;
  if(scalar @$ctrlRules > 0) {
    #print("HAS blocking_ctrl_rules to check\n");
    foreach my $ctrlrule (@{$ctrlRules}) {
      #use this method because the condition_analysis objects can be
      #network distributed to a different database so use it's adaptor to get
      #the AnalysisStats object
      #$ctrlrule->print_rule;
      my $condAnalysis = $ctrlrule->condition_analysis;
      my $condStats = $condAnalysis->stats if($condAnalysis);
      $allRulesDone = 0 unless($condStats and $condStats->status eq 'DONE');
      #print("  "); $condStats->print_stats;
    }
480

481 482 483
    if($allRulesDone) {
      if($stats->status eq 'BLOCKED') {
        #print("  UNBLOCK analysis : all conditions met\n");
484
        $stats->update_status('LOADING'); #trigger sync
485 486 487 488
      }
    } else {
      #print("  RE-BLOCK analysis : some conditions failed\n");
      $stats->update_status('BLOCKED');
489 490
    }
  }
491 492 493
}


494 495 496
sub get_hive_current_load {
  my $self = shift;
  my $sql = "SELECT sum(1/analysis_stats.hive_capacity) FROM hive, analysis_stats ".
497 498
            "WHERE hive.analysis_id=analysis_stats.analysis_id and cause_of_death ='' ".
            "AND analysis_stats.hive_capacity>0";
499 500 501 502 503 504 505 506 507 508
  my $sth = $self->prepare($sql);
  $sth->execute();
  (my $load)=$sth->fetchrow_array();
  $sth->finish;
  $load=0 unless($load);
  print("current hive load = $load\n");
  return $load;
}


509 510 511 512 513 514 515 516 517 518 519 520
sub get_num_running_workers {
  my $self = shift;
  my $sql = "SELECT count(*) FROM hive WHERE cause_of_death =''";
  my $sth = $self->prepare($sql);
  $sth->execute();
  (my $runningCount)=$sth->fetchrow_array();
  $sth->finish;
  $runningCount=0 unless($runningCount);
  print("current hive num_running_workers = $runningCount\n");
  return $runningCount;
}

521 522 523 524 525 526 527
=head2 get_num_needed_workers

  Example    : $count = $queen->get_num_needed_workers();
  Description: Runs through the analyses in the system which are waiting
               for workers to be created for them.  Calculates the maximum
               number of workers needed to fill the current needs of the system
  Exceptions : none
528
  Caller     : beekeepers and other external processes
529 530 531

=cut

532
sub get_num_needed_workers {
533 534
  my $self = shift;

535 536 537 538 539
  my $statsDBA = $self->db->get_AnalysisStatsAdaptor;
  my $neededAnals = $statsDBA->fetch_by_needed_workers();
  my $deeper_stats_list = $statsDBA->fetch_by_status('LOADING', 'BLOCKED');
  push @$neededAnals, @$deeper_stats_list;

540
  return 0 unless($neededAnals);
541

542 543 544 545 546
  my $availableLoad = 1.0 - $self->get_hive_current_load();
  return 0 if($availableLoad <0.0);

  my $numWorkers = 0;
  foreach my $analysis_stats (@{$neededAnals}) {
547 548
    #$analysis_stats->print_stats();

549 550 551 552 553 554 555 556 557
    #digging deeper under the surface so need to sync
    if(($analysis_stats->status eq 'LOADING') or ($analysis_stats->status eq 'BLOCKED')) {
      $self->synchronize_AnalysisStats($analysis_stats);
      $self->check_blocking_control_rules_for_AnalysisStats($analysis_stats);
    }

    next if($analysis_stats->status eq 'BLOCKED');
    next if($analysis_stats->num_required_workers == 0);

558 559 560 561 562 563
    my $thisLoad = 0.0;
    if($analysis_stats->hive_capacity>0) {
      $thisLoad = $analysis_stats->num_required_workers * (1/$analysis_stats->hive_capacity);
    }

    if(($analysis_stats->hive_capacity<=0) or ($thisLoad < $availableLoad)) {
564 565 566
      $numWorkers += $analysis_stats->num_required_workers;
      $availableLoad -= $thisLoad;
      $analysis_stats->print_stats();
567
      printf("  %5d workers (%1.3f remaining-hive-load)\n", $numWorkers, $availableLoad);
568
    } else {
569
      my $workerCount = POSIX::ceil($availableLoad * $analysis_stats->hive_capacity);
570 571 572
      $numWorkers += $workerCount;
      $availableLoad -=  $workerCount * (1/$analysis_stats->hive_capacity);
      $analysis_stats->print_stats();
573
      printf("  %5d workers (%1.3f remaining-hive-load) use only %3d workers\n", $numWorkers, $availableLoad, $workerCount);
574
      last;
575
    }
576
    last if($availableLoad <= 0.0);
577
  }
578

579
  printf("need $numWorkers workers (availLoad=%1.5f)\n", $availableLoad);
580
  return $numWorkers;
581 582 583
}


584 585 586 587 588 589 590 591
sub get_hive_progress
{
  my $self = shift;
  my $sql = "SELECT sum(done_job_count ), sum(total_job_count) FROM analysis_stats";
  my $sth = $self->prepare($sql);
  $sth->execute();
  my ($done, $total) = $sth->fetchrow_array();
  $sth->finish;
592 593 594
  $done=0 unless($done);
  $total=0 unless($total);
  my $completed=0.0;
595 596
  $completed = ((100.0 * $done)/$total)  if($total>0);
  printf("hive %1.3f%% complete (%d done / %d total)\n", $completed, $done, $total);
597 598 599
  return $done, $total;
}

600
sub print_hive_status
601 602 603 604 605 606 607 608
{
  my $self = shift;
  $self->print_analysis_status;
  $self->print_running_worker_status;
}


sub print_analysis_status
609 610 611 612
{
  my $self = shift;

  my $allStats = $self->db->get_AnalysisStatsAdaptor->fetch_all();
613
 
614 615 616
  foreach my $analysis_stats (@{$allStats}) {
    $analysis_stats->print_stats;
  }
617 618 619 620 621 622
}


sub print_running_worker_status
{
  my $self = shift;
623

624
  my $total = 0;
625 626 627 628 629 630 631 632
  print("HIVE LIVE WORKERS====\n");
  my $sql = "select logic_name, count(*) from hive, analysis ".
            "where hive.analysis_id=analysis.analysis_id and hive.cause_of_death='' ".
            "group by hive.analysis_id";
  my $sth = $self->prepare($sql);
  $sth->execute();
  while((my $logic_name, my $count)=$sth->fetchrow_array()) {
    printf("%20s : %d workers\n", $logic_name, $count);
633
    $total += $count;
634
  }
635
  printf("  %d total workers\n", $total);
636
  print("=====================\n");
637
  $sth->finish;
638
}
639

640

Jessica Severin's avatar
Jessica Severin committed
641 642 643 644 645
#
# INTERNAL METHODS
#
###################

646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662
sub _pick_best_analysis_for_new_worker {
  my $self = shift;

  my $statsDBA = $self->db->get_AnalysisStatsAdaptor;
  return undef unless($statsDBA);

  my ($stats) = @{$statsDBA->fetch_by_needed_workers(1)};
  return $stats if($stats);

  # ok so no analyses 'need' workers.
  # see if anything needs an update, in case there are
  # hidden jobs that haven't made it into the summary stats

  print("QUEEN: no obvious needed workers, need to dig deeper\n");
  my $stats_list = $statsDBA->fetch_by_status('LOADING', 'BLOCKED');
  foreach $stats (@$stats_list) {
    #$stats->print_stats();
663
    $self->safe_synchronize_AnalysisStats($stats);
664 665 666 667 668 669 670 671 672 673 674 675 676
    $self->check_blocking_control_rules_for_AnalysisStats($stats);   
    #$stats->print_stats();

    return $stats if(($stats->status eq 'READY') and ($stats->num_required_workers > 0));
  }

  ($stats) = @{$statsDBA->fetch_by_needed_workers(1)};
  return $stats if($stats);

  return undef;
}


Jessica Severin's avatar
Jessica Severin committed
677 678 679 680
=head2 _fetch_by_hive_id

  Arg [1]    : int $id
               the unique database identifier for the feature to be obtained
681
  Example    : $feat = $queen->fetch_by_dbID(1234);
Jessica Severin's avatar
Jessica Severin committed
682 683 684 685 686 687 688 689
  Description: Returns the feature created from the database defined by the
               the id $id.
  Returntype : Bio::EnsEMBL::Hive::Worker
  Exceptions : thrown if $id is not defined
  Caller     : general

=cut

690
sub _fetch_by_hive_id {
Jessica Severin's avatar
Jessica Severin committed
691 692 693
  my ($self,$id) = @_;

  unless(defined $id) {
694
    throw("fetch_by_dbID must have an id");
Jessica Severin's avatar
Jessica Severin committed
695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784
  }

  my @tabs = $self->_tables;

  my $constraint = "h.hive_id = $id";

  #return first element of _generic_fetch list
  my ($obj) = @{$self->_generic_fetch($constraint)};
  return $obj;
}

=head2 _generic_fetch

  Arg [1]    : (optional) string $constraint
               An SQL query constraint (i.e. part of the WHERE clause)
  Arg [2]    : (optional) string $logic_name
               the logic_name of the analysis of the features to obtain
  Example    : $fts = $a->_generic_fetch('contig_id in (1234, 1235)', 'Swall');
  Description: Performs a database fetch and returns feature objects in
               contig coordinates.
  Returntype : listref of Bio::EnsEMBL::SeqFeature in contig coordinates
  Exceptions : none
  Caller     : BaseFeatureAdaptor, ProxyDnaAlignFeatureAdaptor::_generic_fetch

=cut

sub _generic_fetch {
  my ($self, $constraint, $join) = @_;

  my @tables = $self->_tables;
  my $columns = join(', ', $self->_columns());

  if ($join) {
    foreach my $single_join (@{$join}) {
      my ($tablename, $condition, $extra_columns) = @{$single_join};
      if ($tablename && $condition) {
        push @tables, $tablename;

        if($constraint) {
          $constraint .= " AND $condition";
        } else {
          $constraint = " $condition";
        }
      }
      if ($extra_columns) {
        $columns .= ", " . join(', ', @{$extra_columns});
      }
    }
  }

  #construct a nice table string like 'table1 t1, table2 t2'
  my $tablenames = join(', ', map({ join(' ', @$_) } @tables));

  my $sql = "SELECT $columns FROM $tablenames";

  my $default_where = $self->_default_where_clause;
  my $final_clause = $self->_final_clause;

  #append a where clause if it was defined
  if($constraint) {
    $sql .= " WHERE $constraint ";
    if($default_where) {
      $sql .= " AND $default_where ";
    }
  } elsif($default_where) {
    $sql .= " WHERE $default_where ";
  }

  #append additional clauses which may have been defined
  $sql .= " $final_clause";

  my $sth = $self->prepare($sql);
  $sth->execute;

#  print STDERR $sql,"\n";

  return $self->_objs_from_sth($sth);
}

sub _tables {
  my $self = shift;

  return (['hive', 'h']);
}

sub _columns {
  my $self = shift;

  return qw (h.hive_id
             h.analysis_id
785
             h.beekeeper
Jessica Severin's avatar
Jessica Severin committed
786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808
             h.host
             h.process_id
             h.work_done
             h.born
             h.last_check_in
             h.died
             h.cause_of_death
            );
}

sub _objs_from_sth {
  my ($self, $sth) = @_;

  my %column;
  $sth->bind_columns( \( @column{ @{$sth->{NAME_lc} } } ));

  my @workers = ();

  while ($sth->fetch()) {
    my $worker = new Bio::EnsEMBL::Hive::Worker;
    $worker->init;

    $worker->hive_id($column{'hive_id'});
809
    $worker->beekeeper($column{'beekeeper'});
Jessica Severin's avatar
Jessica Severin committed
810 811 812 813 814 815 816
    $worker->host($column{'host'});
    $worker->process_id($column{'process_id'});
    $worker->work_done($column{'work_done'});
    $worker->born($column{'born'});
    $worker->last_check_in($column{'last_check_in'});
    $worker->died($column{'died'});
    $worker->cause_of_death($column{'cause_of_death'});
817
    $worker->queen($self);
Jessica Severin's avatar
Jessica Severin committed
818 819
    $worker->db($self->db);

820 821
    if($column{'analysis_id'} and $self->db->get_AnalysisAdaptor) {
      $worker->analysis($self->db->get_AnalysisAdaptor->fetch_by_dbID($column{'analysis_id'}));
Jessica Severin's avatar
Jessica Severin committed
822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842
    }

    push @workers, $worker;
  }
  $sth->finish;

  return \@workers
}

sub _default_where_clause {
  my $self = shift;
  return '';
}

sub _final_clause {
  my $self = shift;
  return '';
}


1;