This file is indexed.

/usr/bin/pegasus-analyzer is in pegasus-wms 4.0.1+dfsg-8.

This file is owned by root:root, with mode 0o755.

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
 394
 395
 396
 397
 398
 399
 400
 401
 402
 403
 404
 405
 406
 407
 408
 409
 410
 411
 412
 413
 414
 415
 416
 417
 418
 419
 420
 421
 422
 423
 424
 425
 426
 427
 428
 429
 430
 431
 432
 433
 434
 435
 436
 437
 438
 439
 440
 441
 442
 443
 444
 445
 446
 447
 448
 449
 450
 451
 452
 453
 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
 480
 481
 482
 483
 484
 485
 486
 487
 488
 489
 490
 491
 492
 493
 494
 495
 496
 497
 498
 499
 500
 501
 502
 503
 504
 505
 506
 507
 508
 509
 510
 511
 512
 513
 514
 515
 516
 517
 518
 519
 520
 521
 522
 523
 524
 525
 526
 527
 528
 529
 530
 531
 532
 533
 534
 535
 536
 537
 538
 539
 540
 541
 542
 543
 544
 545
 546
 547
 548
 549
 550
 551
 552
 553
 554
 555
 556
 557
 558
 559
 560
 561
 562
 563
 564
 565
 566
 567
 568
 569
 570
 571
 572
 573
 574
 575
 576
 577
 578
 579
 580
 581
 582
 583
 584
 585
 586
 587
 588
 589
 590
 591
 592
 593
 594
 595
 596
 597
 598
 599
 600
 601
 602
 603
 604
 605
 606
 607
 608
 609
 610
 611
 612
 613
 614
 615
 616
 617
 618
 619
 620
 621
 622
 623
 624
 625
 626
 627
 628
 629
 630
 631
 632
 633
 634
 635
 636
 637
 638
 639
 640
 641
 642
 643
 644
 645
 646
 647
 648
 649
 650
 651
 652
 653
 654
 655
 656
 657
 658
 659
 660
 661
 662
 663
 664
 665
 666
 667
 668
 669
 670
 671
 672
 673
 674
 675
 676
 677
 678
 679
 680
 681
 682
 683
 684
 685
 686
 687
 688
 689
 690
 691
 692
 693
 694
 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
 785
 786
 787
 788
 789
 790
 791
 792
 793
 794
 795
 796
 797
 798
 799
 800
 801
 802
 803
 804
 805
 806
 807
 808
 809
 810
 811
 812
 813
 814
 815
 816
 817
 818
 819
 820
 821
 822
 823
 824
 825
 826
 827
 828
 829
 830
 831
 832
 833
 834
 835
 836
 837
 838
 839
 840
 841
 842
 843
 844
 845
 846
 847
 848
 849
 850
 851
 852
 853
 854
 855
 856
 857
 858
 859
 860
 861
 862
 863
 864
 865
 866
 867
 868
 869
 870
 871
 872
 873
 874
 875
 876
 877
 878
 879
 880
 881
 882
 883
 884
 885
 886
 887
 888
 889
 890
 891
 892
 893
 894
 895
 896
 897
 898
 899
 900
 901
 902
 903
 904
 905
 906
 907
 908
 909
 910
 911
 912
 913
 914
 915
 916
 917
 918
 919
 920
 921
 922
 923
 924
 925
 926
 927
 928
 929
 930
 931
 932
 933
 934
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
#!/usr/bin/env python

"""
Pegasus utility for pasing jobstate.log and reporting succesful and failed jobs

Usage: pegasus-analyzer [options]

"""

##
#  Copyright 2007-2012 University Of Southern California
#
#  Licensed under the Apache License, Version 2.0 (the "License");
#  you may not use this file except in compliance with the License.
#  You may obtain a copy of the License at
#
#  http://www.apache.org/licenses/LICENSE-2.0
#
#  Unless required by applicable law or agreed to in writing,
#  software distributed under the License is distributed on an "AS IS" BASIS,
#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#  See the License for the specific language governing permissions and
#  limitations under the License.
##

# Revision : $Revision: 2012 $

# Import Python modules
import os
import re
import sys
import time
import errno
import logging
import commands
import optparse
import traceback
import subprocess
import tempfile

# Initialize logging object
logger = logging.getLogger()
# Set default level to WARNING
logger.setLevel(logging.WARNING)

# Use pegasus-config to find our lib path
bin_dir = os.path.normpath(os.path.join(os.path.dirname(sys.argv[0])))
pegasus_config = os.path.join(bin_dir, "pegasus-config") + " --noeoln --python"
lib_dir = subprocess.Popen(pegasus_config, stdout=subprocess.PIPE, shell=True).communicate()[0]
pegasus_config = os.path.join(bin_dir, "pegasus-config") + " --noeoln --python-externals"
lib_ext_dir = subprocess.Popen(pegasus_config, stdout=subprocess.PIPE, shell=True).communicate()[0]

# Insert this directory in our search path
os.sys.path.insert(0, lib_ext_dir)
os.sys.path.insert(0, lib_dir)

# Import our modules
import Pegasus.common
from Pegasus.tools import utils
from Pegasus.tools import db_utils
from Pegasus.tools import kickstart_parser
from netlogger.analysis.workflow import stampede_statistics
from netlogger.analysis.schema.schema_check import SchemaVersionError

# --- regular expressions -------------------------------------------------------------

re_parse_property = re.compile(r'([^:= \t]+)\s*[:=]?\s*(.*)')
re_parse_script_pre = re.compile(r'^SCRIPT PRE (\S+) (.*)')
re_parse_condor_subs = re.compile(r'(\S+)="([^"]+)"')
re_collapse_condor_subs = re.compile(r'\$\([^\)]*\)') 

# --- classes -------------------------------------------------------------------------

class Job:

    def __init__(self, job_name, job_state=""):
	"""
	Initializes the Job class, setting job name,
	and state, if provided
	"""
	self.name = job_name		# Job name
	self.state = job_state		# Job state
        self.sub_file = ""	    	# Submit file for this job
        self.out_file = ""		# Output file for this job
        self.err_file = ""		# Error file for this job
        self.cluster = ""		# Cluster id for this job (from Condor)
        self.process = ""		# Process id for this job (from Condor)
        self.sub_file_parsed = False 	# Flag to tell if we were able to parse this job's submit file
        self.site = ""			# Site where the job ran
        self.executable = ""		# Job's executable
        self.arguments = ""		# Job's arguments
        self.initial_dir = ""		# Job's initial dir (from submit file)
        self.transfer_input_files = ""	# Files to transfer when debugging a job
        self.retries = None             # Keep track of how many times a job is submitted
        self.is_subdax = False          # Flag to tell if job is a SUBDAX/pegasus-plan job
        self.is_subdag = False		# Flag to tell if job is a SUBDAG job in the dag file
        self.subdag_dir = ""		# Subdag directory from a SUBDAG job in the dag file
        self.dag_path = ""		# Full path to the dag file from a SUBDAG job
        self.dagman_out = ""		# dagman.out file for this job (only for clustered jobs)
        self.pre_script = ""		# SCRIPT PRE line from the dag file
        self.condor_subs = {}		# Lits of condor substitutions rom DAG VARS line

    def set_state(self, new_state):
	"""
	This function updates a job state
	"""
	self.state = new_state

# --- constants -----------------------------------------------------------------------

MAXLOGFILE = 1000				# For log file rotation, check files .000 to .999
debug_level = logging.WARNING			# For now

# --- global variables ----------------------------------------------------------------

prog_base = os.path.split(sys.argv[0])[1]	# Name of this program
input_dir = None				# Directory given in -i command line option
dag_path = None					# Path of the dag file
tsdl_path = None				# Path to monitord's log file
temp_dir = None					# Temporary log file created
debug_job = None				# Path of a submit file to debug
debug_dir = None				# Temp directory to use while debugging a job
workflow_type = None				# Type of the workflow being debugged
workflow_base_dir = ""				# Workflow submit_dir or dirname(jsd) from braindump file
run_monitord = 0				# Run monitord before trying to analyze the output
output_dir = None                               # Output_dir for all files written by monitord
top_dir = None                                  # Top_dir of the main workflow, for obtaining the db location
use_files = False                               # Flag for using files in the workflow dir instead of the db
quiet_mode = 0					# Prints out/err filenames instead of dumping their contents
strict_mode = 0					# Gets out/err filenames from submit file
summary_mode = 0                                # Print just the summary output
debug_mode = 0					# Mode that enables debugging a single job
print_invocation = 0				# Prints invocation command for failed jobs
print_pre_script = 0				# Prints the SCRIPT PRE line for failed jobs, if present
jsdl_filename = "jobstate.log"			# Default name of the log file to use
jobs = {}					# List of jobs found in the jobstate.log file
total = 0					# Number of total jobs
success = 0					# Number of successful jobs
failed = 0					# Number of failed jobs
unsubmitted = 0					# Number of unsubmitted jobs
unknown = 0					# Number of jobs in an unknown state
failed_jobs = []				# List of jobs that failed
unknown_jobs = []				# List of jobs that neither succeeded nor failed

# --- functions -----------------------------------------------------------------------

def get_jsdl_filename(input_dir):
    """
    This function parses the braindump file in the input_dir,
    retrieving the wf_uuid and assembling the filename for the
    jobstate.log file.
    """
    try:
        my_wf_params = utils.slurp_braindb(input_dir)
    except:
        logger.error("cannot read braindump.txt file... exiting...")
        sys.exit(1)

    if "wf_uuid" in my_wf_params:
        return my_wf_params["wf_uuid"]+"-"+jsdl_filename

    logger.error("braindump.txt does not contain wf_uuid... exiting...")
    sys.exit(1)

def create_temp_logfile(name):
    """
    This function uses tempfile.mkstemp to create a temporary 
    log filename in the /tmp directory
    """
    try:
	tmp_file = tempfile.mkstemp(prefix="%s-" % (name), suffix=".log", dir="/tmp")
    except:
	return None

    # Close file, we will use it later
    os.close(tmp_file[0])

    # Return filename
    return tmp_file[1]

def has_seen(job_name):
    """
    This function returns true if we are already tracking job_name
    """
    if job_name in jobs:
	return True
    return False

def add_job(job_name, job_state=""):
    """
    This function adds a job to our list
    """
    # Don't add the same job twice
    if job_name in jobs:
	return

    newjob = Job(job_name, job_state)
    jobs[job_name] = newjob

def update_job_state(job_name, job_state=""):
    """
    This function updates the job state of a given job
    """
    # Make sure we have this job
    if not job_name in jobs:
	# Print a warning message
	logger.error("could not find job %s" % (job_name))
	return

    jobs[job_name].set_state(job_state)

def update_job_condor_info(job_name, condor_id="-"):
    """
    This function updates a job's condor_id (it splits it into process
    and cluster)
    """
    # Make sure we have this job
    if not job_name in jobs:
	# Print a warning message
	logger.error("could not find job %s" % (job_name))
	return

    # Nothing to do if condor_id is not defined
    if condor_id == "-":
	return

    my_split = condor_id.split(".")

    # First part is cluster id
    jobs[job_name].cluster = my_split[0]

    # If we have two pieces, second piece is process
    if len(my_split) >= 2:
	jobs[job_name].process = my_split[1]

def analyze():
    """
    This function processes all currently known jobs, generating some statistics
    """
    global total, success, failed, unsubmitted, unknown

    for my_job in jobs:
        total = total + 1
	if (jobs[my_job].state == "POST_SCRIPT_SUCCESS" or
	    jobs[my_job].state == "JOB_SUCCESS"):
	    success = success + 1
	elif (jobs[my_job].state == "POST_SCRIPT_FAILURE" or
	      jobs[my_job].state == "JOB_FAILURE"):
	    failed_jobs.append(my_job)
            failed = failed + 1
	elif (jobs[my_job].state == "UNSUBMITTED"):
	    unsubmitted = unsubmitted + 1
	else:
	    # It seems we don't have a final result for this job
	    unknown_jobs.append(my_job)
            unknown = unknown + 1

def parse_submit_file(my_job):
    """
    This function opens a submit file and reads site
    and condor dagman log information
    """
    # First we check if this is a SUBDAG job from the dag file
    if my_job.is_subdag:
	# Nothing to do here
	return

    if my_job.sub_file == "":
	# Create full path for the submit file if we already don't have the sub file set up
	my_job.sub_file = os.path.join(input_dir, my_job.name + ".sub")
    my_job.out_file = os.path.join(input_dir, my_job.name + ".out")
    my_job.err_file = os.path.join(input_dir, my_job.name + ".err")

    # Try to access submit file
    if os.access(my_job.sub_file, os.R_OK):
	# Open submit file
	try:
	    SUB = open(my_job.sub_file, "r")
	except:
	    # print "error opening submit file: %s" % (my_job.sub_file)
	    # fail silently for now...
	    return
	
	# submit file found
	my_job.sub_file_parsed = True

	# Check if this job includes sub workflows
	if my_job.is_subdax:
	    has_sub_workflow = True
	else:
	    has_sub_workflow = False

	# Parse submit file
	for line in SUB:
	    # First we need to do some trimming...
	    line = line.strip(" \t") # Remove leading and trailing spaces
	    if line.startswith('#'):
		# Skip comments
		continue
	    line = line.rstrip("\n\r") # Remove new lines, if any
	    line = line.split('#')[0] # Remove inline comments too
	    line = line.strip() # Remove any remaining spaces at both ends
	    if len(line) == 0:
		# Skip empty lines
		continue
	    prop = re_parse_property.search(line)
	    if prop:
		# Parse successful
		k = prop.group(1)
		v = prop.group(2)

		# See if it is one of the properties we are looking for...
		if k == "+pegasus_site":
		    my_job.site = v.strip('"')
		    continue
		if k == "arguments":
		    my_job.arguments = v.strip('"')
		if k == "executable":
		    my_job.executable = v
		if k == "environment" and has_sub_workflow:
		    # Ok, we need to find the CONDOR_DAGMAN_LOG entry now...
		    sub_props = v.split(';')
		    for sub_prop_line in sub_props:
			sub_prop_line = sub_prop_line.strip() # Remove any spaces
			if len(sub_prop_line) == 0:
			    continue
			sub_prop = re_parse_property.search(sub_prop_line)
			if sub_prop:
			    if sub_prop.group(1) == "_CONDOR_DAGMAN_LOG":
				my_job.dagman_out = sub_prop.group(2)
                                my_job.dagman_out = os.path.normpath(my_job.dagman_out)
                                if my_job.dagman_out.find(workflow_base_dir) >= 0:
                                    # Path to dagman_out file includes original submit_dir, let's try to change it
                                    my_job.dagman_out = os.path.normpath(my_job.dagman_out.replace((workflow_base_dir + os.sep), '', 1))
                                    # Join with current input_dir
                                    my_job.dagman_out = os.path.join(input_dir, my_job.dagman_out)

                                    # Now, figure out the correct directory, accounting for
                                    # replanning and rescue modes
                                    
                                    # Split filename into dir and base names
                                    my_dagman_dir = os.path.dirname(my_job.dagman_out)
                                    my_dagman_file = os.path.basename(my_job.dagman_out)
                                    my_retry = my_job.retries
                                    if my_retry is None:
                                        logger.warning("sub-workflow retry counter not initialized... continuing...")
                                        continue

                                    # Compose directory... assuming replanning mode
                                    my_retry_dir = my_dagman_dir + ".%03d" % (my_retry)
                                    # If directory doesn't exist, let's change to rescue mode
                                    if not os.path.isdir(my_retry_dir):
                                        logger.debug("sub-workflow directory %s does not exist, shifting to rescue mode..." % (my_retry_dir))
                                        my_retry_dir = my_dagman_dir + ".000"
                                        if not os.path.isdir(my_retry_dir):
                                            # Still not able to find it, output warning message
                                            logger.warning("sub-workflow directory %s does not exist!"
                                                           % (my_retry_dir))
                                            continue

                                    # Found sub-workflow directory, let's compose the final path to the new dagman.out file...
                                    my_job.dagman_out = os.path.join(my_retry_dir, my_dagman_file)



		# Only parse following keys if we are running in strict mode
		if strict_mode:
		    # Get initial dir
		    if k == "initialdir":
			my_job.initial_dir = v
		    # Parse error and output keys
		    if k == "output" or k == "error":
			# Take care of basic substitutions first
			v = v.replace("$(cluster)", my_job.cluster)
			v = v.replace("$(process)", my_job.process)

			# Now we do any substitutions from the DAG's VAR line (if any)
			for my_key in my_job.condor_subs:
			    v = v.replace("$(%s)" % (my_key), my_job.condor_subs[my_key])

			# Now, we collapse any remaining substitutions (not found in the VAR line)
			v = re_collapse_condor_subs.sub('', v)

			# Make sure we have an absolute path
			if not os.path.isabs(v):
			    v = os.path.join(input_dir, v)

			# Done! Replace out/err filenames with what we have
			if k == "output":
			    my_job.out_file = v
			else:
			    my_job.err_file = v
		# Only parse following keys if we are debugging a job
		if debug_mode:
		    # Get transfer input files
		    if k == "transfer_input_files":
			my_job.transfer_input_files = v
	SUB.close()
	# If initialdir was specified, we need to make both error and output files relative to that
	if len(my_job.initial_dir):
	    my_job.out_file = os.path.join(my_job.initial_dir, my_job.out_file)
	    my_job.err_file = os.path.join(my_job.initial_dir, my_job.err_file)
    else:
	# Was not able to access submit file
	# fail silently for now...
	# print "cannot access submit file: %s" % (my_job.sub_file)
	pass

def find_file(input_dir, file_type):
    """
    This function finds a file with the suffix file_type
    in the input directory. We assume there is just one
    file of the requested type in the directory (otherwise
    the function will return the first file matching the type
    """
    try:
	file_list = os.listdir(input_dir)
    except:
	logger.error("cannot read directory: %s" % (input_dir))
	sys.exit(1)

    for file in file_list:
	if file.endswith(file_type):
	    return os.path.join(input_dir, file)

    logger.error("could not find any %s file in %s" % (file_type, input_dir))
    sys.exit(1)

def parse_dag_file(dag_fn):
    """
    This function walks through the dag file, learning about
    all jobs before hand.
    """
    # Open dag file
    try:
	DAG = open(dag_fn, "r")
    except:
	logger.error("could not open dag file %s: exiting..." % (dag_fn))
	sys.exit(1)

    # Loop through the dag file
    for line in DAG:
	line = line.strip(" \t")
	if line.startswith("#"):
	    # Skip comments
	    continue
	line = line.rstrip("\n\r") # Remove new lines, if any
	line = line.split('#')[0] # Remove inline comments too
	line = line.strip() # Remove any remaining spaces at both ends
	if len(line) == 0:
	    # Skip empty lines
	    continue
	if line.startswith("JOB"):
	    # This is a job line, let's parse it
	    my_job = line.split()
	    if len(my_job) != 3:
		logger.warn("confused parsing dag line: %s" % (line))
		continue
	    if not has_seen(my_job[1]):
		add_job(my_job[1], "UNSUBMITTED")
		# Get submit file information from dag file
		jobs[my_job[1]].sub_file = os.path.join(input_dir, my_job[2])
                if my_job[1].startswith("pegasus-plan") or my_job[1].startswith("subdax_"):
                    # Mark job as subdax
                    jobs[my_job[1]].is_subdax = True
	    else:
		logger.warn("job appears twice in dag file: %s" % (my_job[1]))
	if line.startswith("SUBDAG EXTERNAL"):
	    # This is a subdag line, parse it to get job name and directory
	    my_job = line.split()
	    if len(my_job) != 6:
		logger.warn("confused parsing dag line: %s" % (line))
		continue
	    if not has_seen(my_job[2]):
		add_job(my_job[2], "UNSUBMITTED")
		jobs[my_job[2]].is_subdag = True
		jobs[my_job[2]].dag_path = my_job[3]
		jobs[my_job[2]].subdag_dir = my_job[5]
	    else:
		logger.warn("job appears twice in dag file: %s" % (my_job[2]))
	if line.startswith("SCRIPT PRE"):
	    # This is a SCRIPT PRE line, parse it to get the script for the job
	    my_script = re_parse_script_pre.search(line)
	    if my_script is None:
		# Couldn't parse line
		logger.warn("confused parsing dag line: %s" % (line))
		continue
	    # Get job name, and check if we have it
	    my_job = my_script.group(1)
	    if not has_seen(my_job):
		# Cannot find this job, ignore this line
		logger.warn("couldn't find job: %s for PRE SCRIPT line in dag file" % (my_job))
		continue
	    # Good, copy PRE script line to our job structure
	    jobs[my_job].pre_script = my_script.group(2)
	if line.startswith("VARS"):
	    # This is a VARS line, parse it to get the condor substitutions
	    if len(line.split()) > 2:
		# Line looks promising...
		my_job = line.split()[1]
		if not has_seen(my_job):
		    # Cannot find this job, ignore this line
		    logger.warn("couldn't find job: %s for VARS line in dag file" % (my_job))
		    continue
		# Good, parse the condor substitutions, and create substitution dictionary
		for my_key, my_val in re_parse_condor_subs.findall(line):
		    jobs[my_job].condor_subs[my_key] = my_val

def parse_jobstate_log(jobstate_fn):
    """
    This function parses the jobstate.log file, loading all job information
    """
    # Open log file
    try:
	JSDL = open(jobstate_fn, "r")
    except:
	logger.error("could not open file %s: exiting..." % (jobstate_fn))
	sys.exit(1)

    # Loop through the log file
    for line in JSDL:
	sp = line.split()
	# Skip lines that don't have enough items
	if len(sp) < 6:
	    continue
	# Skip monitord comments
	if sp[1] == "INTERNAL":
	    continue

	# Ok, we have a valid job
	jobname = sp[1]
	jobstate = sp[2]
	condor_id = sp[3]

	# Add to job list if we have never seen this job before
	if not has_seen(jobname):
	    logger.warn("job %s not present in dag file" % (jobname))
	    add_job(jobname, jobstate)
            if jobname.startswith("pegasus-plan") or jobname.startswith("subdax_"):
                # Mark job as subdax
                jobs[jobname].is_subdax = True
	else:
	    # Update job state
	    update_job_state(jobname, jobstate)

	# Update condor id if we reached the SUBMIT state
	if jobstate == "SUBMIT":
	    update_job_condor_info(jobname, condor_id)
            # Keep track of retries
            if jobs[jobname].retries is None:
                jobs[jobname].retries = 0
            else:
                jobs[jobname].retries = jobs[jobname].retries + 1

    # Close log file
    JSDL.close()

def find_latest_log(log_file_base):
    """
    This function tries to locate the latest log file
    """
    last_log = None
    curr_log = None

    if os.access(log_file_base, os.F_OK):
	last_log = log_file_base

    # Starts from .000
    sf = 0

    while (sf < MAXLOGFILE):
	curr_log = log_file_base + ".%03d" % (sf)
	if os.access(curr_log, os.F_OK):
	    last_log = curr_log
	    sf = sf + 1
	else:
	    break

    return last_log

def invoke_monitord(dagman_out_file, output_dir):
    """
    This function runs monitord on the given dagman_out_file.
    """
    monitord_cmd = "pegasus-monitord -r --no-events"
    if output_dir is not None:
        # Add output_dir, if given
        monitord_cmd = monitord_cmd + " --output-dir " + output_dir
    monitord_cmd = monitord_cmd + " " + dagman_out_file
    logger.info("running: %s" % (monitord_cmd))

    try:
	status, output = commands.getstatusoutput(monitord_cmd)
    except:
	logger.error("could not invoke monitord, exiting...")
	sys.exit(1)

def dump_file(file):
    """
    This function dumps a file to our stdout
    """
    if file is not None:
	try:
	    OUT = open(file, 'r')
	except:
	    logger.warn("*** Cannot access: %s" % (file))
	    print
	else:
	    print os.path.split(file)[1].center(80, '-')
	    print
	    # Dump file contents to terminal
	    line = OUT.readline()
	    while line:
		line = line.strip()
		print line
		line = OUT.readline()

	    OUT.close()
	    print

def print_output_error(job):
    """
    This function outputs both output and error files for a given job.
    """
    out_file = find_latest_log(job.out_file)
    err_file = find_latest_log(job.err_file)

    my_parser = kickstart_parser.Parser(out_file)
    my_output = my_parser.parse_stdout_stderr()
    my_task_id = 0

    if len(my_output) > 0:
        # Ok, we got valid kickstart records, output stdout and stderr for tasks that failed
        for entry in my_output:
            # Count tasks, the same way as pegasus-monitord for Stampede
            my_task_id = my_task_id + 1
            if not "derivation" in entry or not "transformation" in entry:
                continue
            if not "exitcode" in entry and not "error" in entry:
                continue
            if "exitcode" in entry:
                try:
                    if int(entry["exitcode"]) == 0:
                        # Skip tasks with exitcode equals to zero
                        continue
                except:
                    logger.warn("couldn't convert exitcode to integer!")
                    continue
            else:
                # We must have "error" in entry
                pass
            # Got a task with a non-zero exitcode
            print ("Task #" + str(my_task_id) + " - Summary").center(80, '-')
            print
            if "resource" in entry:
                print "site        : %s" % (entry["resource"])
            if "hostname" in entry:
                print "hostname    : %s" % (entry["hostname"])
            if "name" in entry:
                print "executable  : %s" % (entry["name"])
            if "argument-vector" in entry:
                print "arguments   : %s" % (entry["argument-vector"])
            if "exitcode" in entry:
                print "exitcode    : %s" % (entry["exitcode"])
            else:
                if "error" in entry:
                    print "error       : %s" % (entry["error"])
            if "cwd" in entry:
                print "working dir : %s" % (entry["cwd"])
            print
            # Now let's display stdout and stderr
            if "stdout" in entry:
                if len(entry["stdout"]) > 0:
                    # Something to display
                    print ("Task #" + str(my_task_id) + " - " + entry["transformation"] + " - " + entry["derivation"] + " - stdout").center(80, '-')
                    print
                    print entry["stdout"]
                    print
            if "stderr" in entry:
                if len(entry["stderr"]) > 0:
                    # Something to display
                    print ("Task #" + str(my_task_id) + " - " + entry["transformation"] + " - " + entry["derivation"] + " - stderr").center(80, '-')
                    print
                    print entry["stderr"]
                    print
    else:
        # Not able to parse the kickstart output file, let's just dump the out and err files

	# Print outfile to screen
        dump_file(out_file)

	# Print errfile to screen
        dump_file(err_file)

def print_job_info(job):
    """
    This function prints the information about a particular job
    """
    print
    print job.center(80, '=')
    print
    print " last state: %s" % (jobs[job].state)
    parse_submit_file(jobs[job])

    # Handle subdag jobs from the dag file
    if jobs[job].is_subdag == True:
	print " This is a SUBDAG job:"
	print " For more information, please run the following command:"
	user_cmd = " %s -s " % (prog_base)
        if output_dir is not None:
            user_cmd = user_cmd + " --output-dir %s" % (output_dir)
	print "%s -f %s" % (user_cmd, jobs[job].dag_path)
	print
	return

    if jobs[job].sub_file_parsed == False:
	print "       site: submit file not available"
    else:
	print "       site: %s" % (jobs[job].site or '-')
    print "submit file: %s" % (jobs[job].sub_file)
    print "output file: %s" % (find_latest_log(jobs[job].out_file))
    print " error file: %s" % (find_latest_log(jobs[job].err_file))
    if print_invocation:
	print
	print "To re-run this job, use: %s %s" % (jobs[job].executable, jobs[job].arguments)
	print
    if print_pre_script and len(jobs[job].pre_script) > 0:
	print
	print "SCRIPT PRE:"
	print jobs[job].pre_script
	print
    if len(jobs[job].dagman_out) > 0:
	# This job has a sub workflow
	print " This job contains sub workflows!"
	print " Please run the command below for more information:"
	user_cmd = " %s" % (prog_base)
        if output_dir is not None:
            user_cmd = user_cmd + " --output-dir %s" % (output_dir)
	print "%s -d %s" % (user_cmd, os.path.split(jobs[job].dagman_out)[0])
	print
    print

    # Now dump file contents to screen if we are not in quiet mode
    if not quiet_mode:
        print_output_error(jobs[job])

def print_top_summary():
    """
    This function prints the summary for the analyzer report,
    which is the same for the long and short output versions
    """
    print
    print "Summary".center(80, '*')
    print
    print " Total jobs         : % 6d (%3.2f%%)" % (total, 100 * (1.0 * total/(total or 1)))
    print " # jobs succeeded   : % 6d (%3.2f%%)" % (success, 100 * (1.0 * success/(total or 1)))
    print " # jobs failed      : % 6d (%3.2f%%)" % (failed, 100 * (1.0 * failed/(total or 1)))
    print " # jobs unsubmitted : % 6d (%3.2f%%)" % (unsubmitted, 100 * (1.0 * unsubmitted/(total or 1)))
    if unknown > 0:
	print " # jobs unknown     : % 6d (%3.2f%%)" % (unknown, 100 * (1.0 * unknown/(total or 1)))
    print

def print_summary():
    """
    This function prints the analyzer report summary
    """

    # First print the summary section
    print_top_summary()

    # Print information about failed jobs
    if len(failed_jobs):
	print "Failed jobs' details".center(80, '*')
	for job in failed_jobs:
	    print_job_info(job)

    # Print information about unknown jobs
    if len(unknown_jobs):
	print "Unknown jobs' details".center(80, '*')
	for job in unknown_jobs:
	    print_job_info(job)

def analyze_files():
    """
    This function runs the analyzer using the files in the workflow
    directory as the data source.
    """
    jsdl_path = None                            # Path of the jobstate.log file
    run_directory_writable = False              # Flag to indicate if we can write to the run directory
    dagman_out_path = None			# Path to the dagman.out file

    global workflow_base_dir, dag_path

    # Get the dag file if it was not specified by the user
    if dag_path is None:
        dag_path = find_file(input_dir, ".dag")
        logger.info("using %s, use the --dag option to override" % (dag_path))

    # Build dagman.out path
    dagman_out_path = dag_path + ".dagman.out"

    # Check if we can write to the run directory
    run_directory_writable = os.access(input_dir, os.W_OK)

    # Invoke monitord if requested
    if run_monitord:
        if output_dir is not None:
            # If output_dir is specified, invoke monitord with that path
            invoke_monitord("%s.dagman.out" % (dag_path), output_dir)
            # jobstate.log file uses wf_uuid as prefix
            jsdl_path = os.path.join(output_dir, get_jsdl_filename(input_dir))
        else:
            if run_directory_writable:
                # Run directory is writable, write monitord output to jobstate.log file
                jsdl_path = os.path.join(input_dir, jsdl_filename)
                # Invoke monitord
                invoke_monitord("%s.dagman.out" % (dag_path), None)
            else:
                # User must provide the --output-dir option
                logger.error("%s is not writable" % (input_dir))
                logger.error("user must specify directory for new monitord logs with the --output-dir option")
                logger.error("exiting...")
                sys.exit(1)
    else:
        if output_dir is not None:
            # jobstate.log file uses wf_uuid as prefix and is inside output_dir
            jsdl_path = os.path.join(output_dir, get_jsdl_filename(input_dir))
        else:
            jsdl_path = os.path.join(input_dir, jsdl_filename)

    # Compare timestamps of jsdl_path with dagman_out_path
    try:
        jsdl_stat = os.stat(jsdl_path)
    except:
        logger.error("could not access %s, exiting..." % (jsdl_path))
        sys.exit(1)

    try:
        dagman_out_stat = os.stat(dagman_out_path)
    except:
        logger.error("could not access %s, exiting..." % (dagman_out_path))
        sys.exit(1)

    # Compare mtime for both files
    if dagman_out_stat[8] > jsdl_stat[8]:
        logger.warning("jobstate.log older than the dagman.out file, workflow logs may not be up to date...")

    # Try to parse workflow parameters from braindump.txt file
    wfparams = utils.slurp_braindb(input_dir)
    if "submit_dir" in wfparams:
        workflow_base_dir = os.path.normpath(wfparams["submit_dir"])
    elif "jsd" in wfparams:
        workflow_base_dir = os.path.dirname(os.path.normpath(wfparams["jsd"]))

    # First we learn about jobs by going through the dag file
    parse_dag_file(dag_path)

    # Read logfile
    parse_jobstate_log(jsdl_path)

    # Process our jobs
    analyze()

    # Print summary of our analysis
    if summary_mode :
        print_top_summary()
    else :
        # This is non summary mode despite of the name (go figure)    
        print_summary()

    if failed > 0:
        # Workflow has failures, exit with exitcode 2
        sys.exit(2)
    # Workflow has no failures, exit with exitcode 0
    sys.exit(0)

def analyze_db(config_properties):
    """
    This function runs the analyzer using data from the database.
    """
    global total, success, failed, unsubmitted, unknown

    # Get the database URL
    output_db_url, wf_uuid = db_utils.get_db_url_wf_uuid(input_dir, config_properties, top_dir)

    # Nothing to do if we cannot resolve the database URL
    if output_db_url is None:
        logger.error("cannot find database URL, exiting...")
        sys.exit(1)

    # Now, let's try to access the database
    try:
        workflow_stats = stampede_statistics.StampedeStatistics(output_db_url, False)
        workflow_stats.initialize(wf_uuid)
    except SchemaVersionError:
        logger.error("------------------------------------------------------")
        logger.error("Database schema mismatch! Please run the upgrade tool")
        logger.error("to upgrade the database to the latest schema version.")
        sys.exit(1)
    except:
        logger.error("Failed to load the database." + output_db_url )
        logger.warning(traceback.format_exc())
        sys.exit(1)

    total = workflow_stats.get_total_jobs_status()
    success = workflow_stats.get_total_succeeded_jobs_status()
    failed = workflow_stats.get_total_failed_jobs_status()
    unsubmitted = total - success - failed

    # Let's print the results
    
    print_top_summary()

    # Exit if summary mode is on
    if summary_mode :
         if failed > 0:
             # Workflow has failures, exit with exitcode 2
             sys.exit(2)
         # Workflow has no failures, exit with exitcode 0  
         sys.exit(0)

    # Now, print information about jobs that failed...
    if failed > 0:
        # Get list of failed jobs from database
        my_failed_jobs = workflow_stats.get_failed_job_instances(final=True, all_jobs=True)
        
        # Print header
	print "Failed jobs' details".center(80, '*')

        # Now process one by one...
        for my_job in my_failed_jobs:
            my_info = workflow_stats.get_job_instance_info(my_job[0])[0]
            my_tasks = workflow_stats.get_invocation_info(my_job[0])

            # Unquote stdout and stderr
            my_info.stdout_text = utils.unquote(my_info.stdout_text or "")
            my_info.stderr_text = utils.unquote(my_info.stderr_text or "")
            my_info.stdout_text = my_info.stdout_text.strip(" \n\r\t")
            my_info.stderr_text = my_info.stderr_text.strip(" \n\r\t")
            
            if my_job[0] == my_info.job_instance_id:

                print
                print my_info.job_name.center(80, '=')
                print
                print " last state: %s" % (my_info.state or '-')

                print "       site: %s" % (my_info.site or '-')
                print "submit file: %s" % (my_info.submit_file or '-')
                print "output file: %s" % (my_info.stdout_file or '-')
                print " error file: %s" % (my_info.stderr_file or '-')
                if print_invocation:
                    print
                    print "To re-run this job, use: %s %s" % ((my_info.executable or '-'), (my_info.argv or '-'))
                    print
                if print_pre_script and len((my_info.pre_executable or "")) > 0:
                    print
                    print "SCRIPT PRE:"
                    print "%s %s" % ((my_info.pre_executable or ""), (my_info.pre_argv or ""))
                    print
                if my_info.subwf_dir is not None:
                    # This job has a sub workflow
                    print " This job contains sub workflows!"
                    print " Please run the command below for more information:"
                    user_cmd = " %s" % (prog_base)
                    my_wfdir = os.path.normpath(my_info.subwf_dir)
                    if my_wfdir.find(my_info.submit_dir) >= 0:
                        # Path to dagman_out file includes original submit_dir, let's try to change it...
                        my_wfdir = os.path.normpath(my_wfdir.replace((my_info.submit_dir + os.sep), '', 1))
                        my_wfdir = os.path.join(input_dir, my_wfdir)
                    print "%s -d %s --top-dir %s" % (user_cmd, my_wfdir, (top_dir or input_dir))
                    print
                print

                # Now, print task information
                for my_task in my_tasks:
                    if my_task[0] < 1:
                        # Skip pre, and post script tasks
                        continue
                    if my_task[1] == 0:
                        # Skip tasks that succeeded
                        continue

                    # Got a task with a non-zero exitcode
                    my_exitcode = utils.raw_to_regular(my_task[1])

                    # Print task summary
                    print ("Task #" + str(my_task[0]) + " - Summary").center(80, '-')
                    print
                    print "site        : %s" % (my_info.site or '-')
                    print "hostname    : %s" % (my_info.hostname or '-')
                    print "executable  : %s" % (str(my_task[2] or '-'))
                    print "arguments   : %s" % (str(my_task[3] or '-'))
                    print "exitcode    : %s" % (str(my_exitcode))
                    print "working dir : %s" % (my_info.work_dir or '-')
                    print

                    if quiet_mode:
                        continue

                    # Now, print task stdout and stderr, if anything is there
                    my_stdout_str = "#@ %d stdout" % (my_task[0])
                    my_stderr_str = "#@ %d stderr" % (my_task[0])

                    # Start with stdout
                    my_stdout_start = my_info.stdout_text.find(my_stdout_str)
                    if my_stdout_start >= 0:
                        my_stdout_start = my_stdout_start + len(my_stdout_str) + 1
                        my_stdout_end = my_info.stdout_text.find("#@", my_stdout_start)
                        if my_stdout_end < 0:
                            # Next comment not found, possibly the last entry
                            my_stdout_end = len(my_info.stdout_text)
                        else:
                            my_stdout_end = my_stdout_end - 1

                        if my_stdout_end - my_stdout_start > 0:
                            # Something to display
                            print ("Task #" + str(my_task[0]) + " - " + str(my_task[4]) + " - " + str(my_task[5]) + " - stdout").center(80, '-')
                            print
                            print my_info.stdout_text[my_stdout_start:my_stdout_end]
                            print

                    # Now print stderr (from the kickstart output file)
                    my_stderr_start = my_info.stdout_text.find(my_stderr_str)
                    if my_stderr_start >= 0:
                        my_stderr_start = my_stderr_start + len(my_stderr_str) + 1
                        my_stderr_end = my_info.stdout_text.find("#@", my_stderr_start)
                        if my_stderr_end < 0:
                            # Next comment not found, possibly the last entry
                            my_stderr_end = len(my_info.stdout_text)
                        else:
                            my_stderr_end = my_stderr_end - 1

                        if my_stderr_end - my_stderr_start > 0:
                            # Something to display
                            print ("Task #" + str(my_task[0]) + " - " + str(my_task[4]) + " - " + str(my_task[5]) + " - Kickstart stderr").center(80, '-')
                            print
                            print my_info.stdout_text[my_stderr_start:my_stderr_end]
                            print

                    # Now print the stderr output from the .err file
                    if my_info.stderr_text.strip("\n\t \r") != "":
                        # Something to display
                        print ("Task #" + str(my_task[0]) + " - " + str(my_task[4]) + " - " + str(my_task[5]) + " - stderr").center(80, '-')
                        print
                        print my_info.stderr_text
                        print
            else:
                log.error("unexpected job instance returned by database!")
                log.error("returned: %d - expected: %d" % (my_info[0], my_job[0]))
                continue

    # Done with the database
    workflow_stats.close()

    if failed > 0:
        # Workflow has failures, exit with exitcode 2
        sys.exit(2)

    # Workflow has no failures, exit with exitcode 0
    sys.exit(0)

def debug_condor(my_job):
    """
    This function is used to debug a condor job. It creates a
    shell script in the debug_dir directory that is used to
    copy all necessary files to the (local) debug_dir directory
    and then execute the job locally.
    """
    global strict_mode

    # Set strict mode in order to parse everything in the submit file
    strict_mode = 1
    # Parse submit file
    parse_submit_file(my_job)

    # Create script name
    debug_script_basename = "debug_" + my_job.name + ".sh"
    debug_script_name = os.path.join(debug_dir, debug_script_basename)

    job_executable = os.path.join(debug_dir, my_job.executable) + my_job.arguments

    try:
	debug_script = open(debug_script_name, "w")
    except:
	logger.error("cannot create debug script %s" % (debug_script))
	sys.exit(1)

    try:
	# Start with the bash line
	debug_script.write("#!/bin/bash\n")
	debug_script.write("\n")
	debug_script.write("set -e\n")
	debug_script.write("\n")
	debug_script.write("# Copy any files that are needed\n")

	debug_script.write('echo "copying input files..."\n')
	debug_script.write("\n")
	# Copy all files that we need
	for my_file in my_job.transfer_input_files.split(","):
	    if len(my_file):
		if len(my_job.initial_dir):
		    # Add the initial dir to all files to be copied
		    my_file = os.path.join(my_job.initial_dir, my_file)
		debug_script.write("cp %s %s\n" % (my_file, debug_dir))

	# Extra newline before executing the job
	debug_script.write("\n")
	debug_script.write('echo "copying input files completed."\n')
	debug_script.write("\n")
	debug_script.write("# Set the execute bit on the executable\n")
	debug_script.write("chmod +x %s\n" % (os.path.join(debug_dir, my_job.executable)))
	debug_script.write("\n")
	debug_script.write('echo "executing job: %s"\n' % (job_executable))
	debug_script.write("\n")
	debug_script.write("# Now, execute the job\n")
	debug_script.write("%s\n" % (job_executable))
	debug_script.write("\n")
        # Remember not to put anything between running the executable
        # and checking the exit code, otherwise $? will break...
        debug_script.write("# Check error code\n")
        debug_script.write("if [ $? -eq 0 ]; then\n")
        debug_script.write('   echo "executable ran successfully"\n')
        debug_script.write("else\n")
        debug_script.write('   echo "executable failed with error $?"\n')
    except:
	logger.error("cannot write to file %s" % (debug_script))
	sys.exit(1)

    # We are done writing the file!
    debug_script.close()

    try:
	# Make our debug script executable
	os.chmod(debug_script_name, 0755)
    except:
	logger.error("cannot change permissions for the debug script %s" % (debug_script))
	sys.exit(1)

    # Print next step
    print
    print "%s: finished generating job debug script!" % (prog_base)
    print
    print "To run it, you need to type:"
    print "   $ cd %s" % (debug_dir)
    print "   $ ./%s" % (debug_script_basename)
    print

def debug_workflow():
    """
    This function handles the mode where the analyzer
    is used to debug a job in a workflow
    """
    global debug_job, debug_dir

    # Check if we can find this job's submit file
    if not debug_job.endswith(".sub"):
	debug_job = debug_job + ".sub"
    # Figure out job name
    jobname = os.path.basename(debug_job)
    jobname = jobname[0:jobname.find(".sub")]
    # Create job class
    my_job = Job(jobname)
    my_job.sub_file = debug_job

    if not os.access(debug_job, os.R_OK):
	logger.error("cannot access job submit file: %s" % (debug_job))
	sys.exit(1)

    # Handle the temporary directory option
    if debug_dir is None:
	# Create temporary directory
	try:
	    debug_dir = tempfile.mkdtemp()
	except:
	    logger.error("could not create temporary directory!")
	    sys.exit(1)
    else:
	# Make sure directory specified is writable
	debug_dir = os.path.abspath(debug_dir)
	if not os.access(debug_dir, os.F_OK):
	    # Create directory if it does not exist
	    try:
		os.mkdir(debug_dir)
	    except:
		logger.error("cannot create debug directory: %s" % (debug_dir))

	# Check if we can write to the debug directory
	if not os.access(debug_dir, os.W_OK):
	    logger.error("not able to write to temporary directory: %s" % (debug_dir))
	    sys.exit(1)


    # Handle workflow type
    if workflow_type is not None:
	if workflow_type.lower() == "condor":
	    logger.info("debugging condor type workflow")
	    debug_condor(my_job)
	else:
	    logger.error("workflow type %s not supported!" % (workflow_type))
	    sys.exit(1)
    else:
	logger.info("debugging condor type workflow")
	debug_condor(my_job)

    # All done, in case we are back here!
    sys.exit(0)

# --- main ----------------------------------------------------------------------------

# Configure command line option parser
prog_usage = "usage: %s [options] workflow_directory" % (prog_base)
parser = optparse.OptionParser(usage=prog_usage)
parser.add_option("-v", "--verbose", action="count", default=0, dest="vb",
                  help="Increase verbosity, repeatable")
parser.add_option("-i", "-d", "--dir", action = "store", type = "string", dest = "input_dir",
		  help = "input directory where the jobstate.log file is located, default is the current directory")
parser.add_option("--dag", action = "store", type = "string", dest = "dag_filename",
		  help = "full path to the dag file to use -- this option overrides the -d option")
parser.add_option("-f", "--files", action = "store_const", const = 1, dest = "use_files",
		  help = "disables the database mode and forces the use of workflow directory files")
parser.add_option("-m", "-t", "--monitord", action = "store_const", const = 1, dest = "run_monitord",
		  help = "run pegasus-monitord before analyzing the output")
parser.add_option("-o", "--output-dir", action = "store", type = "string", dest = "output_dir",
                  help = "provides an output directory for all monitord log files")
parser.add_option("--top-dir", action = "store", type = "string", dest = "top_dir",
                  help = "provides the location of the top-level workflow directory, needed to analyze sub-workflows")
parser.add_option("-c","--conf", action = "store", type = "string", dest = "config_properties",
                  help = "Specifies the properties file to use. This overrides all other property files.")
parser.add_option("-q", "--quiet", action = "store_const", const = 1, dest = "quiet_mode",
		  help = "output out/err filenames instead of their contents")
parser.add_option("-p", "--print", action = "store", type = "string" , dest = "print_options",
		  help = "specifies print options from pre,invocation")
parser.add_option("-s", "--strict", action = "store_const", const = 1, dest = "strict_mode",
		  help = "gets a job's out and err files from the submit file")
parser.add_option("-S", "--summary", action = "store_const", const = 1, dest = "summary_mode",
		  help = "Just print the summary and exit")
parser.add_option("--debug-job", action = "store", type = "string", dest = "debug_job",
		  help = "specifies a job to debug (can be either the job base name or the submit file name) -- this option enables debugging a single job")
parser.add_option("--debug-dir", action = "store", type = "string", dest = "debug_dir",
		  help = "specifies the directory to use as debug directory (default is to create a random directory in /tmp)")
parser.add_option("--type", action = "store", type = "string", dest = "workflow_type",
		  help = "specifies what type of workflow we are debugging (available types: condor)")

# Parse command line options
(options, args) = parser.parse_args()

print "%s: initializing..." % (prog_base)

# Copy options from the command line parser

if options.vb == 0:
    lvl = logging.WARN
elif options.vb == 1:
    lvl = logging.INFO
else:
    lvl = logging.DEBUG
# Set logging level
logger.setLevel(lvl)

if options.run_monitord is not None:
    run_monitord = options.run_monitord
if options.strict_mode is not None:
    strict_mode = options.strict_mode
if options.summary_mode is not None:
    summary_mode = options.summary_mode
if options.quiet_mode is not None:
    quiet_mode = options.quiet_mode
if options.use_files is not None:
    use_files = True
if options.print_options is not None:
    my_options = options.print_options.split(",")
    if "pre" in my_options or "all" in my_options:
	print_pre_script = 1
    if "invocation" in my_options or "all" in my_options:
	print_invocation = 1
if options.output_dir is not None:
    output_dir = options.output_dir
if options.top_dir is not None:
    top_dir = os.path.abspath(options.top_dir)
if options.debug_job is not None:
    debug_job = options.debug_job
    # Enables the debugging mode
    debug_mode = 1
if options.debug_dir is not None:
    debug_dir = options.debug_dir
if options.workflow_type is not None:
    workflow_type = options.workflow_type

if options.dag_filename is not None:
    dag_path = options.dag_filename
    input_dir = os.path.abspath(os.path.split(dag_path)[0])
    # Assume current directory if input dir is empty
    if len(input_dir) == 0:
	input_dir = os.getcwd()
else:
    # Select directory where jobstate.log is located
    if options.input_dir is not None:
	input_dir = os.path.abspath(options.input_dir)
    else:
        if len(args) > 1:
            parser.print_help()
            sys.exit(1)
        elif len(args) == 1:
            input_dir = args[0]
        else:
            input_dir = os.getcwd()

if debug_mode == 1:
    # Enter debug mode if job name given
    # This function does not return
    debug_workflow()

# Run the analyzer
if use_files:
    analyze_files()
else:
    analyze_db(options.config_properties)

# Done!
print "Done".center(80, '*')
print
print "%s: end of status report" % (prog_base)
print