DQM_Tier0Wrapper_tf Namespace Reference

Functions
def	getSubFileMap (fname, nevts=0)
def	publish_success_to_mq (run, ptag, stream, incr, ami, procpass, hcfg, isprod, parmap)
def	genmd5sum (filename)
def	dq_combined_trf (jsonfile, outmap)
def	dq_trf_wrapper (jsonfile)

Variables
	jsonfile = sys.argv[1][len('--argJSON='):]
	main() More...

Function Documentation

dq_combined_trf()

def DQM_Tier0Wrapper_tf.dq_combined_trf	(		jsonfile,
			outmap
	)

Definition at line 140 of file DQM_Tier0Wrapper_tf.py.

def dq_combined_trf(jsonfile, outmap):
 
  print("\n##################################################################")
  print("## STEP 1: creating file with list of root files ...")
  print("##################################################################\n")
  
  nfiles=0
 
  try:
    # extract parameters from json file
    print("Using json file ", jsonfile, " for input parameters")
    f = open(jsonfile, 'r')
    parmap = json.load(f)
    f.close()
 
    print("\nFull Tier-0 run options:\n")
    pprint.pprint(parmap)
 
    inputfilelist = parmap.get('inputHistFiles', [])
    nfiles = len(inputfilelist) 
    histMergeCompressionLevel=parmap.get('histMergeCompressionLevel', 1)
    histMergeDebugLevel=parmap.get('histMergeDebugLevel', 0)
  except: 
    outmap['exitCode'] = 101
    outmap['exitAcronym'] = 'TRF_NOINPUT'
    outmap['exitMsg'] = 'Trouble reading json input dict.'
    traceback.print_exc()
    return
  
  if not nfiles :   # problem with job definition or reading json file
    outmap['exitCode'] = 102
    outmap['exitAcronym'] = 'TRF_NOINPUT'
    outmap['exitMsg'] = 'Empty input file list.'
    return
 
  histtmpflist = []
  histtmpdsname = ''
  nevts = 0
 
  try:
    if isinstance(inputfilelist[0], six.text_type) :  
      histtmpdsname = (inputfilelist[0]).split('#')[0]
      for val in inputfilelist :
        histtmpflist.append(val.split('#')[1])
    
    elif isinstance(inputfilelist[0], dict) :
      histtmpdsname = inputfilelist[0]['dsn']
      for fdict in inputfilelist :
        histtmpflist.append(fdict['lfn'])
        nevt = fdict.get('events', 0)
        if nevt is None:
          nevt=0
          print("WARNING Can't get number of events from input json file")
        nevts+=nevt
  
    f = open('hist_merge.list', 'w')
    txtstr = ""
    for hf in histtmpflist : 
      txtstr += "%s\n" % hf
    f.write(txtstr)
    f.close()
 
    cmd = "cat hist_merge.list"
    (s,o) = getstatusoutput(cmd)
    print("\nContents of file hist_merge.list:\n")
    print(o)
  except:
    outmap['exitCode'] = 103
    outmap['exitAcronym'] = 'TRF_INPUTINFO'
    outmap['exitMsg'] = 'ERROR: crash in assembling input file list (STEP 1)'
    traceback.print_exc()
    return
 
  try:
    print("\n##################################################################")
    print("## STEP 2: determining job parameters...")
    print("##################################################################\n")
 
    skipMerge = parmap.get('skipMerge', 'False')
    if skipMerge == 'True':
      if nfiles != 1:
        print("ERROR: skipMerge specified but something other than one input file specified")
        outmap['exitCode'] = 108
        outmap['exitAcronym'] = 'TRF_INPUTINFO'
        outmap['exitMsg'] = 'ERROR: skipMerge specified but something other than one input file specified (STEP 1)'
        return
      histdsname = histtmpdsname
      histfile = histtmpflist[0]
    else:
      # normal output file determination
      histdsname = (parmap['outputHistFile']).split('#')[0]
      histfile = (parmap['outputHistFile']).split('#')[1]
    amitag = histdsname.split('.')[5]
 
 
    # incremental mode on/off
    incr = parmap.get('incrementalMode', 'False') 
    
    # post-processing on/off
    postproc = parmap.get('postProcessing', 'True')
 
    # database uploading on/off
    allowCOOLUpload = parmap.get('allowCOOLUpload', 'True')
 
    # do web display
    doWebDisplay = parmap.get('doWebDisplay', 'True')
 
    # extra parameters for merge
    mergeParams = parmap.get('mergeParams', '')
 
    # production mode
    productionMode = parmap.get('productionMode', 'True')
    if productionMode != 'True' and incr == 'True':
      print("Production mode is not True, turning off incremental mode")
      incr = 'False'
    
    # server list
    servers = parmap.get('servers', 'False')
    os.environ['DQC_SERVERS'] = servers
 
    # get file paths, put into environment vars
    filepaths = parmap.get('filepaths', None)
    if filepaths and isinstance(filepaths, dict):
      if 'basename' not in filepaths:
        print("Improperly formed 'filepaths' (no 'basename')")
      else:
        for evtclass in ('Collisions', 'Cosmics', 'HeavyIons'):
          if evtclass not in filepaths:
            print("Improperly formed 'filepaths' (no '%s')" % evtclass)
          else:
            clinfo = filepaths[evtclass]
            for timeclass in ('run', 'minutes10', 'minutes30'):
              if timeclass not in clinfo:
                print("Improperly formed 'filepaths[%s]' (no '%s')" % (evtclass, timeclass))
              else:
                dqcenvvar = 'DQC_HCFG_%s_%s' % (evtclass.upper(), timeclass.upper())
                fpath = os.path.join(filepaths['basename'], clinfo[timeclass])
                print("Setting %s = %s" % (dqcenvvar, fpath))
                os.environ[dqcenvvar] = fpath
 
    # extract info from dataset name
    # AMI project name
    # override if tag has been specified in parmap
    try :
      dqproject = histdsname.split('.')[0]
    except :
      dqproject = 'data_test'
    dqproject = parmap.get('projectTag', dqproject)
    
    # run number
    if 'runNumber' in parmap : 
      runnr = parmap['runNumber']
    else :
      try :
        runnr = int(histdsname.split('.')[1])
      except :
        runnr = 1234567890
 
    # stream name  
    if 'streamName' in parmap : 
      stream = parmap['streamName']
    else :
      try :
        stream = histdsname.split('.')[2]
      except :
        stream = 'test_dummy'
    
    # processing pass number  
    procnumber = 99
    MAX_XMLRPC_TRIES = 5 
    if 'procNumber' in parmap : 
      procnumber = parmap['procNumber']
    else :
      n_xmlrpc_tries = 1
      while n_xmlrpc_tries <= MAX_XMLRPC_TRIES :
        procnumber = 99
        try :
          xmlrpcserver = xmlrpclib.ServerProxy('http://proc-pass-atlasdqm.app.cern.ch')
          #xmlrpcserver = xmlrpclib.ServerProxy('http://atlasdqm.cern.ch:8888') ##old server
          procnumber = xmlrpcserver.get_next_proc_pass(runnr, stream, 'tier0')
          break 
        except :
          print('Web service connection failed, attempt', n_xmlrpc_tries, 'of', MAX_XMLRPC_TRIES)
          n_xmlrpc_tries += 1
          if n_xmlrpc_tries <= MAX_XMLRPC_TRIES:
            time.sleep(20*2**n_xmlrpc_tries)
 
 
    print("Job parameters:\n")
    print("  Run number:      ", runnr)
    print("  Stream name:     ", stream)
    print("  Processing pass: ", procnumber)
    print("  Incremental mode:", incr)
    print("  Post-processing: ", postproc)
    print("  COOL uploads:    ", allowCOOLUpload)
    print("  Production mode: ", productionMode)
    print("  Skip merge:      ", skipMerge)
    print("  Merge parameters:", mergeParams)
    if servers == []:
        print("  EOS only:    True")
 
  except:
    outmap['exitCode'] = 104
    outmap['exitAcronym'] = 'TRF_JOBPARS'
    outmap['exitMsg'] = 'Error in determining job parameters (STEP 2).'
    traceback.print_exc()
    return
 
  # environment setting
  os.environ['DQPRODUCTION'] = '1' if productionMode == 'True' else '0'
  print("Setting env variable DQPRODUCTION to %s\n" % os.environ['DQPRODUCTION'])
  os.environ['DQ_STREAM'] = stream
  print("Setting env variable DQ_STREAM to %s\n" % os.environ['DQ_STREAM'])
  os.environ['COOLUPLOADS'] = '1' if allowCOOLUpload == 'True' and productionMode == 'True' else '0'
  print("Setting env variable COOLUPLOADS to %s\n" % os.environ['COOLUPLOADS'])
 
  if skipMerge != 'True':
    try:
      print("\n##################################################################")
      print("## STEP 3: running histogram merging procedure ...")
      print("##################################################################\n")
 
 
      if postproc == 'True' :
        if incr == 'True':
          cmd = "python -u `which DQHistogramMerge.py` hist_merge.list %s 1 1 %d %d " % (histfile,histMergeCompressionLevel,histMergeDebugLevel)
        else:        
          cmd = "python -u `which DQHistogramMerge.py` hist_merge.list %s 1 0 %d %d"  % (histfile,histMergeCompressionLevel,histMergeDebugLevel)
      else :  
        cmd = "python -u `which DQHistogramMerge.py` hist_merge.list %s 0 0 %d %d"    % (histfile,histMergeCompressionLevel,histMergeDebugLevel)
      
      cmd += (f' {mergeParams}' if mergeParams else '')
 
      print("Histogram merging command:\n")
      print(cmd)
      print("\n##################################################################\n")
      
      print("## ... logfile from DQHistogramMerge.py: ")
      print("--------------------------------------------------------------------------------")
      tstart = time.time()
      # execute command
      retcode1 = os.system(cmd)
      print("--------------------------------------------------------------------------------")
      t1 = time.time()
      dt1 = int(t1 - tstart)
      
      print("\n## DQHistogramMerge.py finished with retcode = %s" % retcode1)
      print("## ... elapsed time: ", dt1, " sec")
 
      if retcode1 != 0 :
        outmap['exitCode'] = retcode1
        outmap['exitAcronym'] = 'TRF_DQMHISTMERGE_EXE'
        outmap['exitMsg'] = 'ERROR: DQHistogramMerge.py execution problem! (STEP 3).'
        print("ERROR: DQHistogramMerge.py execution problem!")
        retcode = retcode1
        txt = 'DQHistogramMerge.py execution problem'
        try:
          try:
            with open('hist_merge.list','r') as infilelist:
              for infname in infilelist:
                genmd5sum(infname.rstrip(os.linesep))
          except:
            pass
          genmd5sum(histfile)
          DQResFile="DQResourceUtilization.txt"
          if os.path.exists(DQResFile):
            print("dumping resource utilization log")
            with open(DQResFile) as resfile:
              for resline in resfile:
                print(resline, end=' ')
        except:
          outmap['exitMsg'] = 'ERROR: DQHistogramMerge.py execution problem + problem dumping DQResourceUtilization! (STEP 3).'
          traceback.print_exc()
          print("ERROR: DQHistogramMerge.py execution problem + problem dumping DQResourceUtilization!")
        return
 
      if postproc == 'True' and incr == 'False':
        print("\n##################################################################")
        print("## STEP 3b: copying postprocessing output to AFS ...")
        print("##################################################################\n")
 
        cmd = "python -u `which DQFileMove.py` %s %s_%s_%s" % (dqproject, runnr, stream, procnumber)
 
        print("File move command:\n")
        print(cmd)
        print("\n##################################################################\n")
 
        print("## ... logfile from DQFileMove.py: ")
        print("--------------------------------------------------------------------------------")
        # execute command
        retcode1b = os.system(cmd)
        print("--------------------------------------------------------------------------------")
        t1b = time.time()
        dt1b = int(t1b - t1)
        t1 = t1b
 
        print("\n## DQFileMove.py finished with retcode = %s" % retcode1b)
        print("## ... elapsed time: ", dt1b, " sec")
    except:
        outmap['exitCode'] = 105
        outmap['exitAcronym'] = 'TRF_DQMHISTMERGE_EXE'
        outmap['exitMsg'] = 'ERROR: Failure in histogram merging or copying postprocessing output to AFS (STEP 3/3b).'
        traceback.print_exc()
        return
  else:
    print("\n##################################################################")
    print("## HISTOGRAM MERGE/POSTPROCESSING SKIPPED BY USER REQUEST")
    print("##################################################################\n")
    t1 = time.time()
 
  try:
    retcode2 = 0
    dt2 = 0
    if doWebDisplay == 'True':
      print("\n##################################################################")
      print("## STEP 4: running web-display creation procedure ...")
      print("##################################################################\n")
 
      cmd = "python -u `which DQWebDisplay.py` %s %s %s %s stream=%s" % (histfile, dqproject, procnumber, incr, stream)
 
      print("Web display creation command:\n")
      print(cmd)
      print("\n##################################################################\n")
 
      print("## ... logfile from DQWebDisplay.py: ")
      print("--------------------------------------------------------------------------------")
      # execute command
      retcode2 = os.system(cmd)
      print('DO NOT REPORT "Error in TH1: cannot merge histograms" ERRORS! THESE ARE IRRELEVANT!')
      print("--------------------------------------------------------------------------------")
      t2 = time.time()
      dt2 = int(t2 - t1)
 
      print("\n## DQWebDisplay.py finished with retcode = %s" % retcode2)
      print("## ... elapsed time: ", dt2, " sec")
      if not (retcode2 >> 8) in (0, 5) :
        print("ERROR: DQWebDisplay.py execution problem!")
        outmap['exitCode'] = retcode2
        outmap['exitAcronym'] = 'TRF_DQMDISPLAY_EXE'
        outmap['exitMsg'] = 'ERROR: DQWebDisplay.py execution problem! (STEP 4).'
        try:
          with open('hist_merge.list','r') as infilelist:
            for infname in infilelist:
              genmd5sum(infname.rstrip(os.linesep))
        except:
          pass
        genmd5sum(histfile)
        return
      if productionMode == 'True': 
          try:
              print('Publishing to message service')
              publish_success_to_mq(runnr, dqproject, stream, incr=(incr=='True'), ami=amitag, procpass=procnumber, hcfg=filepaths, isprod=(productionMode=='True'), parmap=parmap)
          except:
              outmap['exitCode'] = 106
              outmap['exitAcronym'] = 'TRF_DQMDISPLAY_EXE'
              outmap['exitMsg'] = 'ERROR: Failure in publishing info to messaging service (STEP 4).'
              traceback.print_exc()
              return
    else:
      print("\n##################################################################")
      print("## WEB DISPLAY CREATION SKIPPED BY USER REQUEST")
      print("##################################################################\n")
      print('Web display off, not publishing to message service')
  except: 
    outmap['exitCode'] = 106
    outmap['exitAcronym'] = 'TRF_DQMDISPLAY_EXE'
    outmap['exitMsg'] = 'ERROR: Failure in web-display creation procedure (STEP 4).'
    print('ERROR: Failure in web-display creation procedure (STEP 4).')
    traceback.print_exc()
    return
  
  print("\n##################################################################")
  print("## STEP 5: finishing the job ...")
  print("##################################################################\n")
        
  # get info for report json file
  try:
    outfiles = [getSubFileMap(histfile, nevts=nevts)]
    # assemble job report map
    outmap['files']['output'][0]['dataset'] = histdsname
    outmap['files']['output'][0]['subFiles'] = outfiles
    outmap['resource']['transform']['processedEvents'] = int(nevts)
    return
  except: 
    outmap['exitCode'] = 107
    outmap['exitAcronym'] = 'TRF_JOBREPORT'
    outmap['exitMsg'] = 'ERROR: in job report creation (STEP 5)'
    print("ERROR: in job report creation (STEP 5) !")
    traceback.print_exc()
    return
 

dq_trf_wrapper()

def DQM_Tier0Wrapper_tf.dq_trf_wrapper

(

jsonfile

)

Definition at line 531 of file DQM_Tier0Wrapper_tf.py.

def dq_trf_wrapper(jsonfile):
  print("\n##################################################################")
  print("##              ATLAS Tier-0 Offline DQM Processing             ##")
  print("##################################################################\n")
 
  outmap = { 'exitAcronym' : 'OK',
               'exitCode' : 0,
               'exitMsg' : 'trf finished OK',
               'files' : { 'output' : [{ 'dataset' : '',
                                         'subFiles' : [ {},
                                                      ]}
                                    ] },
                'resource' : { 'transform' : { 'processedEvents' : 0 } }
                 }
 
  # dq_combined_trf will update outmap
  tstart = time.time()
  dq_combined_trf(jsonfile, outmap)
  outmap['resource']['transform']['wallTime'] = int(time.time() - tstart) 
   
  # dump json report map
  f = open('jobReport.json', 'w')
  json.dump(outmap, f)
  f.close()
 
  # summarize status
  print("\n## ... job finished with retcode : %s" % outmap['exitCode'])
  print("## ... error acronym: ", outmap['exitAcronym'])
  print("## ... job status message: ", outmap['exitMsg'])
  print("## ... elapsed time: ", outmap['resource']['transform']['wallTime'], "sec")
  print("##")
  print("##################################################################")
  print("## End of job.")
  print("##################################################################\n")
 
 

genmd5sum()

def DQM_Tier0Wrapper_tf.genmd5sum

(

filename

)

Definition at line 125 of file DQM_Tier0Wrapper_tf.py.

def genmd5sum(filename):
  md5summer=hashlib.md5()
  if os.path.isfile(filename):
    try:
      with open(filename,'rb') as infil:
        while True:
          fs=infil.read(8192)
          if not fs:
            break
          md5summer.update(fs)
    except:
      pass
  print("md5 sum of the \"%s\" is %s"%(filename,md5summer.hexdigest()))
  return
      

getSubFileMap()

def DQM_Tier0Wrapper_tf.getSubFileMap	(		fname,
			nevts = 0
	)

Definition at line 78 of file DQM_Tier0Wrapper_tf.py.

def getSubFileMap(fname, nevts=0) :
    if os.path.isfile(fname) :
        sz = os.path.getsize(fname)
        map = { 'name': fname,
                'file_size' : sz,
                'nentries' : nevts,
              }
    else : 
        map = {}
    return map
 

publish_success_to_mq()

def DQM_Tier0Wrapper_tf.publish_success_to_mq	(		run,
			ptag,
			stream,
			incr,
			ami,
			procpass,
			hcfg,
			isprod,
			parmap
	)

Definition at line 89 of file DQM_Tier0Wrapper_tf.py.

def publish_success_to_mq(run, ptag, stream, incr, ami, procpass, hcfg, isprod, parmap):
  import stomp, json
  from DataQualityUtils import stompconfig
  dest='/topic/atlas.dqm.progress'
  conn=stomp.Connection([('atlas-mb.cern.ch', 61013)])
  conn.connect(wait=True, **stompconfig.config())
 
  eos_only = False
 
  servers = parmap.get('servers', 'False')
  if servers == [] or servers == '':
     eos_only = True
 
 
  body = {
    'run': run,
    'project_tag': ptag,
    'stream': stream,
    'ami': ami,
    'pass': procpass,
    'hcfg': hcfg,
    'eos_only': eos_only,
    }
  headers = {
    'MsgClass':'DQ', 
    'MsgType': (('' if isprod else 'Development') +
                ('WebDisplayRunComplete' if not incr else 'WebDisplayIncremental')),
    'type':'textMessage', 
    'persistent': 'true',
    'destination': dest,
    }
  conn.send(body=json.dumps(body), destination=dest,headers=headers,ack='auto')
  conn.disconnect()
 

Variable Documentation

jsonfile

DQM_Tier0Wrapper_tf.jsonfile = sys.argv[1][len('--argJSON='):]

main()

Definition at line 601 of file DQM_Tier0Wrapper_tf.py.