| 1 |
paus |
1.1.2.1 |
#!/usr/bin/env python
|
| 2 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 3 |
|
|
# Script to automatically download a MIT dataset to our local cluster
|
| 4 |
|
|
#
|
| 5 |
|
|
# The download of the MIT dataset is organized in accordance with the dataset production logic. In
|
| 6 |
|
|
# general it is allowed to download the dataset from any location of a properly configured storage
|
| 7 |
|
|
# element. The script will do all most obvious tests to ensure efficient and safe download. For
|
| 8 |
|
|
# performance reason a checksum is not calculated. This ommission is considered completely safe as
|
| 9 |
|
|
# failures will be identified in the analysis phase and the rare occasions will be more effective to
|
| 10 |
|
|
# fix by hand.
|
| 11 |
|
|
#
|
| 12 |
|
|
# At present the download proceeds in one thread (one file at a time) which for performance reasons
|
| 13 |
|
|
# might not be optimal.
|
| 14 |
|
|
#
|
| 15 |
|
|
# Author: C.Paus (July 1, 2008)
|
| 16 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 17 |
|
|
# Missing but desired features:
|
| 18 |
|
|
# + accounting of size of each file
|
| 19 |
|
|
# + accounting of locally available files (avoid copying already existing files)
|
| 20 |
|
|
# + determine full list of files before starting to copy
|
| 21 |
|
|
# + minimal success check of the copy
|
| 22 |
|
|
# + calculate total data volume (to copy, already copied etc.)
|
| 23 |
|
|
# + add feature to check the castor status
|
| 24 |
|
|
# - add time estimates and progressions for copies
|
| 25 |
|
|
# - multi downloads to enhance performance
|
| 26 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 27 |
|
|
import os,sys,getopt,re,string
|
| 28 |
|
|
|
| 29 |
|
|
def Seconds():
|
| 30 |
|
|
for secs in os.popen('date +%s').readlines():
|
| 31 |
|
|
secs = int(secs[:-1])
|
| 32 |
|
|
return secs
|
| 33 |
|
|
|
| 34 |
|
|
def InSkipList(file,list):
|
| 35 |
|
|
for entry in list:
|
| 36 |
|
|
if entry == file:
|
| 37 |
|
|
return True
|
| 38 |
|
|
return False
|
| 39 |
|
|
|
| 40 |
|
|
def DecodeSrmLs(line):
|
| 41 |
|
|
line = line.strip()
|
| 42 |
|
|
f = line.split(" ")
|
| 43 |
|
|
size = f[0]
|
| 44 |
|
|
f = f[1].split("/")
|
| 45 |
|
|
file = f.pop()
|
| 46 |
|
|
list = [ size, file ]
|
| 47 |
|
|
return list
|
| 48 |
|
|
|
| 49 |
|
|
def DecodeRfDir(line):
|
| 50 |
|
|
line = line.strip()
|
| 51 |
|
|
f = line.split(" ")
|
| 52 |
|
|
file = f.pop()
|
| 53 |
|
|
size = f[4]
|
| 54 |
|
|
list = [ size, file ]
|
| 55 |
|
|
return list
|
| 56 |
|
|
|
| 57 |
|
|
def BuildFileList(cmd):
|
| 58 |
|
|
isSrm = 1
|
| 59 |
|
|
f = cmd.split(" ")
|
| 60 |
|
|
if f[0] == 'rfdir':
|
| 61 |
|
|
isSrm = 0
|
| 62 |
|
|
|
| 63 |
|
|
fileList = {}
|
| 64 |
|
|
|
| 65 |
|
|
if debug == 1:
|
| 66 |
|
|
print ' Debug:: list: ' + cmd
|
| 67 |
|
|
for line in os.popen(cmd).readlines(): # run command
|
| 68 |
|
|
line = line[:-1] # strip '\n'
|
| 69 |
|
|
#print ' Line: ' + line
|
| 70 |
|
|
f = line.split(" ")
|
| 71 |
|
|
##if isSrm == 1:
|
| 72 |
|
|
## f = DecodeSrmLs(line)
|
| 73 |
|
|
##else:
|
| 74 |
|
|
## f = DecodeRfDir(line)
|
| 75 |
|
|
size = f[0]
|
| 76 |
|
|
file = f[1]
|
| 77 |
|
|
f = file.split("/")
|
| 78 |
|
|
file = f[-1]
|
| 79 |
|
|
if debug == 1:
|
| 80 |
|
|
print ' Debug:: adding: ' + file + ' with size ' + size
|
| 81 |
|
|
fileList[file] = int(size)
|
| 82 |
|
|
|
| 83 |
|
|
return fileList
|
| 84 |
|
|
|
| 85 |
|
|
def BuildStagedFileList(storagePath,allFileList,cacheFile):
|
| 86 |
|
|
|
| 87 |
|
|
# initialize the basics
|
| 88 |
|
|
fileList = {}
|
| 89 |
|
|
f = storagePath.split("=");
|
| 90 |
|
|
rfPath = f[-1]
|
| 91 |
|
|
|
| 92 |
|
|
# if this is not castor, trick it and mark them as staged
|
| 93 |
|
|
if not (re.search('/castor/',rfPath)):
|
| 94 |
|
|
for file, size in allFileList.iteritems():
|
| 95 |
|
|
fileList[file] = 'STAGED'
|
| 96 |
|
|
return fileList
|
| 97 |
|
|
|
| 98 |
|
|
# here we deal with castor
|
| 99 |
|
|
if debug == 1:
|
| 100 |
|
|
print ' Debug:: rfpath: ' + rfPath
|
| 101 |
|
|
if os.path.exists(cacheFile) and noCache == 0:
|
| 102 |
|
|
print ' Using the cached stager queries at ' + cacheFile
|
| 103 |
|
|
for file, size in allFileList.iteritems():
|
| 104 |
|
|
fullFile = rfPath + '/' + file
|
| 105 |
|
|
if debug == 1:
|
| 106 |
|
|
print ' Debug:: full file name: ' + fullFile
|
| 107 |
|
|
if os.path.exists(cacheFile) and noCache == 0:
|
| 108 |
|
|
cmd = 'grep ' + file + ' ' + cacheFile
|
| 109 |
|
|
else:
|
| 110 |
|
|
cmd = 'stager_qry -M ' + fullFile
|
| 111 |
|
|
fileList[file] = 'undefined'
|
| 112 |
|
|
for line in os.popen(cmd).readlines(): # run command
|
| 113 |
|
|
line = line[:-1]
|
| 114 |
|
|
f = line.split(" ")
|
| 115 |
|
|
if f[0] == fullFile:
|
| 116 |
|
|
f = line.split(" ")
|
| 117 |
|
|
status = f[-1]
|
| 118 |
|
|
fileList[file] = status
|
| 119 |
|
|
|
| 120 |
|
|
return fileList
|
| 121 |
|
|
|
| 122 |
|
|
def CacheStagedFileList(cacheFile,storagePath,stagedFileList):
|
| 123 |
|
|
print ' Caching stager query status to ' + cacheFile
|
| 124 |
|
|
f = storagePath.split("=");
|
| 125 |
|
|
rfPath = f[-1]
|
| 126 |
|
|
fileOutput = open(cacheFile,'w')
|
| 127 |
|
|
for file, status in stagedFileList.iteritems():
|
| 128 |
|
|
line = rfPath + '/' + file + ' xyz@castorns ' + status + '\n'
|
| 129 |
|
|
fileOutput.write(line)
|
| 130 |
|
|
fileOutput.close()
|
| 131 |
|
|
|
| 132 |
|
|
def CopyFile(storageEle,storagePath,storageUrl,file,localDir):
|
| 133 |
|
|
deltaT = 0
|
| 134 |
|
|
print ' working on file: ' + file + ' to ' + localDir + \
|
| 135 |
|
|
' (size: %d MB) '%(int(size)/1024/1024)
|
| 136 |
|
|
if storageEle == 'srm-cms.cern.ch':
|
| 137 |
|
|
f = storagePath.split("=");
|
| 138 |
|
|
rfPath = f[-1]
|
| 139 |
|
|
cpy = 'rfcp ' + rfPath + '/' + file + ' ' + localPath + '/' \
|
| 140 |
|
|
+ mitCfg + '/' + version + '/' + mitDataset + '/' + file
|
| 141 |
|
|
#print ' using rfcp.... ' + cpy
|
| 142 |
|
|
#sys.exit(0)
|
| 143 |
|
|
else:
|
| 144 |
|
|
#storageUrl = 'srm://' + storageEle + ':8443' + storagePath
|
| 145 |
|
|
cpy = 'lcg-cp ' + storageUrl + '/' + file + ' file:////' + localPath + '/' \
|
| 146 |
|
|
+ mitCfg + '/' + version + '/' + mitDataset + '/' + file
|
| 147 |
|
|
|
| 148 |
|
|
# Check whether the file size make sense (zero length files are probably not yet ready to
|
| 149 |
|
|
# copy and will not be transfered
|
| 150 |
|
|
if size < 1:
|
| 151 |
|
|
print ' WARNING - file size is <1b. Probably this file is not yet ready. Stop copy.'
|
| 152 |
|
|
else:
|
| 153 |
|
|
if debug == 1:
|
| 154 |
|
|
print ' Debug:: copy: ' + cpy
|
| 155 |
|
|
start = Seconds()
|
| 156 |
|
|
status = os.system(cpy)
|
| 157 |
|
|
end = Seconds()
|
| 158 |
|
|
deltaT = end - start
|
| 159 |
|
|
|
| 160 |
|
|
return deltaT
|
| 161 |
|
|
|
| 162 |
|
|
def StageFile(storagePath,storageUrl,file):
|
| 163 |
|
|
print ' staging in file: ' + file
|
| 164 |
|
|
if storageEle == 'srm-cms.cern.ch':
|
| 165 |
|
|
f = storagePath.split("=");
|
| 166 |
|
|
rfPath = f[-1]
|
| 167 |
|
|
stg = 'stager_get -M ' + rfPath + '/' + file
|
| 168 |
|
|
else:
|
| 169 |
|
|
#storageUrl = 'srm://' + storageEle + ':8443' + storagePath
|
| 170 |
|
|
stg = 'echo lcg-cp ' + storageUrl + '/' + file + ' file:////' + localPath + '/' \
|
| 171 |
|
|
+ mitCfg + '/' + version + '/' + mitDataset + '/' + file
|
| 172 |
|
|
|
| 173 |
|
|
if debug == 1:
|
| 174 |
|
|
print ' Debug:: stage: ' + stg
|
| 175 |
|
|
status = os.system(stg)
|
| 176 |
|
|
|
| 177 |
|
|
#===================================================================================================
|
| 178 |
|
|
# Main starts here
|
| 179 |
|
|
#===================================================================================================
|
| 180 |
|
|
# Define string to explain usage of the script
|
| 181 |
|
|
usage = "Usage: downloadSample.py --cmsDataset=<name> | --mitDataset=<name>\n"
|
| 182 |
|
|
usage += " --mitCfg=<name>\n"
|
| 183 |
|
|
usage += " --version=<version>\n"
|
| 184 |
|
|
usage += " --cmssw=<name>\n"
|
| 185 |
|
|
usage += " --localStorageUrl=<name>\n"
|
| 186 |
|
|
usage += " --localPath=<dir>\n"
|
| 187 |
|
|
usage += " --skip=<file list>\n"
|
| 188 |
|
|
usage += " --backward\n"
|
| 189 |
|
|
usage += " --debug\n"
|
| 190 |
|
|
usage += " --help\n"
|
| 191 |
|
|
|
| 192 |
|
|
# Define the valid options which can be specified and check out the command line
|
| 193 |
|
|
valid = ['cmsDataset=','mitDataset=','mitCfg=','version=','cmssw=','pattern=','localStorageUrl=',
|
| 194 |
|
|
'localPath=','noCache','skip=',
|
| 195 |
|
|
'forceCopy','backward',
|
| 196 |
|
|
'debug','help']
|
| 197 |
|
|
try:
|
| 198 |
|
|
opts, args = getopt.getopt(sys.argv[1:], "", valid)
|
| 199 |
|
|
except getopt.GetoptError, ex:
|
| 200 |
|
|
print usage
|
| 201 |
|
|
print str(ex)
|
| 202 |
|
|
sys.exit(1)
|
| 203 |
|
|
|
| 204 |
|
|
# --------------------------------------------------------------------------------------------------
|
| 205 |
|
|
# Get all parameters for the production
|
| 206 |
|
|
# --------------------------------------------------------------------------------------------------
|
| 207 |
|
|
# Set defaults for each option
|
| 208 |
|
|
cmsDataset = None
|
| 209 |
|
|
mitDataset = None
|
| 210 |
|
|
skip = ''
|
| 211 |
|
|
skipList = []
|
| 212 |
|
|
mitCfg = 'filler'
|
| 213 |
|
|
version = '012'
|
| 214 |
|
|
cmssw = ''
|
| 215 |
|
|
blockLocal = 0
|
| 216 |
|
|
localStorageUrl = ''
|
| 217 |
|
|
localPath = '/server/02b/mitprod'
|
| 218 |
|
|
pattern = ''
|
| 219 |
|
|
noCache = 0
|
| 220 |
|
|
backward = ''
|
| 221 |
|
|
forceCopy = False
|
| 222 |
|
|
debug = 0
|
| 223 |
|
|
cmsswCfg = 'cmssw.cfg'
|
| 224 |
|
|
|
| 225 |
|
|
# Read new values from the command line
|
| 226 |
|
|
for opt, arg in opts:
|
| 227 |
|
|
if opt == '--help':
|
| 228 |
|
|
print usage
|
| 229 |
|
|
sys.exit(0)
|
| 230 |
|
|
if opt == '--cmsDataset':
|
| 231 |
|
|
cmsDataset = arg
|
| 232 |
|
|
if opt == '--mitDataset':
|
| 233 |
|
|
mitDataset = arg
|
| 234 |
|
|
if opt == '--mitCfg':
|
| 235 |
|
|
mitCfg = arg
|
| 236 |
|
|
if opt == '--version':
|
| 237 |
|
|
version = arg
|
| 238 |
|
|
if opt == '--cmssw':
|
| 239 |
|
|
cmssw = arg
|
| 240 |
|
|
if opt == '--pattern':
|
| 241 |
|
|
pattern = arg
|
| 242 |
|
|
if opt == '--localStorageUrl':
|
| 243 |
|
|
localStorageUrl = arg
|
| 244 |
|
|
if opt == '--localPath':
|
| 245 |
|
|
blockLocal = 1
|
| 246 |
|
|
localPath = arg
|
| 247 |
|
|
if opt == '--skip':
|
| 248 |
|
|
skip = arg
|
| 249 |
|
|
skipList = skip.split(',')
|
| 250 |
|
|
if opt == '--noCache':
|
| 251 |
|
|
noCache = 1
|
| 252 |
|
|
if opt == '--backward':
|
| 253 |
|
|
backward = ' -r '
|
| 254 |
|
|
if opt == '--forceCopy':
|
| 255 |
|
|
forceCopy = True
|
| 256 |
|
|
if opt == '--debug':
|
| 257 |
|
|
debug = 1
|
| 258 |
|
|
|
| 259 |
|
|
# Deal with obvious problems
|
| 260 |
|
|
if cmsDataset == None and mitDataset == None:
|
| 261 |
|
|
cmd = '--cmsDataset option not provided. This is required.'
|
| 262 |
|
|
raise RuntimeError, cmd
|
| 263 |
|
|
|
| 264 |
|
|
crabFile = mitCfg + '/' + version + '/' + 'crab.cfg'
|
| 265 |
|
|
if not os.path.exists(crabFile):
|
| 266 |
|
|
cmd = 'Crab file not found: %s' % crabFile
|
| 267 |
|
|
raise RuntimeError, cmd
|
| 268 |
|
|
cmsswFile = mitCfg + '/' + version + '/' + cmsswCfg
|
| 269 |
|
|
if not os.path.exists(cmsswFile):
|
| 270 |
|
|
cmd = 'Cmssw file not found: %s' % cmsswFile
|
| 271 |
|
|
cmsswCfg = 'cmssw.py'
|
| 272 |
|
|
cmsswFile = mitCfg + '/' + version + '/' + cmsswCfg
|
| 273 |
|
|
if not os.path.exists(cmsswFile):
|
| 274 |
|
|
cmd = 'Cmssw file not found: %s' % cmsswFile
|
| 275 |
|
|
cmd = ' XXXX ERROR no valid configuration found XXXX'
|
| 276 |
|
|
raise RuntimeError, cmd
|
| 277 |
|
|
|
| 278 |
|
|
# Resolve the other mitCfg parameters from the configuration file
|
| 279 |
|
|
cmd = 'cat ' + mitCfg + '/' + version + '/' + 'Productions'
|
| 280 |
|
|
if cmssw != '':
|
| 281 |
|
|
cmd = cmd + '.' + cmssw
|
| 282 |
|
|
|
| 283 |
|
|
join = 0
|
| 284 |
|
|
if cmsDataset == None:
|
| 285 |
|
|
cmsDataset = ''
|
| 286 |
|
|
else:
|
| 287 |
|
|
mitDataset = ''
|
| 288 |
|
|
|
| 289 |
|
|
fullLine = ''
|
| 290 |
|
|
bSlash = '\\';
|
| 291 |
|
|
for line in os.popen(cmd).readlines(): # run command
|
| 292 |
|
|
line = line[:-1]
|
| 293 |
|
|
#print 'Line: "' + line + '"'
|
| 294 |
|
|
# get ride of empty or commented lines
|
| 295 |
|
|
if line == '' or line[0] == '#':
|
| 296 |
|
|
continue
|
| 297 |
|
|
|
| 298 |
|
|
# join lines
|
| 299 |
|
|
if join == 1:
|
| 300 |
|
|
fullLine += line
|
| 301 |
|
|
else:
|
| 302 |
|
|
fullLine = line
|
| 303 |
|
|
|
| 304 |
|
|
# determine if finished or more is coming
|
| 305 |
|
|
if fullLine[-1] == bSlash:
|
| 306 |
|
|
join = 1
|
| 307 |
|
|
fullLine = fullLine[:-1]
|
| 308 |
|
|
else:
|
| 309 |
|
|
join = 0
|
| 310 |
|
|
# test whether there is a directory
|
| 311 |
|
|
names = fullLine.split() # splitting every blank
|
| 312 |
|
|
#print "FullLine: " + fullLine
|
| 313 |
|
|
#print "Datasets: " + mitDataset + ' -> ' + cmsDataset + "\n"
|
| 314 |
|
|
if names[0] == cmsDataset:
|
| 315 |
|
|
mitDataset = names[1] # this is the equivalent MIT name of the dataset
|
| 316 |
|
|
nevents = int(names[2]) # number of events to be used in the production
|
| 317 |
|
|
if names[4] != "-" and blockLocal == 0:
|
| 318 |
|
|
localPath = names[4]
|
| 319 |
|
|
#print "\n Sample Info: " + fullLine + "\n"
|
| 320 |
|
|
#print "\n Local path : " + localPath + ' -> ' + names[4] + "\n"
|
| 321 |
|
|
if names[1] == mitDataset:
|
| 322 |
|
|
cmsDataset = names[0] # this is the equivalent CMS name of the dataset
|
| 323 |
|
|
nevents = int(names[2]) # number of events to be used in the production
|
| 324 |
|
|
if names[4] != "-" and blockLocal == 0:
|
| 325 |
|
|
localPath = names[4]
|
| 326 |
|
|
#print "\n Sample Info: " + fullLine + "\n"
|
| 327 |
|
|
#print "\n Local path : " + localPath + ' -> ' + names[4] + "\n"
|
| 328 |
|
|
|
| 329 |
|
|
if mitDataset == "":
|
| 330 |
|
|
print "ERROR - dataset not defined."
|
| 331 |
|
|
sys.exit(0)
|
| 332 |
|
|
|
| 333 |
|
|
#cmd = 'grep ' + cmsDataset + ' ' + mitCfg + '/' + version + '/' + 'Productions'
|
| 334 |
|
|
#for file in os.popen(cmd).readlines(): # run command
|
| 335 |
|
|
# line = file[:-1] # strip '\n'
|
| 336 |
|
|
# # test whether there is a directory
|
| 337 |
|
|
# names = line.split() # splitting every blank
|
| 338 |
|
|
# mitDataset = names[1] # this is the equivalent MIT name of the dataset
|
| 339 |
|
|
# nevents = int(names[2]) # number of events to be used in the production
|
| 340 |
|
|
|
| 341 |
|
|
# Say what we do now
|
| 342 |
|
|
print '\n Preparing dataset for transfer: ' + cmsDataset + ' [MIT: ' + mitDataset + ']\n'
|
| 343 |
|
|
|
| 344 |
|
|
# --------------------------------------------------------------------------------------------------
|
| 345 |
|
|
# Deal with storage element area
|
| 346 |
|
|
# --------------------------------------------------------------------------------------------------
|
| 347 |
|
|
pMitDset = re.compile('XX-MITDATASET-XX')
|
| 348 |
|
|
pMitCfg = re.compile('XX-MITCFG-XX')
|
| 349 |
|
|
pMitVers = re.compile('XX-MITVERSION-XX')
|
| 350 |
|
|
# find the forseen storage place
|
| 351 |
|
|
crabFile = mitCfg + '/' + version + '/' + 'crab.cfg'
|
| 352 |
|
|
cmd = 'grep ^storage_element ' + crabFile
|
| 353 |
|
|
for file in os.popen(cmd).readlines(): # run command
|
| 354 |
|
|
line = file[:-1] # strip '\n'
|
| 355 |
|
|
# decode the storage element name
|
| 356 |
|
|
names = line.split("=") # splitting every '='
|
| 357 |
|
|
storageEle = names.pop()
|
| 358 |
|
|
storageEle = re.sub("\s", "",storageEle)
|
| 359 |
|
|
# Compile search and replacement sequences just for the path
|
| 360 |
|
|
cmd = 'grep ^storage_path ' + crabFile
|
| 361 |
|
|
for file in os.popen(cmd).readlines(): # run command
|
| 362 |
|
|
line = file[:-1] # strip '\n'
|
| 363 |
|
|
line = pMitDset.sub(mitDataset,line);
|
| 364 |
|
|
line = pMitCfg .sub(mitCfg, line);
|
| 365 |
|
|
line = pMitVers.sub(version, line);
|
| 366 |
|
|
# decode the storage directory name
|
| 367 |
|
|
names = line.split("=") # splitting every '='
|
| 368 |
|
|
names = names[1:]
|
| 369 |
|
|
storagePath = "=".join(names)
|
| 370 |
|
|
storagePath = re.sub("\s", "",storagePath)
|
| 371 |
|
|
storageUrl = 'srm://' + storageEle + ':8443' + storagePath
|
| 372 |
|
|
|
| 373 |
|
|
cmd = 'grep ^user_remote_dir ' + crabFile
|
| 374 |
|
|
for file in os.popen(cmd).readlines(): # run command
|
| 375 |
|
|
line = file[:-1] # strip '\n'
|
| 376 |
|
|
line = pMitDset.sub(mitDataset,line);
|
| 377 |
|
|
line = pMitCfg .sub(mitCfg, line);
|
| 378 |
|
|
line = pMitVers.sub(version, line);
|
| 379 |
|
|
# decode the storage directory name
|
| 380 |
|
|
names = line.split("=") # splitting every '='
|
| 381 |
|
|
names = names[1:]
|
| 382 |
|
|
userRemoteDir = "=".join(names)
|
| 383 |
|
|
userRemoteDir = re.sub("\s","",userRemoteDir)
|
| 384 |
|
|
userRemoteDir = re.sub("/XX-CRABID-XX","",userRemoteDir)
|
| 385 |
|
|
|
| 386 |
|
|
if userRemoteDir != '':
|
| 387 |
|
|
storagePath += userRemoteDir
|
| 388 |
|
|
storageUrl += userRemoteDir
|
| 389 |
|
|
|
| 390 |
|
|
if localStorageUrl != '':
|
| 391 |
|
|
storageEle = ''
|
| 392 |
|
|
storagePath = ''
|
| 393 |
|
|
storageUrl = localStorageUrl
|
| 394 |
|
|
|
| 395 |
|
|
print ' --> StorageUrl: ' + storageUrl
|
| 396 |
|
|
|
| 397 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 398 |
|
|
# create the local storage area
|
| 399 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 400 |
|
|
print ' Make local path: ' + localPath
|
| 401 |
|
|
localDir = localPath + '/' + mitCfg + '/' + version + '/' + mitDataset
|
| 402 |
|
|
mkd = 'mkdir -p ' + localDir
|
| 403 |
|
|
status = os.system(mkd)
|
| 404 |
|
|
|
| 405 |
|
|
if status != 0:
|
| 406 |
|
|
print ' ERROR - could not create local directory ' + localDir
|
| 407 |
|
|
sys.exit(1)
|
| 408 |
|
|
|
| 409 |
|
|
print ' --> LocalDir: ' + localDir
|
| 410 |
|
|
|
| 411 |
|
|
cmd = 'df --block-size=1 ' + localDir + ' | tr -s \' \' | tail -1'
|
| 412 |
|
|
for line in os.popen(cmd).readlines(): # run command
|
| 413 |
|
|
line = line.strip()
|
| 414 |
|
|
f = line.split(" ")
|
| 415 |
|
|
if line[0:0] == '/' or line[0:4] == 'fuse':
|
| 416 |
|
|
free = int(f[3])
|
| 417 |
|
|
else:
|
| 418 |
|
|
free = int(f[2])
|
| 419 |
|
|
|
| 420 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 421 |
|
|
# create a list af all files to be copied
|
| 422 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 423 |
|
|
cmd = ''
|
| 424 |
|
|
f = storagePath.split('=')
|
| 425 |
|
|
path = f.pop()
|
| 426 |
|
|
cmd = 'list ' + path + ' | grep root | sort ' + backward
|
| 427 |
|
|
|
| 428 |
|
|
##if storageEle == 'srm.cern.ch' or storageEle == 'srm-cms.cern.ch':
|
| 429 |
|
|
## cmd = 'rfdir ' + path + ' | grep root | tr -s \' \' | sort ' + backward
|
| 430 |
|
|
##else:
|
| 431 |
|
|
## cmd = 'list ' + path + ' | grep root | sort ' + backward
|
| 432 |
|
|
## #cmd = 'srmls ' + storageUrl + ' | grep root | sort ' + backward
|
| 433 |
|
|
|
| 434 |
|
|
if pattern != "":
|
| 435 |
|
|
cmd += ' | grep ' + pattern
|
| 436 |
|
|
|
| 437 |
|
|
print ' Find file: ' + cmd
|
| 438 |
|
|
cacheFile = '/tmp/.cache_' + mitDataset
|
| 439 |
|
|
allFileList = BuildFileList(cmd)
|
| 440 |
|
|
stagedFileList = BuildStagedFileList(storagePath,allFileList,cacheFile)
|
| 441 |
|
|
cacheStaged = CacheStagedFileList(cacheFile,storagePath,stagedFileList)
|
| 442 |
|
|
|
| 443 |
|
|
#cmd = 'find ' + localPath + '/' + mitCfg + '/' + version + '/' + mitDataset + \
|
| 444 |
|
|
# ' -maxdepth 1 -type f -printf "%s %f\n"'
|
| 445 |
|
|
print 'List: ' + cmd
|
| 446 |
|
|
cmd = 'list ' + localPath + '/' + mitCfg + '/' + version + '/' + mitDataset + ' | grep root'
|
| 447 |
|
|
doneFileList = BuildFileList(cmd)
|
| 448 |
|
|
|
| 449 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 450 |
|
|
# go through the lists: first check files are consistent, then copy the remaining files
|
| 451 |
|
|
#---------------------------------------------------------------------------------------------------
|
| 452 |
|
|
# initialize data volumes
|
| 453 |
|
|
b2G = 1.0/(1024.*1024.*1024)
|
| 454 |
|
|
nTotal = 0
|
| 455 |
|
|
totalDataVolume = 0
|
| 456 |
|
|
nDone = 0
|
| 457 |
|
|
doneDataVolume = 0
|
| 458 |
|
|
|
| 459 |
|
|
for file, size in allFileList.iteritems():
|
| 460 |
|
|
nTotal += 1
|
| 461 |
|
|
totalDataVolume += size
|
| 462 |
|
|
if (file in doneFileList) and (doneFileList[file] == size):
|
| 463 |
|
|
nDone += 1
|
| 464 |
|
|
doneDataVolume += size
|
| 465 |
|
|
|
| 466 |
|
|
print ' '
|
| 467 |
|
|
print ' Summary of data volume\n'
|
| 468 |
|
|
print ' --> number of files to copy: %8d (total: %d) '%(nTotal-nDone,nTotal)
|
| 469 |
|
|
print ' --> volume to copy [GB]: %8.2f (total: %.2f) '%(b2G*(totalDataVolume-doneDataVolume), \
|
| 470 |
|
|
b2G*totalDataVolume)
|
| 471 |
|
|
print ' --> free volume [GB]: %8.2f '%(b2G*free)
|
| 472 |
|
|
print ' '
|
| 473 |
|
|
|
| 474 |
|
|
if free*0.85 < (totalDataVolume-doneDataVolume):
|
| 475 |
|
|
print ' ERROR - probably no enough space on volume. See above (some safety assumed)!'
|
| 476 |
|
|
sys.exit(1)
|
| 477 |
|
|
|
| 478 |
|
|
for file, size in doneFileList.iteritems():
|
| 479 |
|
|
if file in allFileList:
|
| 480 |
|
|
#print ' --> file is done: ' + file
|
| 481 |
|
|
if allFileList[file] != size:
|
| 482 |
|
|
print ' ERROR - file sizes did not match: ' + file + \
|
| 483 |
|
|
' [ local: %10d, remote: %10d ]'%(size,allFileList[file])
|
| 484 |
|
|
sys.exit(1)
|
| 485 |
|
|
else:
|
| 486 |
|
|
print ' ERROR - file from done list is not in the all files list. File: ' + file
|
| 487 |
|
|
sys.exit(1)
|
| 488 |
|
|
|
| 489 |
|
|
totalSizeMb = 0.
|
| 490 |
|
|
totalTimeSc = 0.
|
| 491 |
|
|
for file, size in allFileList.iteritems():
|
| 492 |
|
|
if debug == 1:
|
| 493 |
|
|
print ' Debug:: ' + file + ' -> size %d'%size
|
| 494 |
|
|
|
| 495 |
|
|
totalDataVolume += size
|
| 496 |
|
|
if file in doneFileList:
|
| 497 |
|
|
print ' --> done, size match: %10d - %s'%(size,file)
|
| 498 |
|
|
doneDataVolume = +size
|
| 499 |
|
|
else:
|
| 500 |
|
|
if not InSkipList(file,skipList):
|
| 501 |
|
|
print ' --> copying file: %10d - %s (castor stat: %s)'% \
|
| 502 |
|
|
(size,file,stagedFileList[file])
|
| 503 |
|
|
if stagedFileList[file] == "STAGED" or forceCopy:
|
| 504 |
|
|
|
| 505 |
|
|
sizeMb = size/1024./1024.
|
| 506 |
|
|
deltaT = CopyFile(storageEle,storagePath,storageUrl,file,localDir)
|
| 507 |
|
|
if deltaT > 0:
|
| 508 |
|
|
print ' time required [sec]: %7d rate [MB/sec]: %9.3f'%\
|
| 509 |
|
|
(deltaT,sizeMb/deltaT)
|
| 510 |
|
|
else:
|
| 511 |
|
|
print ' time required [sec]: %7d rate [MB/sec]: ?'%(deltaT)
|
| 512 |
|
|
totalTimeSc += deltaT
|
| 513 |
|
|
totalSizeMb += sizeMb
|
| 514 |
|
|
else:
|
| 515 |
|
|
print ' skipping file: %s'%(stagedFileList[file])
|
| 516 |
|
|
StageFile(storagePath,storageUrl,file)
|
| 517 |
|
|
|
| 518 |
|
|
else:
|
| 519 |
|
|
print ' --> skipping file: %10d - %s'%(size,file)
|
| 520 |
|
|
|
| 521 |
|
|
print ''
|
| 522 |
|
|
if totalTimeSc > 0:
|
| 523 |
|
|
print ' Performance: volume copied [GB] %9.3f; time [sec] %9d; -> rate [MB/sec] %9.3f'%\
|
| 524 |
|
|
(totalSizeMb/1024.,totalTimeSc,totalSizeMb/totalTimeSc)
|
| 525 |
|
|
else:
|
| 526 |
|
|
print ' Performance: volume copied [GB] %9.3f; time [sec] %9d; -> rate [MB/sec] ?'%\
|
| 527 |
|
|
(totalSizeMb/1024.,totalTimeSc)
|
| 528 |
|
|
print ''
|
