forked from CMSCompOps/WmAgentScripts
/
resubmit.py
executable file
·238 lines (214 loc) · 8.42 KB
/
resubmit.py
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
#!/usr/bin/env python
"""
This script clones a given workflow
Usage:
python resubmit.py [options] WORKFLOW_NAME [USER GROUP]
Options:
-b --backfill, creates a clone
-v --verbose, prints schemas and responses
USER: the user for creating the clone, if empty it will
use the OS user running the script
GROUP: the group for creating the clone, if empty it will
use 'DATAOPS' by default
This script depends on WMCore code, so WMAgent environment
,libraries and voms proxy need to be loaded before running it.
"""
import pprint
import os, datetime, pwd
import sys
import urllib
import httplib
import re
import json
from optparse import OptionParser
from pprint import pprint
try:
import changePriorityWorkflow
import reqMgrClient
from WMCore.WMSpec.WMWorkload import WMWorkloadHelper
from WMCore.Wrappers import JsonWrapper
except:
print "WMCore libraries not loaded, run the following command:"
print "source /data/srv/wmagent/current/apps/wmagent/etc/profile.d/init.sh"
sys.exit(0)
reqmgrCouchURL = "https://cmsweb.cern.ch/couchdb/reqmgr_workload_cache"
def modifySchema(helper, user, group, backfill=False):
"""
Adapts schema to right parameters.
If the original workflow points to DBS2, DBS3 URL is fixed instead.
if backfill is True, modifies RequestString, ProcessingString, AcquisitionEra
and Campaign to say Backfill, and restarts requestDate.
"""
result = {}
for (key, value) in helper.data.request.schema.dictionary_().items():
#previous versions of tags
if key == 'ProcConfigCacheID':
result['ConfigCacheID'] = value
elif key == 'RequestSizeEvents':
result['RequestSizeEvents'] = value
#requestor info
elif key == 'Requestor':
result['Requestor'] = user
elif key == 'Group':
result['Group'] = group
#if emtpy
elif key in ["RunWhitelist", "RunBlacklist", "BlockWhitelist", "BlockBlacklist"] and not value:
result[key]=[]
#replace old DBS2 URL
elif value == "http://cmsdbsprod.cern.ch/cms_dbs_prod_global/servlet/DBSServlet":
result[key] = 'https://cmsweb.cern.ch/dbs/prod/global/DBSReader'
#copy the right LFN base
elif key == 'MergedLFNBase':
result['MergedLFNBase'] = helper.getMergedLFNBase()
# convert LumiList to dict
#elif key == 'LumiList':
# result['LumiList'] = JsonWrapper.loads(value)
# result['LumiList'] = eval(value)
#TODO deleting timeout so they will move to running-close as soon as they can
#elif key == 'OpenRunningTimeout':
#delete entry
# continue
#skip empty entries
elif value != None:
result[key] = value
elif not value:
continue
#check MonteCarlo
if result['RequestType']=='MonteCarlo':
#check assigning parameters
#seek for events per job on helper
try:
splitting = helper.listJobSplittingParametersByTask()
except AttributeError:
splitting = {}
eventsPerJob = 120000
eventsPerLumi = 100000
for k, v in splitting.items():
#print k,":",v
if k.endswith('/Production'):
if 'events_per_job' in v:
eventsPerJob = v['events_per_job']
elif 'events_per_lumi' in v:
eventsPerLumi = v['events_per_lumi']
result['EventsPerJob'] = eventsPerJob
#result['EventsPerLumi'] = eventsPerLumi
#check MonteCarloFromGen
elif result['RequestType']=='MonteCarloFromGEN':
#seek for lumis per job on helper
splitting = helper.listJobSplittingParametersByTask()
lumisPerJob = 300
for k, v in splitting.items():
if k.endswith('/Production'):
if 'lumis_per_job' in v:
lumisPerJob = v['lumis_per_job']
result['LumisPerJob'] = lumisPerJob
#Algorithm = lumi based?
result["SplittingAlgo"] = "LumiBased"
#Merged LFN
if 'MergedLFNBase' not in result:
result['MergedLFNBase'] = helper.getMergedLFNBase()
#update information from reqMgr
# Add AcquisitionEra, ProcessingString and increase ProcessingVersion by 1
result["ProcessingString"] = helper.getProcessingString()
result["AcquisitionEra"] = helper.getAcquisitionEra()
#try to parse processing version as an integer, if don't, assign 2.
try:
result["ProcessingVersion"] = int(helper.getProcessingVersion()) + 1
except ValueError:
result["ProcessingVersion"] = 2
#modify for backfill
if backfill:
#Modify ProcessingString, AcquisitionEra, Campaign and Request string (if they don't
#have the word 'backfill' in it
result["ProcessingString"] = "BACKFILL"
if "backfill" not in result["AcquisitionEra"].lower():
result["AcquisitionEra"] = helper.getAcquisitionEra()+"Backfill"
if "backfill" not in result["Campaign"].lower():
result["Campaign"] = result["Campaign"]+"-Backfill"
if "backfill" not in result["RequestString"].lower():
#Word backfill in the middle of the request strin
parts = result["RequestString"].split('-')
result["RequestString"] = '-'.join(parts[:2]+["Backfill"]+parts[2:])
if "PrepID" in result:
#delete entry
del result["PrepID"]
#reset the request date
now = datetime.datetime.utcnow()
result["RequestDate"] = [now.year, now.month, now.day, now.hour, now.minute]
return result
def cloneWorkflow(workflow, user, group, verbose=False, backfill=False):
"""
clones a workflow
"""
# Get info about the workflow to be cloned
helper = reqMgrClient.retrieveSchema(workflow)
# get info from reqMgr
schema = modifySchema(helper, user, group, backfill)
if verbose:
pprint(schema)
print 'Submitting workflow'
# Sumbit cloned workflow to ReqMgr
response = reqMgrClient.submitWorkflow(url,schema)
if verbose:
print "RESPONSE", response
#find the workflow name in response
m = re.search("details\/(.*)\'",response)
if m:
newWorkflow = m.group(1)
print 'Cloned workflow: '+newWorkflow
if verbose:
print response
print 'Approving request response:'
#TODO only for debug
#response = reqMgrClient.setWorkflowSplitting(url, schema)
#print "RESPONSE", response
#schema['requestName'] = requestName
#schema['splittingTask'] = '/%s/%s' % (requestName, taskName)
#schema['splittingAlgo'] = splittingAlgo
# Move the request to Assignment-approved
data = reqMgrClient.setWorkflowApproved(url, newWorkflow)
if verbose:
print data
#return the name of new workflow
return newWorkflow
else:
if verbose:
print response
else:
print "Couldn't clone the workflow."
return None
"""
__Main__
"""
url = 'cmsweb.cern.ch'
def main():
#Create option parser
usage = "\n python %prog [options] WORKFLOW_NAME [USER GROUP]\n"\
"USER: the user for creating the clone, if empty it will\n"\
" use the OS user running the script\n"\
"GROUP: the group for creating the clone, if empty it will\n"\
" use 'DATAOPS' by default"
parser = OptionParser(usage=usage)
parser.add_option("-b","--backfill",action="store_true", dest="backfill", default=False,
help="Creates a clone for backfill test purposes.")
parser.add_option("-v","--verbose",action="store_true", dest="verbose", default=False,
help="Prints all query information.")
(options, args) = parser.parse_args()
# Check the arguments, get info from them
if len(args) == 3:
user = args[2]
group = args[3]
elif len(args) == 1:
#get os username by default
uinfo = pwd.getpwuid(os.getuid())
user = uinfo.pw_name
#group by default DATAOPS
group = 'DATAOPS'
else:
parser.error("Provide the workflow name and/or user name and group.")
sys.exit(1)
workflow = args[0]
cloneWorkflow(workflow, user, group, options.verbose, options.backfill)
sys.exit(0)
if __name__ == "__main__":
main()