forked from kernelci/lava-ci-staging
-
Notifications
You must be signed in to change notification settings - Fork 0
/
lava-job-runner.py
executable file
·327 lines (299 loc) · 14.8 KB
/
lava-job-runner.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
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
#!/usr/bin/python
#
# Copyright (C) 2014, 2015, 2016 Linaro Limited
# Author: Tyler Baker <[email protected]>
# Author: Anders Roxell <[email protected]>
#
# Copyright (C) 2016 Baylibre SAS
# Author: Marc Titinger <[email protected]>
#
# This module is free software; you can redistribute it and/or modify it under
# the terms of the GNU Lesser General Public License as published by the Free
# Software Foundation; either version 2.1 of the License, or (at your option)
# any later version.
#
# This library is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
# details.
#
# You should have received a copy of the GNU Lesser General Public License
# along with this library; if not, write to the Free Software Foundation, Inc.,
# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
# <variable> = required
# [variable] = optional
# Usage ./lava-job-runner.py <username> <token> <lava_server_url> [job_repo] [bundle_stream]
import os
import xmlrpclib
import json
import subprocess
import fnmatch
import time
import re
import argparse
import httplib
from lib import utils
from lib import configuration
job_map = {}
def poll_jobs(connection, timeout):
run = True
submitted_jobs = {}
finished_jobs = {}
start_time = time.time()
for job in job_map:
if job_map[job] is not None:
# Push
for job_id in job_map[job]:
job_name = os.path.basename(job) + '-' + job_id
submitted_jobs[job_name] = job_id
while run:
if timeout < (time.time() - start_time) and timeout != -1:
print 'Polling timeout reached, collecting completed results'
run = False
break
if not submitted_jobs:
run = False
break
for job in submitted_jobs:
try:
status = connection.scheduler.job_status(submitted_jobs[job])
if status['job_status'] == 'Complete':
print 'job-id-' + str(job) + ' : pass'
# Pop
if status['bundle_sha1']:
finished_jobs[submitted_jobs[job]] = {'result': 'PASS', 'bundle': status['bundle_sha1']}
else:
finished_jobs[submitted_jobs[job]] = {'result': 'PASS', 'bundle': None}
submitted_jobs.pop(job, None)
break
elif status['job_status'] == 'Incomplete' or status['job_status'] == 'Canceled' or status['job_status'] == 'Canceling':
print 'job-id-' + str(job) + ' : fail'
# Pop
if status['bundle_sha1']:
finished_jobs[submitted_jobs[job]] = {'result': 'FAIL', 'bundle': status['bundle_sha1']}
else:
finished_jobs[submitted_jobs[job]] = {'result': 'FAIL', 'bundle': None}
submitted_jobs.pop(job, None)
break
else:
print str(submitted_jobs[job]) + ' - ' + str(status['job_status'])
time.sleep(10)
except (xmlrpclib.ProtocolError, xmlrpclib.Fault, IOError, httplib.ResponseNotReady) as e:
print "POLLING ERROR!"
print e
continue
return finished_jobs
def submit_jobs(connection, server, bundle_stream=None):
online_devices, offline_devices = gather_devices(connection)
online_device_types, offline_device_types = gather_device_types(connection)
print "Submitting Jobs to Server..."
for job in job_map:
try:
with open(job, 'rb') as stream:
job_data = stream.read()
# Injection
if bundle_stream is not None:
job_data = re.sub('LAVA_SERVER', server, job_data)
job_data = re.sub('BUNDLE_STREAM', bundle_stream, job_data)
job_info = json.loads(job_data)
# Check if request device(s) are available
if 'target' in job_info:
if job_info['target'] in offline_devices:
print "%s is OFFLINE skipping submission" % job_info['target']
print os.path.basename(job) + ': skip'
elif job_info['target'] in online_devices:
pass
jobs = connection.scheduler.submit_job(job_data)
if isinstance(jobs, int):
jobs = str(jobs).split()
job_map[job] = jobs
else:
print "No target available on server, skipping..."
print os.path.basename(job) + ' : skip'
elif 'device_type' in job_info:
if job_info['device_type'] in offline_device_types:
print "All device types: %s are OFFLINE, skipping..." % job_info['device_type']
print os.path.basename(job) + ' : skip'
elif job_info['device_type'] in online_device_types:
pass
jobs = connection.scheduler.submit_job(job_data)
if isinstance(jobs, int):
jobs = str(jobs).split()
job_map[job] = jobs
else:
print "No device-type available on server, skipping..."
print os.path.basename(job) + ' : skip'
elif 'device_group' in job_info:
print "Multinode Job Detected! Checking if required devices are available..."
multinode_online_device_types = online_device_types
server_has_required_devices = True
for groups in job_info['device_group']:
if groups['device_type'] in offline_device_types:
print "All device types: %s are OFFLINE, skipping..." % groups['device_type']
server_has_required_devices = False
print os.path.basename(job) + ' : skip'
break
elif groups['device_type'] in online_device_types:
if groups['count'] > multinode_online_device_types[groups['device_type']]:
print "Server does not have enough online devices to submit job!"
print os.path.basename(job) + ' : skip'
server_has_required_devices = False
break
elif groups['count'] <= multinode_online_device_types[groups['device_type']]:
print "Server has enough devices for this group!"
multinode_online_device_types[groups['device_type']] = multinode_online_device_types[groups['device_type']] - groups['count']
else:
print "Should never get here!"
print os.path.basename(job) + ' : skip'
server_has_required_devices = False
break
else:
print "No device-type available on server, skipping..."
print os.path.basename(job) + ' : skip'
if server_has_required_devices:
print "Submitting Multinode Job!"
jobs = connection.scheduler.submit_job(job_data)
if isinstance(jobs, int):
jobs = str(jobs).split()
job_map[job] = jobs
elif 'vm_group' in job_info:
print "VMGroup Job Detected! Checking if required devices are available..."
vmgroup_online_device_types = online_device_types
server_has_required_devices = True
for groups in job_info['vm_group']:
if job_info['vm_group'][groups]:
if 'device_type' in job_info['vm_group'][groups]:
if job_info['vm_group'][groups]['device_type'] in offline_device_types:
print "All device types: %s are OFFLINE, skipping..." % groups['device_type']
server_has_required_devices = False
print os.path.basename(job) + ' : skip'
break
elif job_info['vm_group'][groups]['device_type'] in online_device_types:
if 'count' in job_info['vm_group'][groups]:
count = job_info['vm_group'][groups]['device_type']['count']
else:
count = 1
if count > vmgroup_online_device_types[job_info['vm_group'][groups]['device_type']]:
print "Server does not have enough online devices to submit job!"
print os.path.basename(job) + ' : skip'
server_has_required_devices = False
break
elif count <= vmgroup_online_device_types[job_info['vm_group'][groups]['device_type']]:
print "Server has enough devices for this group!"
vmgroup_online_device_types[job_info['vm_group'][groups]['device_type']] = vmgroup_online_device_types[job_info['vm_group'][groups]['device_type']] - count
else:
print "Should never get here!"
print os.path.basename(job) + ' : skip'
server_has_required_devices = False
break
else:
print "No device-type available on server, skipping..."
print os.path.basename(job) + ' : skip'
if server_has_required_devices:
print "Submitting VMGroups Job!"
jobs = connection.scheduler.submit_job(job_data)
if isinstance(jobs, int):
jobs = str(jobs).split()
job_map[job] = jobs
else:
print "Should never get here"
print os.path.basename(job) + ' : skip'
except (xmlrpclib.ProtocolError, xmlrpclib.Fault, IOError, ValueError) as e:
print "JSON VALIDATION ERROR!"
print job
print e
continue
def load_jobs(top):
for root, dirnames, filenames in os.walk(top):
for filename in fnmatch.filter(filenames, '*.json'):
job_map[os.path.join(root, filename)] = None
def retrieve_jobs(jobs):
cmd = 'git clone %s' % jobs
try:
print "Cloning LAVA Jobs..."
subprocess.check_output(cmd, shell=True)
print "Clone Successful!"
print "clone-jobs : pass"
except subprocess.CalledProcessError as e:
print "ERROR!"
print "Unable to clone %s" % jobs
print "clone-jobs : fail"
exit(1)
def gather_devices(connection):
online_devices = {}
offline_devices = {}
print "Gathering Devices..."
all_devices = connection.scheduler.all_devices()
for device in all_devices:
if device[2] == 'offline':
offline_devices[device[0]] = 1
else:
online_devices[device[0]] = 1
print "Gathered Devices Successfully!"
return online_devices, offline_devices
def gather_device_types(connection):
online_device_types = {}
offline_device_types = {}
print "Gathering Device Types..."
all_device_types = connection.scheduler.all_device_types()
for device_type in all_device_types:
# Only use dictionary data structures
if isinstance(device_type, dict):
# Retired
if device_type['idle'] == 0 and device_type['busy'] == 0 and device_type['offline'] == 0:
offline_device_types[device_type['name']] = 0
# Running
elif device_type['idle'] > 0 or device_type['busy'] > 0:
online_device_types[device_type['name']] = device_type['idle'] + device_type['busy']
# Offline
else:
offline_device_types[device_type['name']] = device_type['offline']
print "Gathered Device Types Successfully!"
return online_device_types, offline_device_types
def main(args):
config = configuration.get_config(args)
url = utils.validate_input(config.get("username"), config.get("token"), config.get("server"))
connection = utils.connect(url)
if config.get("repo"):
retrieve_jobs(config.get("repo"))
if config.get("jobs"):
load_jobs(config.get("jobs"))
print "Loading jobs from top folder " + str(config.get("jobs"))
else:
load_jobs(os.getcwd())
start_time = time.time()
bundle_stream = None
if config.get("stream"):
bundle_stream = config.get("stream")
submit_jobs(connection, config.get("server"), bundle_stream=bundle_stream)
if config.get("poll"):
jobs = poll_jobs(connection, config.get("timeout"))
end_time = time.time()
if config.get("bisect"):
for job_id in jobs:
if 'result' in jobs[job_id]:
if jobs[job_id]['result'] == 'FAIL':
exit(1)
jobs['duration'] = end_time - start_time
jobs['username'] = config.get("username")
jobs['token'] = config.get("token")
jobs['server'] = config.get("server")
results_directory = os.getcwd() + '/results'
utils.mkdir(results_directory)
utils.write_json(config.get("poll"), results_directory, jobs)
exit(0)
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument("--config", help="configuration for the LAVA server")
parser.add_argument("--section", default="default", help="section in the LAVA config file")
parser.add_argument("--jobs", help="absolute path to top jobs folder (default scans the whole cwd)")
parser.add_argument("--username", help="username for the LAVA server")
parser.add_argument("--token", help="token for LAVA server api")
parser.add_argument("--server", help="server url for LAVA server")
parser.add_argument("--stream", help="bundle stream for LAVA server")
parser.add_argument("--repo", help="git repo for LAVA jobs")
parser.add_argument("--poll", help="poll the submitted LAVA jobs, dumps info into specified json")
parser.add_argument("--timeout", type=int, default=-1, help="polling timeout in seconds. default is -1.")
parser.add_argument('--bisect', help="bisection mode, returns 1 on any job failures", action='store_true')
args = vars(parser.parse_args())
main(args)