/
submit.py
434 lines (393 loc) · 12.3 KB
/
submit.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
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
from __future__ import print_function
import sys
import json
import logging
from optparse import OptionParser, OptionGroup
from datetime import datetime
from pyspark import SparkContext, SparkConf
from pyspark.streaming import StreamingContext
from pyspark.streaming.kafka import KafkaUtils
import adapters as AllAdapters
###############
# Config class
###############
class Config(object):
# spark setting
PROJECT_NAME = 'KSE'
CHECK_POINT = False
CHECK_POINT_PATH = None
CHECK_POINT_INTERVAL = None
TTL = None
# spark streaming
INTERVAL = 1
REMEMBER = None
WINDOW = 10
# es setting
ES_NODES = None
# app setting
mode = None
logpath = None
hostname = None
zkQuorum = None
port = None
topic = None
# options
usage = 'Usage: %prog <-m [mode]> [options]'
mode_list = ['network', 'kafka']
parser_system_options = [
# system
{
"short": "-e",
"long": "--esnode",
"action": "store",
"dest": "es",
"default": None,
"help": "elasticsearch node address",
"type": "string"
},
{
"short": "-l",
"long": "--log",
"action": "store",
"dest": "log",
"default": '/data/logs/KSE/',
"help": "log path(endwith '/')",
"type": "string"
},
{
"short": "--checkpoint",
"action": "store_true",
"dest": "checkpoint",
"default": False,
"help": "enable spark checkpoint",
"type": None
},
{
"short": "--checkpoint-path",
"action": "store",
"dest": "checkpointpath",
"default": '/data/checkpoint/KSE/',
"help": "spark checkpoint directory(endwith '/')",
"type": "string"
},
{
"short": "--checkpoint-interval",
"action": "store",
"dest": "checkpointinterval",
"default": 10,
"help": "spark checkpoint interval(default 10 seconds)",
"type": "int"
},
{
"short": "--spark-cleaner-ttl",
"action": "store",
"dest": "ttl",
"default": "300",
"help": "how long will data remain in memory(default 300 seconds)",
"type": "string"
},
{
"short": "--ssc-remember",
"action": "store",
"dest": "sscremember",
"default": 240,
"help": "how long will spark-streaming context remember \
input data or persisted rdds.(default 240 seconds)",
"type": "int"
},
{
"short": "--ssc-window",
"action": "store",
"dest": "sscwindow",
"default": 10,
"help": "how long will spark-streaming context window \
be. Note this must be multiples of the batch interval \
of the value --ssc-interval.(default 10 seconds)",
"type": "int"
},
{
"short": "--ssc-interval",
"action": "store",
"dest": "interval",
"default": 2,
"help": "how long is the interval of spark-streaming \
getting input data.(default 2 seconds)",
"type": "int"
},
]
parser_options = [
# mode
{
"short": "-m",
"long": "--mode",
"action": "store",
"dest": "mode",
"default": None,
"help": "KSE mode",
"type": "string"
},
{
"short": "--hostname",
"action": "store",
"dest": "hostname",
"default": None,
"help": "network mode hostname",
"type": "string"
},
{
"short": "--port",
"action": "store",
"dest": "port",
"default": None,
"help": "network mode port",
"type": "string"
},
{
"short": "--zkquorum",
"action": "store",
"dest": "zkquorum",
"default": None,
"help": "kafka mode zkQuorum",
"type": "string"
},
{
"short": "--topic",
"action": "store",
"dest": "topic",
"default": None,
"help": "kafka mode topic",
"type": "string"
}
]
def __init__(self, args):
self.verify(args)
def verify(self, args):
parser = OptionParser(usage=self.usage)
for opt in self.parser_options:
parser.add_option(
opt['short'],
opt.get('long', None),
action=opt['action'],
dest=opt['dest'],
default=opt['default'],
help=opt['help'],
type=opt['type'])
group = OptionGroup(
parser, "System Options",
"Caution: These options usually use default values.")
for opt in self.parser_system_options:
group.add_option(
opt['short'],
opt.get('long', None),
action=opt['action'],
dest=opt['dest'],
default=opt['default'],
help=opt['help'],
type=opt['type'])
parser.add_option_group(group)
(options, args) = parser.parse_args(args)
self.TTL = options.ttl
self.CHECK_POINT = options.checkpoint
self.CHECK_POINT_PATH = options.checkpointpath
self.CHECK_POINT_INTERVAL = options.checkpointinterval
self.INTERVAL = options.interval
self.REMEMBER = options.sscremember
self.WINDOW = options.sscwindow
self.ES_NODES = options.es
self.mode = options.mode
self.logpath = options.log
self.hostname = options.hostname
self.zkQuorum = options.zkquorum
self.port = options.port
self.topic = options.topic
if self.mode not in self.mode_list:
parser.error("-m mast be one of [network, kafka]")
exit(-1)
if self.ES_NODES is None:
parser.error("-e should not be null value")
exit(-1)
if self.mode == 'network' and (
self.hostname is None or self.port is None):
parser.error("--hostname and --port should not be null value")
exit(-1)
elif self.mode == 'kafka' and (
self.zkQuorum is None or self.topic is None):
parser.error("--zkquorum and --topic should not be null value")
exit(-1)
def getSparkConf(self):
conf = SparkConf()
conf.setAppName(self.PROJECT_NAME)
conf.set(
"spark.serializer", "org.apache.spark.serializer.KryoSerializer")
conf.set("spark.cleaner.ttl", self.TTL)
# es
conf.set("es.index.auto.create", "true")
conf.set("es.nodes", self.ES_NODES)
return conf
def getReader(self, ssc):
if self.mode == 'network':
lines = self.networkReader(ssc)
elif self.mode == 'kafka':
lines = self.kafkaReader(ssc)
return lines
def networkReader(self, ssc):
lines = ssc.socketTextStream(self.hostname, int(self.port))
if self.CHECK_POINT:
lines.checkpoint(self.CHECK_POINT_INTERVAL)
return lines
def kafkaReader(self, ssc):
kvs = KafkaUtils.createStream(
ssc,
self.zkQuorum,
"spark-streaming-log",
{self.topic: 1}
)
lines = kvs.map(lambda x: x[1])
if self.CHECK_POINT:
lines.checkpoint(self.CHECK_POINT_INTERVAL)
return lines
###############
# Json Decoder
###############
class logDecoder(json.JSONDecoder):
def __init__(self):
json.JSONDecoder.__init__(self, object_hook=self.dict_to_object)
def dict_to_object(self, d):
for key in d:
if isinstance(d[key], list):
d[key] = tuple(d[key])
return d
###############
# ES
###############
class EsClient(object):
# es setting
ES_NODES = None
def __init__(self, nodes):
self.ES_NODES = nodes
def getESRDD(self, index=None, doc_type=None, query=None):
if index is None or doc_type is None or self.ES_NODES is None:
return None
conf = {
"es.resource": "%s/%s" % (index, doc_type),
"es.nodes": self.ES_NODES,
}
if query is not None:
conf["es.query"] = json.dumps(query)
try:
es_rdd = self.SC.newAPIHadoopRDD(
inputFormatClass="org.elasticsearch.hadoop.mr.EsInputFormat",
keyClass="org.apache.hadoop.io.NullWritable",
valueClass="org.elasticsearch.hadoop.mr.LinkedMapWritable",
conf=conf)
return es_rdd
except Exception as e:
log(e, level='ERROR')
return None
def saveTOES(self, rdd, index=None, doc_type=None):
if index is None or doc_type is None or self.ES_NODES is None:
return False
conf = {
"es.resource": "%s/%s" % (index, doc_type),
"es.nodes": self.ES_NODES,
}
rdd.saveAsNewAPIHadoopFile(
path='-',
outputFormatClass="org.elasticsearch.hadoop.mr.EsOutputFormat",
keyClass="org.apache.hadoop.io.NullWritable",
valueClass="org.elasticsearch.hadoop.mr.LinkedMapWritable",
conf=conf)
return True
@classmethod
def fastSaveTOES(cls, rdd, resource, es):
conf = {
"es.resource": resource,
"es.nodes": es,
}
rdd.saveAsNewAPIHadoopFile(
path='-',
outputFormatClass="org.elasticsearch.hadoop.mr.EsOutputFormat",
keyClass="org.apache.hadoop.io.NullWritable",
valueClass="org.elasticsearch.hadoop.mr.LinkedMapWritable",
conf=conf)
###############
# Output
###############
def output(msg):
logger = logging.getLogger('output')
logger.info(msg)
def stdout(msg):
print(msg)
###############
# Deal logic
###############
def linegrok(line):
ret = []
for adapter in AllAdapters.AdapterHelper.getAdapters():
ext = adapter.act(line)
if len(ext) > 0:
ext = [(adapter.es_doc_type, ('key', i)) for i in ext]
ret.extend(ext)
return ret
def streamingOutput(rdd, es):
if rdd.isEmpty():
return
rdd.persist()
for res, key in AllAdapters.AdapterHelper.getAdaptersEsResource():
group = rdd.filter(lambda i: i[0] == key).map(lambda i: i[1])
if not group.isEmpty():
EsClient.fastSaveTOES(group, resource=res, es=es)
count = rdd.countByKey().items()
for i in count:
output("[%s] %d" % (i[0], i[1]))
rdd.unpersist()
def deal(lines, conf):
es = conf.ES_NODES
lines = lines.flatMap(linegrok)
lines.foreachRDD(lambda rdd: streamingOutput(rdd, es=es))
###############
# Main logic
###############
def main(conf):
ssc = None
if conf.CHECK_POINT:
ssc = StreamingContext.getOrCreate(
conf.CHECK_POINT_PATH,
lambda: createContext(conf))
else:
ssc = createContext(conf)
ssc.start()
ssc.awaitTermination()
return ssc
def createContext(conf):
spConf = conf.getSparkConf()
sc = SparkContext(conf=spConf)
ssc = StreamingContext(sc, conf.INTERVAL)
ssc.remember(conf.REMEMBER)
# get reader
lines = conf.getReader(ssc)
# use window
lines = lines.window(conf.WINDOW, conf.WINDOW)
lines = lines.map(lambda line: jsonDecode(line))
deal(lines, conf)
return ssc
def jsonDecode(line):
try:
return logDecoder().decode(line)
except Exception as e:
return {}
if __name__ == "__main__":
conf = Config(sys.argv)
# logger
logger = logging.getLogger('output')
d = datetime.now().strftime('%Y%m%d%H%M%S')
f = logging.FileHandler("%s%s.log" % (conf.logpath, d))
logger.addHandler(f)
formatter = logging.Formatter(
fmt="[%(levelname)s][%(asctime)s]%(message)s",
datefmt="%Y-%m-%d %H:%M:%S"
)
f.setFormatter(formatter)
logger.setLevel(logging.INFO)
# spark
main(conf)