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
|
###############################################################################
# Copyright (c) 2017 Koren Lev (Cisco Systems), Yaron Yogev (Cisco Systems) #
# and others #
# #
# All rights reserved. This program and the accompanying materials #
# are made available under the terms of the Apache License, Version 2.0 #
# which accompanies this distribution, and is available at #
# http://www.apache.org/licenses/LICENSE-2.0 #
###############################################################################
import argparse
import signal
import time
from multiprocessing import Process, Manager as SharedManager
import os
from discover.events.listeners.default_listener import DefaultListener
from discover.events.listeners.listener_base import ListenerBase
from discover.manager import Manager
from utils.constants import OperationalStatus, EnvironmentFeatures
from utils.inventory_mgr import InventoryMgr
from utils.logging.file_logger import FileLogger
from utils.mongo_access import MongoAccess
class EventManager(Manager):
# After EventManager receives a SIGTERM,
# it will try to terminate all listeners.
# After this delay, a SIGKILL will be sent
# to each listener that is still alive.
SIGKILL_DELAY = 5 # in seconds
DEFAULTS = {
"mongo_config": "",
"collection": "environments_config",
"inventory": "inventory",
"interval": 5,
"loglevel": "INFO"
}
LISTENERS = {
'Mirantis-6.0': DefaultListener,
'Mirantis-7.0': DefaultListener,
'Mirantis-8.0': DefaultListener,
'RDO-Mitaka': DefaultListener,
'RDO-Liberty': DefaultListener,
}
def __init__(self):
self.args = self.get_args()
super().__init__(log_directory=self.args.log_directory,
mongo_config_file=self.args.mongo_config)
self.db_client = None
self.interval = None
self.processes = []
@staticmethod
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument("-m", "--mongo_config", nargs="?", type=str,
default=EventManager.DEFAULTS["mongo_config"],
help="Name of config file with MongoDB server access details")
parser.add_argument("-c", "--collection", nargs="?", type=str,
default=EventManager.DEFAULTS["collection"],
help="Environments collection to read from "
"(default: '{}')"
.format(EventManager.DEFAULTS["collection"]))
parser.add_argument("-y", "--inventory", nargs="?", type=str,
default=EventManager.DEFAULTS["inventory"],
help="name of inventory collection "
"(default: '{}')"
.format(EventManager.DEFAULTS["inventory"]))
parser.add_argument("-i", "--interval", nargs="?", type=float,
default=EventManager.DEFAULTS["interval"],
help="Interval between collection polls "
"(must be more than {} seconds. Default: {})"
.format(EventManager.MIN_INTERVAL,
EventManager.DEFAULTS["interval"]))
parser.add_argument("-l", "--loglevel", nargs="?", type=str,
default=EventManager.DEFAULTS["loglevel"],
help="Logging level \n(default: '{}')"
.format(EventManager.DEFAULTS["loglevel"]))
parser.add_argument("-d", "--log_directory", nargs="?", type=str,
default=FileLogger.LOG_DIRECTORY,
help="File logger directory \n(default: '{}')"
.format(FileLogger.LOG_DIRECTORY))
args = parser.parse_args()
return args
def configure(self):
self.db_client = MongoAccess()
self.inv = InventoryMgr()
self.inv.set_collections(self.args.inventory)
self.collection = self.db_client.db[self.args.collection]
self.interval = max(self.MIN_INTERVAL, self.args.interval)
self.log.set_loglevel(self.args.loglevel)
self.log.info("Started EventManager with following configuration:\n"
"Mongo config file path: {0}\n"
"Collection: {1}\n"
"Polling interval: {2} second(s)"
.format(self.args.mongo_config, self.collection.name, self.interval))
def get_listener(self, env: str):
env_config = self.inv.get_env_config(env)
return self.LISTENERS.get(env_config.get('distribution'))
def listen_to_events(self, listener: ListenerBase, env_name: str, process_vars: dict):
listener.listen({
'env': env_name,
'mongo_config': self.args.mongo_config,
'inventory': self.args.inventory,
'loglevel': self.args.loglevel,
'environments_collection': self.args.collection,
'process_vars': process_vars
})
def _get_alive_processes(self):
return [p for p in self.processes
if p['process'].is_alive()]
# Get all processes that should be terminated
def _get_stuck_processes(self, stopped_processes: list):
return [p for p in self._get_alive_processes()
if p.get("name") in map(lambda p: p.get("name"), stopped_processes)]
# Give processes time to finish and kill them if they are stuck
def _kill_stuck_processes(self, process_list: list):
if self._get_stuck_processes(process_list):
time.sleep(self.SIGKILL_DELAY)
for process in self._get_stuck_processes(process_list):
self.log.info("Killing event listener '{0}'".format(process.get("name")))
os.kill(process.get("process").pid, signal.SIGKILL)
def _get_operational(self, process: dict) -> OperationalStatus:
try:
return process.get("vars", {})\
.get("operational")
except:
self.log.error("Event listener '{0}' is unreachable".format(process.get("name")))
return OperationalStatus.STOPPED
def _update_operational_status(self, status: OperationalStatus):
self.collection.update_many(
{"name": {"$in": [process.get("name")
for process
in self.processes
if self._get_operational(process) == status]}},
{"$set": {"operational": status.value}}
)
def update_operational_statuses(self):
self._update_operational_status(OperationalStatus.RUNNING)
self._update_operational_status(OperationalStatus.ERROR)
self._update_operational_status(OperationalStatus.STOPPED)
def cleanup_processes(self):
# Query for envs that are no longer eligible for listening
# (scanned == false and/or listen == false)
dropped_envs = [env['name']
for env
in self.collection
.find(filter={'$or': [{'scanned': False},
{'listen': False}]},
projection=['name'])]
live_processes = []
stopped_processes = []
# Drop already terminated processes
# and for all others perform filtering
for process in self._get_alive_processes():
# If env no longer qualifies for listening,
# stop the listener.
# Otherwise, keep the process
if process['name'] in dropped_envs:
self.log.info("Stopping event listener '{0}'".format(process.get("name")))
process['process'].terminate()
stopped_processes.append(process)
else:
live_processes.append(process)
self._kill_stuck_processes(stopped_processes)
# Update all 'operational' statuses
# for processes stopped on the previous step
self.collection.update_many(
{"name": {"$in": [process.get("name")
for process
in stopped_processes]}},
{"$set": {"operational": OperationalStatus.STOPPED.value}}
)
# Keep the living processes
self.processes = live_processes
def do_action(self):
try:
while True:
# Update "operational" field in db before removing dead processes
# so that we keep last statuses of env listeners before they were terminated
self.update_operational_statuses()
# Perform a cleanup that filters out all processes
# that are no longer eligible for listening
self.cleanup_processes()
envs = self.collection.find({'scanned': True, 'listen': True})
# Iterate over environments that don't have an event listener attached
for env in filter(lambda e: e['name'] not in
map(lambda process: process["name"], self.processes),
envs):
env_name = env['name']
if not self.inv.is_feature_supported(env_name, EnvironmentFeatures.LISTENING):
self.log.error("Listening is not supported for env '{}'".format(env_name))
self.collection.update({"name": env_name},
{"$set": {"operational": OperationalStatus.ERROR.value}})
continue
listener = self.get_listener(env_name)
if not listener:
self.log.error("No listener is defined for env '{}'".format(env_name))
self.collection.update({"name": env_name},
{"$set": {"operational": OperationalStatus.ERROR.value}})
continue
# A dict that is shared between event manager and newly created env listener
process_vars = SharedManager().dict()
p = Process(target=self.listen_to_events,
args=(listener, env_name, process_vars,),
name=env_name)
self.processes.append({"process": p, "name": env_name, "vars": process_vars})
self.log.info("Starting event listener '{0}'".format(env_name))
p.start()
# Make sure statuses are up-to-date before event manager goes to sleep
self.update_operational_statuses()
time.sleep(self.interval)
finally:
# Fetch operational statuses before terminating listeners.
# Shared variables won't be available after termination.
stopping_processes = [process.get("name")
for process
in self.processes
if self._get_operational(process) != OperationalStatus.ERROR]
self._update_operational_status(OperationalStatus.ERROR)
# Gracefully stop processes
for process in self._get_alive_processes():
self.log.info("Stopping event listener '{0}'".format(process.get("name")))
process.get("process").terminate()
# Kill all remaining processes
self._kill_stuck_processes(self.processes)
# Updating operational statuses for stopped processes
self.collection.update_many(
{"name": {"$in": stopping_processes}},
{"$set": {"operational": OperationalStatus.STOPPED.value}}
)
if __name__ == "__main__":
EventManager().run()
|
