| 1 |
|
#!/usr/bin/python3 |
| 2 |
|
# -*- coding: utf-8 -*- |
| 3 |
|
|
| 4 |
|
## |
| 5 |
|
# Copyright 2018 Telefonica S.A. |
| 6 |
|
# |
| 7 |
|
# Licensed under the Apache License, Version 2.0 (the "License"); you may |
| 8 |
|
# not use this file except in compliance with the License. You may obtain |
| 9 |
|
# a copy of the License at |
| 10 |
|
# |
| 11 |
|
# http://www.apache.org/licenses/LICENSE-2.0 |
| 12 |
|
# |
| 13 |
|
# Unless required by applicable law or agreed to in writing, software |
| 14 |
|
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT |
| 15 |
|
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the |
| 16 |
|
# License for the specific language governing permissions and limitations |
| 17 |
|
# under the License. |
| 18 |
|
## |
| 19 |
|
|
| 20 |
|
|
| 21 |
|
# DEBUG WITH PDB |
| 22 |
1 |
import os |
| 23 |
1 |
import pdb |
| 24 |
|
|
| 25 |
1 |
import asyncio |
| 26 |
1 |
import yaml |
| 27 |
1 |
import logging |
| 28 |
1 |
import logging.handlers |
| 29 |
1 |
import getopt |
| 30 |
1 |
import sys |
| 31 |
|
|
| 32 |
1 |
from osm_lcm import ns, vim_sdn, netslice |
| 33 |
1 |
from osm_lcm.ng_ro import NgRoException, NgRoClient |
| 34 |
1 |
from osm_lcm.ROclient import ROClient, ROClientException |
| 35 |
|
|
| 36 |
1 |
from time import time |
| 37 |
1 |
from osm_lcm.lcm_utils import versiontuple, LcmException, TaskRegistry, LcmExceptionExit |
| 38 |
1 |
from osm_lcm import version as lcm_version, version_date as lcm_version_date |
| 39 |
|
|
| 40 |
1 |
from osm_common import msglocal, msgkafka |
| 41 |
1 |
from osm_common import version as common_version |
| 42 |
1 |
from osm_common.dbbase import DbException |
| 43 |
1 |
from osm_common.fsbase import FsException |
| 44 |
1 |
from osm_common.msgbase import MsgException |
| 45 |
1 |
from osm_lcm.data_utils.database.database import Database |
| 46 |
1 |
from osm_lcm.data_utils.filesystem.filesystem import Filesystem |
| 47 |
1 |
from osm_lcm.data_utils.lcm_config import LcmCfg |
| 48 |
1 |
from osm_lcm.lcm_hc import get_health_check_file |
| 49 |
1 |
from os import path |
| 50 |
1 |
from random import choice as random_choice |
| 51 |
1 |
from n2vc import version as n2vc_version |
| 52 |
1 |
import traceback |
| 53 |
|
|
| 54 |
1 |
if os.getenv("OSMLCM_PDB_DEBUG", None) is not None: |
| 55 |
0 |
pdb.set_trace() |
| 56 |
|
|
| 57 |
|
|
| 58 |
1 |
__author__ = "Alfonso Tierno" |
| 59 |
1 |
min_RO_version = "6.0.2" |
| 60 |
1 |
min_n2vc_version = "0.0.2" |
| 61 |
|
|
| 62 |
1 |
min_common_version = "0.1.19" |
| 63 |
|
|
| 64 |
|
|
| 65 |
1 |
class Lcm: |
| 66 |
1 |
ping_interval_pace = ( |
| 67 |
|
120 # how many time ping is send once is confirmed all is running |
| 68 |
|
) |
| 69 |
1 |
ping_interval_boot = 5 # how many time ping is sent when booting |
| 70 |
|
|
| 71 |
1 |
main_config = LcmCfg() |
| 72 |
|
|
| 73 |
1 |
def __init__(self, config_file, loop=None): |
| 74 |
|
""" |
| 75 |
|
Init, Connect to database, filesystem storage, and messaging |
| 76 |
|
:param config: two level dictionary with configuration. Top level should contain 'database', 'storage', |
| 77 |
|
:return: None |
| 78 |
|
""" |
| 79 |
1 |
self.db = None |
| 80 |
1 |
self.msg = None |
| 81 |
1 |
self.msg_admin = None |
| 82 |
1 |
self.fs = None |
| 83 |
1 |
self.pings_not_received = 1 |
| 84 |
1 |
self.consecutive_errors = 0 |
| 85 |
1 |
self.first_start = False |
| 86 |
|
|
| 87 |
|
# logging |
| 88 |
1 |
self.logger = logging.getLogger("lcm") |
| 89 |
|
# get id |
| 90 |
1 |
self.worker_id = self.get_process_id() |
| 91 |
|
# load configuration |
| 92 |
1 |
config = self.read_config_file(config_file) |
| 93 |
1 |
self.main_config.set_from_dict(config) |
| 94 |
1 |
self.main_config.transform() |
| 95 |
1 |
self.main_config.load_from_env() |
| 96 |
1 |
self.logger.critical("Loaded configuration:" + str(self.main_config.to_dict())) |
| 97 |
|
# TODO: check if lcm_hc.py is necessary |
| 98 |
1 |
self.health_check_file = get_health_check_file(self.main_config.to_dict()) |
| 99 |
1 |
self.loop = loop or asyncio.get_event_loop() |
| 100 |
1 |
self.ns = ( |
| 101 |
|
self.netslice |
| 102 |
|
) = ( |
| 103 |
|
self.vim |
| 104 |
|
) = self.wim = self.sdn = self.k8scluster = self.vca = self.k8srepo = None |
| 105 |
|
|
| 106 |
|
# logging |
| 107 |
1 |
log_format_simple = ( |
| 108 |
|
"%(asctime)s %(levelname)s %(name)s %(filename)s:%(lineno)s %(message)s" |
| 109 |
|
) |
| 110 |
1 |
log_formatter_simple = logging.Formatter( |
| 111 |
|
log_format_simple, datefmt="%Y-%m-%dT%H:%M:%S" |
| 112 |
|
) |
| 113 |
1 |
if self.main_config.globalConfig.logfile: |
| 114 |
0 |
file_handler = logging.handlers.RotatingFileHandler( |
| 115 |
|
self.main_config.globalConfig.logfile, |
| 116 |
|
maxBytes=100e6, |
| 117 |
|
backupCount=9, |
| 118 |
|
delay=0, |
| 119 |
|
) |
| 120 |
0 |
file_handler.setFormatter(log_formatter_simple) |
| 121 |
0 |
self.logger.addHandler(file_handler) |
| 122 |
1 |
if not self.main_config.globalConfig.to_dict()["nologging"]: |
| 123 |
1 |
str_handler = logging.StreamHandler() |
| 124 |
1 |
str_handler.setFormatter(log_formatter_simple) |
| 125 |
1 |
self.logger.addHandler(str_handler) |
| 126 |
|
|
| 127 |
1 |
if self.main_config.globalConfig.to_dict()["loglevel"]: |
| 128 |
1 |
self.logger.setLevel(self.main_config.globalConfig.loglevel) |
| 129 |
|
|
| 130 |
|
# logging other modules |
| 131 |
1 |
for logger in ("message", "database", "storage", "tsdb"): |
| 132 |
1 |
logger_config = self.main_config.to_dict()[logger] |
| 133 |
1 |
logger_module = logging.getLogger(logger_config["logger_name"]) |
| 134 |
1 |
if logger_config["logfile"]: |
| 135 |
0 |
file_handler = logging.handlers.RotatingFileHandler( |
| 136 |
|
logger_config["logfile"], maxBytes=100e6, backupCount=9, delay=0 |
| 137 |
|
) |
| 138 |
0 |
file_handler.setFormatter(log_formatter_simple) |
| 139 |
0 |
logger_module.addHandler(file_handler) |
| 140 |
1 |
if logger_config["loglevel"]: |
| 141 |
1 |
logger_module.setLevel(logger_config["loglevel"]) |
| 142 |
1 |
self.logger.critical( |
| 143 |
|
"starting osm/lcm version {} {}".format(lcm_version, lcm_version_date) |
| 144 |
|
) |
| 145 |
|
|
| 146 |
|
# check version of N2VC |
| 147 |
|
# TODO enhance with int conversion or from distutils.version import LooseVersion |
| 148 |
|
# or with list(map(int, version.split("."))) |
| 149 |
1 |
if versiontuple(n2vc_version) < versiontuple(min_n2vc_version): |
| 150 |
0 |
raise LcmException( |
| 151 |
|
"Not compatible osm/N2VC version '{}'. Needed '{}' or higher".format( |
| 152 |
|
n2vc_version, min_n2vc_version |
| 153 |
|
) |
| 154 |
|
) |
| 155 |
|
# check version of common |
| 156 |
1 |
if versiontuple(common_version) < versiontuple(min_common_version): |
| 157 |
0 |
raise LcmException( |
| 158 |
|
"Not compatible osm/common version '{}'. Needed '{}' or higher".format( |
| 159 |
|
common_version, min_common_version |
| 160 |
|
) |
| 161 |
|
) |
| 162 |
|
|
| 163 |
1 |
try: |
| 164 |
1 |
self.db = Database(self.main_config.to_dict()).instance.db |
| 165 |
|
|
| 166 |
1 |
self.fs = Filesystem(self.main_config.to_dict()).instance.fs |
| 167 |
1 |
self.fs.sync() |
| 168 |
|
|
| 169 |
|
# copy message configuration in order to remove 'group_id' for msg_admin |
| 170 |
1 |
config_message = self.main_config.message.to_dict() |
| 171 |
1 |
config_message["loop"] = self.loop |
| 172 |
1 |
if config_message["driver"] == "local": |
| 173 |
1 |
self.msg = msglocal.MsgLocal() |
| 174 |
1 |
self.msg.connect(config_message) |
| 175 |
1 |
self.msg_admin = msglocal.MsgLocal() |
| 176 |
1 |
config_message.pop("group_id", None) |
| 177 |
1 |
self.msg_admin.connect(config_message) |
| 178 |
0 |
elif config_message["driver"] == "kafka": |
| 179 |
0 |
self.msg = msgkafka.MsgKafka() |
| 180 |
0 |
self.msg.connect(config_message) |
| 181 |
0 |
self.msg_admin = msgkafka.MsgKafka() |
| 182 |
0 |
config_message.pop("group_id", None) |
| 183 |
0 |
self.msg_admin.connect(config_message) |
| 184 |
|
else: |
| 185 |
0 |
raise LcmException( |
| 186 |
|
"Invalid configuration param '{}' at '[message]':'driver'".format( |
| 187 |
|
self.main_config.message.driver |
| 188 |
|
) |
| 189 |
|
) |
| 190 |
0 |
except (DbException, FsException, MsgException) as e: |
| 191 |
0 |
self.logger.critical(str(e), exc_info=True) |
| 192 |
0 |
raise LcmException(str(e)) |
| 193 |
|
|
| 194 |
|
# contains created tasks/futures to be able to cancel |
| 195 |
1 |
self.lcm_tasks = TaskRegistry(self.worker_id, self.logger) |
| 196 |
|
|
| 197 |
1 |
async def check_RO_version(self): |
| 198 |
0 |
tries = 14 |
| 199 |
0 |
last_error = None |
| 200 |
|
while True: |
| 201 |
0 |
ro_uri = self.main_config.RO.uri |
| 202 |
0 |
if not ro_uri: |
| 203 |
0 |
ro_uri = "" |
| 204 |
0 |
try: |
| 205 |
|
# try new RO, if fail old RO |
| 206 |
0 |
try: |
| 207 |
0 |
self.main_config.RO.uri = ro_uri + "ro" |
| 208 |
0 |
ro_server = NgRoClient(self.loop, **self.main_config.RO.to_dict()) |
| 209 |
0 |
ro_version = await ro_server.get_version() |
| 210 |
0 |
self.main_config.RO.ng = True |
| 211 |
0 |
except Exception: |
| 212 |
0 |
self.main_config.RO.uri = ro_uri + "openmano" |
| 213 |
0 |
ro_server = ROClient(self.loop, **self.main_config.RO.to_dict()) |
| 214 |
0 |
ro_version = await ro_server.get_version() |
| 215 |
0 |
self.main_config.RO.ng = False |
| 216 |
0 |
if versiontuple(ro_version) < versiontuple(min_RO_version): |
| 217 |
0 |
raise LcmException( |
| 218 |
|
"Not compatible osm/RO version '{}'. Needed '{}' or higher".format( |
| 219 |
|
ro_version, min_RO_version |
| 220 |
|
) |
| 221 |
|
) |
| 222 |
0 |
self.logger.info( |
| 223 |
|
"Connected to RO version {} new-generation version {}".format( |
| 224 |
|
ro_version, self.main_config.RO.ng |
| 225 |
|
) |
| 226 |
|
) |
| 227 |
0 |
return |
| 228 |
0 |
except (ROClientException, NgRoException) as e: |
| 229 |
0 |
self.main_config.RO.uri = ro_uri |
| 230 |
0 |
tries -= 1 |
| 231 |
0 |
traceback.print_tb(e.__traceback__) |
| 232 |
0 |
error_text = "Error while connecting to RO on {}: {}".format( |
| 233 |
|
self.main_config.RO.uri, e |
| 234 |
|
) |
| 235 |
0 |
if tries <= 0: |
| 236 |
0 |
self.logger.critical(error_text) |
| 237 |
0 |
raise LcmException(error_text) |
| 238 |
0 |
if last_error != error_text: |
| 239 |
0 |
last_error = error_text |
| 240 |
0 |
self.logger.error( |
| 241 |
|
error_text + ". Waiting until {} seconds".format(5 * tries) |
| 242 |
|
) |
| 243 |
0 |
await asyncio.sleep(5) |
| 244 |
|
|
| 245 |
1 |
async def test(self, param=None): |
| 246 |
0 |
self.logger.debug("Starting/Ending test task: {}".format(param)) |
| 247 |
|
|
| 248 |
1 |
async def kafka_ping(self): |
| 249 |
0 |
self.logger.debug("Task kafka_ping Enter") |
| 250 |
0 |
consecutive_errors = 0 |
| 251 |
0 |
first_start = True |
| 252 |
0 |
kafka_has_received = False |
| 253 |
0 |
self.pings_not_received = 1 |
| 254 |
|
while True: |
| 255 |
0 |
try: |
| 256 |
0 |
await self.msg_admin.aiowrite( |
| 257 |
|
"admin", |
| 258 |
|
"ping", |
| 259 |
|
{ |
| 260 |
|
"from": "lcm", |
| 261 |
|
"to": "lcm", |
| 262 |
|
"worker_id": self.worker_id, |
| 263 |
|
"version": lcm_version, |
| 264 |
|
}, |
| 265 |
|
self.loop, |
| 266 |
|
) |
| 267 |
|
# time between pings are low when it is not received and at starting |
| 268 |
0 |
wait_time = ( |
| 269 |
|
self.ping_interval_boot |
| 270 |
|
if not kafka_has_received |
| 271 |
|
else self.ping_interval_pace |
| 272 |
|
) |
| 273 |
0 |
if not self.pings_not_received: |
| 274 |
0 |
kafka_has_received = True |
| 275 |
0 |
self.pings_not_received += 1 |
| 276 |
0 |
await asyncio.sleep(wait_time, loop=self.loop) |
| 277 |
0 |
if self.pings_not_received > 10: |
| 278 |
0 |
raise LcmException("It is not receiving pings from Kafka bus") |
| 279 |
0 |
consecutive_errors = 0 |
| 280 |
0 |
first_start = False |
| 281 |
0 |
except LcmException: |
| 282 |
0 |
raise |
| 283 |
0 |
except Exception as e: |
| 284 |
|
# if not first_start is the first time after starting. So leave more time and wait |
| 285 |
|
# to allow kafka starts |
| 286 |
0 |
if consecutive_errors == 8 if not first_start else 30: |
| 287 |
0 |
self.logger.error( |
| 288 |
|
"Task kafka_read task exit error too many errors. Exception: {}".format( |
| 289 |
|
e |
| 290 |
|
) |
| 291 |
|
) |
| 292 |
0 |
raise |
| 293 |
0 |
consecutive_errors += 1 |
| 294 |
0 |
self.logger.error( |
| 295 |
|
"Task kafka_read retrying after Exception {}".format(e) |
| 296 |
|
) |
| 297 |
0 |
wait_time = 2 if not first_start else 5 |
| 298 |
0 |
await asyncio.sleep(wait_time, loop=self.loop) |
| 299 |
|
|
| 300 |
1 |
def kafka_read_callback(self, topic, command, params): |
| 301 |
1 |
order_id = 1 |
| 302 |
|
|
| 303 |
1 |
if topic != "admin" and command != "ping": |
| 304 |
0 |
self.logger.debug( |
| 305 |
|
"Task kafka_read receives {} {}: {}".format(topic, command, params) |
| 306 |
|
) |
| 307 |
1 |
self.consecutive_errors = 0 |
| 308 |
1 |
self.first_start = False |
| 309 |
1 |
order_id += 1 |
| 310 |
1 |
if command == "exit": |
| 311 |
0 |
raise LcmExceptionExit |
| 312 |
1 |
elif command.startswith("#"): |
| 313 |
0 |
return |
| 314 |
1 |
elif command == "echo": |
| 315 |
|
# just for test |
| 316 |
0 |
print(params) |
| 317 |
0 |
sys.stdout.flush() |
| 318 |
0 |
return |
| 319 |
1 |
elif command == "test": |
| 320 |
0 |
asyncio.Task(self.test(params), loop=self.loop) |
| 321 |
0 |
return |
| 322 |
|
|
| 323 |
1 |
if topic == "admin": |
| 324 |
1 |
if command == "ping" and params["to"] == "lcm" and params["from"] == "lcm": |
| 325 |
1 |
if params.get("worker_id") != self.worker_id: |
| 326 |
1 |
return |
| 327 |
1 |
self.pings_not_received = 0 |
| 328 |
1 |
try: |
| 329 |
1 |
with open(self.health_check_file, "w") as f: |
| 330 |
1 |
f.write(str(time())) |
| 331 |
0 |
except Exception as e: |
| 332 |
0 |
self.logger.error( |
| 333 |
|
"Cannot write into '{}' for healthcheck: {}".format( |
| 334 |
|
self.health_check_file, e |
| 335 |
|
) |
| 336 |
|
) |
| 337 |
1 |
return |
| 338 |
1 |
elif topic == "pla": |
| 339 |
0 |
if command == "placement": |
| 340 |
0 |
self.ns.update_nsrs_with_pla_result(params) |
| 341 |
0 |
return |
| 342 |
1 |
elif topic == "k8scluster": |
| 343 |
0 |
if command == "create" or command == "created": |
| 344 |
0 |
k8scluster_id = params.get("_id") |
| 345 |
0 |
task = asyncio.ensure_future(self.k8scluster.create(params, order_id)) |
| 346 |
0 |
self.lcm_tasks.register( |
| 347 |
|
"k8scluster", k8scluster_id, order_id, "k8scluster_create", task |
| 348 |
|
) |
| 349 |
0 |
return |
| 350 |
0 |
elif command == "delete" or command == "deleted": |
| 351 |
0 |
k8scluster_id = params.get("_id") |
| 352 |
0 |
task = asyncio.ensure_future(self.k8scluster.delete(params, order_id)) |
| 353 |
0 |
self.lcm_tasks.register( |
| 354 |
|
"k8scluster", k8scluster_id, order_id, "k8scluster_delete", task |
| 355 |
|
) |
| 356 |
0 |
return |
| 357 |
1 |
elif topic == "vca": |
| 358 |
0 |
if command == "create" or command == "created": |
| 359 |
0 |
vca_id = params.get("_id") |
| 360 |
0 |
task = asyncio.ensure_future(self.vca.create(params, order_id)) |
| 361 |
0 |
self.lcm_tasks.register("vca", vca_id, order_id, "vca_create", task) |
| 362 |
0 |
return |
| 363 |
0 |
elif command == "delete" or command == "deleted": |
| 364 |
0 |
vca_id = params.get("_id") |
| 365 |
0 |
task = asyncio.ensure_future(self.vca.delete(params, order_id)) |
| 366 |
0 |
self.lcm_tasks.register("vca", vca_id, order_id, "vca_delete", task) |
| 367 |
0 |
return |
| 368 |
1 |
elif topic == "k8srepo": |
| 369 |
0 |
if command == "create" or command == "created": |
| 370 |
0 |
k8srepo_id = params.get("_id") |
| 371 |
0 |
self.logger.debug("k8srepo_id = {}".format(k8srepo_id)) |
| 372 |
0 |
task = asyncio.ensure_future(self.k8srepo.create(params, order_id)) |
| 373 |
0 |
self.lcm_tasks.register( |
| 374 |
|
"k8srepo", k8srepo_id, order_id, "k8srepo_create", task |
| 375 |
|
) |
| 376 |
0 |
return |
| 377 |
0 |
elif command == "delete" or command == "deleted": |
| 378 |
0 |
k8srepo_id = params.get("_id") |
| 379 |
0 |
task = asyncio.ensure_future(self.k8srepo.delete(params, order_id)) |
| 380 |
0 |
self.lcm_tasks.register( |
| 381 |
|
"k8srepo", k8srepo_id, order_id, "k8srepo_delete", task |
| 382 |
|
) |
| 383 |
0 |
return |
| 384 |
1 |
elif topic == "ns": |
| 385 |
0 |
if command == "instantiate": |
| 386 |
|
# self.logger.debug("Deploying NS {}".format(nsr_id)) |
| 387 |
0 |
nslcmop = params |
| 388 |
0 |
nslcmop_id = nslcmop["_id"] |
| 389 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 390 |
0 |
task = asyncio.ensure_future(self.ns.instantiate(nsr_id, nslcmop_id)) |
| 391 |
0 |
self.lcm_tasks.register( |
| 392 |
|
"ns", nsr_id, nslcmop_id, "ns_instantiate", task |
| 393 |
|
) |
| 394 |
0 |
return |
| 395 |
0 |
elif command == "terminate": |
| 396 |
|
# self.logger.debug("Deleting NS {}".format(nsr_id)) |
| 397 |
0 |
nslcmop = params |
| 398 |
0 |
nslcmop_id = nslcmop["_id"] |
| 399 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 400 |
0 |
self.lcm_tasks.cancel(topic, nsr_id) |
| 401 |
0 |
task = asyncio.ensure_future(self.ns.terminate(nsr_id, nslcmop_id)) |
| 402 |
0 |
self.lcm_tasks.register("ns", nsr_id, nslcmop_id, "ns_terminate", task) |
| 403 |
0 |
return |
| 404 |
0 |
elif command == "vca_status_refresh": |
| 405 |
0 |
nslcmop = params |
| 406 |
0 |
nslcmop_id = nslcmop["_id"] |
| 407 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 408 |
0 |
task = asyncio.ensure_future( |
| 409 |
|
self.ns.vca_status_refresh(nsr_id, nslcmop_id) |
| 410 |
|
) |
| 411 |
0 |
self.lcm_tasks.register( |
| 412 |
|
"ns", nsr_id, nslcmop_id, "ns_vca_status_refresh", task |
| 413 |
|
) |
| 414 |
0 |
return |
| 415 |
0 |
elif command == "action": |
| 416 |
|
# self.logger.debug("Update NS {}".format(nsr_id)) |
| 417 |
0 |
nslcmop = params |
| 418 |
0 |
nslcmop_id = nslcmop["_id"] |
| 419 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 420 |
0 |
task = asyncio.ensure_future(self.ns.action(nsr_id, nslcmop_id)) |
| 421 |
0 |
self.lcm_tasks.register("ns", nsr_id, nslcmop_id, "ns_action", task) |
| 422 |
0 |
return |
| 423 |
0 |
elif command == "update": |
| 424 |
|
# self.logger.debug("Update NS {}".format(nsr_id)) |
| 425 |
0 |
nslcmop = params |
| 426 |
0 |
nslcmop_id = nslcmop["_id"] |
| 427 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 428 |
0 |
task = asyncio.ensure_future(self.ns.update(nsr_id, nslcmop_id)) |
| 429 |
0 |
self.lcm_tasks.register("ns", nsr_id, nslcmop_id, "ns_update", task) |
| 430 |
0 |
return |
| 431 |
0 |
elif command == "scale": |
| 432 |
|
# self.logger.debug("Update NS {}".format(nsr_id)) |
| 433 |
0 |
nslcmop = params |
| 434 |
0 |
nslcmop_id = nslcmop["_id"] |
| 435 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 436 |
0 |
task = asyncio.ensure_future(self.ns.scale(nsr_id, nslcmop_id)) |
| 437 |
0 |
self.lcm_tasks.register("ns", nsr_id, nslcmop_id, "ns_scale", task) |
| 438 |
0 |
return |
| 439 |
0 |
elif command == "heal": |
| 440 |
|
# self.logger.debug("Healing NS {}".format(nsr_id)) |
| 441 |
0 |
nslcmop = params |
| 442 |
0 |
nslcmop_id = nslcmop["_id"] |
| 443 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 444 |
0 |
task = asyncio.ensure_future(self.ns.heal(nsr_id, nslcmop_id)) |
| 445 |
0 |
self.lcm_tasks.register("ns", nsr_id, nslcmop_id, "ns_heal", task) |
| 446 |
0 |
return |
| 447 |
0 |
elif command == "migrate": |
| 448 |
0 |
nslcmop = params |
| 449 |
0 |
nslcmop_id = nslcmop["_id"] |
| 450 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 451 |
0 |
task = asyncio.ensure_future(self.ns.migrate(nsr_id, nslcmop_id)) |
| 452 |
0 |
self.lcm_tasks.register("ns", nsr_id, nslcmop_id, "ns_migrate", task) |
| 453 |
0 |
return |
| 454 |
0 |
elif command == "verticalscale": |
| 455 |
0 |
nslcmop = params |
| 456 |
0 |
nslcmop_id = nslcmop["_id"] |
| 457 |
0 |
nsr_id = nslcmop["nsInstanceId"] |
| 458 |
0 |
task = asyncio.ensure_future(self.ns.vertical_scale(nsr_id, nslcmop_id)) |
| 459 |
0 |
self.logger.debug( |
| 460 |
|
"nsr_id,nslcmop_id,task {},{},{}".format(nsr_id, nslcmop_id, task) |
| 461 |
|
) |
| 462 |
0 |
self.lcm_tasks.register( |
| 463 |
|
"ns", nsr_id, nslcmop_id, "ns_verticalscale", task |
| 464 |
|
) |
| 465 |
0 |
self.logger.debug( |
| 466 |
|
"LCM task registered {},{},{} ".format(nsr_id, nslcmop_id, task) |
| 467 |
|
) |
| 468 |
0 |
return |
| 469 |
0 |
elif command == "show": |
| 470 |
0 |
nsr_id = params |
| 471 |
0 |
try: |
| 472 |
0 |
db_nsr = self.db.get_one("nsrs", {"_id": nsr_id}) |
| 473 |
0 |
print( |
| 474 |
|
"nsr:\n _id={}\n operational-status: {}\n config-status: {}" |
| 475 |
|
"\n detailed-status: {}\n deploy: {}\n tasks: {}" |
| 476 |
|
"".format( |
| 477 |
|
nsr_id, |
| 478 |
|
db_nsr["operational-status"], |
| 479 |
|
db_nsr["config-status"], |
| 480 |
|
db_nsr["detailed-status"], |
| 481 |
|
db_nsr["_admin"]["deployed"], |
| 482 |
|
self.lcm_ns_tasks.get(nsr_id), |
| 483 |
|
) |
| 484 |
|
) |
| 485 |
0 |
except Exception as e: |
| 486 |
0 |
print("nsr {} not found: {}".format(nsr_id, e)) |
| 487 |
0 |
sys.stdout.flush() |
| 488 |
0 |
return |
| 489 |
0 |
elif command == "deleted": |
| 490 |
0 |
return # TODO cleaning of task just in case should be done |
| 491 |
0 |
elif command in ( |
| 492 |
|
"vnf_terminated", |
| 493 |
|
"policy_updated", |
| 494 |
|
"terminated", |
| 495 |
|
"instantiated", |
| 496 |
|
"scaled", |
| 497 |
|
"healed", |
| 498 |
|
"actioned", |
| 499 |
|
"updated", |
| 500 |
|
"migrated", |
| 501 |
|
"verticalscaled", |
| 502 |
|
): # "scaled-cooldown-time" |
| 503 |
0 |
return |
| 504 |
|
|
| 505 |
1 |
elif topic == "nsi": # netslice LCM processes (instantiate, terminate, etc) |
| 506 |
0 |
if command == "instantiate": |
| 507 |
|
# self.logger.debug("Instantiating Network Slice {}".format(nsilcmop["netsliceInstanceId"])) |
| 508 |
0 |
nsilcmop = params |
| 509 |
0 |
nsilcmop_id = nsilcmop["_id"] # slice operation id |
| 510 |
0 |
nsir_id = nsilcmop["netsliceInstanceId"] # slice record id |
| 511 |
0 |
task = asyncio.ensure_future( |
| 512 |
|
self.netslice.instantiate(nsir_id, nsilcmop_id) |
| 513 |
|
) |
| 514 |
0 |
self.lcm_tasks.register( |
| 515 |
|
"nsi", nsir_id, nsilcmop_id, "nsi_instantiate", task |
| 516 |
|
) |
| 517 |
0 |
return |
| 518 |
0 |
elif command == "terminate": |
| 519 |
|
# self.logger.debug("Terminating Network Slice NS {}".format(nsilcmop["netsliceInstanceId"])) |
| 520 |
0 |
nsilcmop = params |
| 521 |
0 |
nsilcmop_id = nsilcmop["_id"] # slice operation id |
| 522 |
0 |
nsir_id = nsilcmop["netsliceInstanceId"] # slice record id |
| 523 |
0 |
self.lcm_tasks.cancel(topic, nsir_id) |
| 524 |
0 |
task = asyncio.ensure_future( |
| 525 |
|
self.netslice.terminate(nsir_id, nsilcmop_id) |
| 526 |
|
) |
| 527 |
0 |
self.lcm_tasks.register( |
| 528 |
|
"nsi", nsir_id, nsilcmop_id, "nsi_terminate", task |
| 529 |
|
) |
| 530 |
0 |
return |
| 531 |
0 |
elif command == "show": |
| 532 |
0 |
nsir_id = params |
| 533 |
0 |
try: |
| 534 |
0 |
db_nsir = self.db.get_one("nsirs", {"_id": nsir_id}) |
| 535 |
0 |
print( |
| 536 |
|
"nsir:\n _id={}\n operational-status: {}\n config-status: {}" |
| 537 |
|
"\n detailed-status: {}\n deploy: {}\n tasks: {}" |
| 538 |
|
"".format( |
| 539 |
|
nsir_id, |
| 540 |
|
db_nsir["operational-status"], |
| 541 |
|
db_nsir["config-status"], |
| 542 |
|
db_nsir["detailed-status"], |
| 543 |
|
db_nsir["_admin"]["deployed"], |
| 544 |
|
self.lcm_netslice_tasks.get(nsir_id), |
| 545 |
|
) |
| 546 |
|
) |
| 547 |
0 |
except Exception as e: |
| 548 |
0 |
print("nsir {} not found: {}".format(nsir_id, e)) |
| 549 |
0 |
sys.stdout.flush() |
| 550 |
0 |
return |
| 551 |
0 |
elif command == "deleted": |
| 552 |
0 |
return # TODO cleaning of task just in case should be done |
| 553 |
0 |
elif command in ( |
| 554 |
|
"terminated", |
| 555 |
|
"instantiated", |
| 556 |
|
"scaled", |
| 557 |
|
"healed", |
| 558 |
|
"actioned", |
| 559 |
|
): # "scaled-cooldown-time" |
| 560 |
0 |
return |
| 561 |
1 |
elif topic == "vim_account": |
| 562 |
0 |
vim_id = params["_id"] |
| 563 |
0 |
if command in ("create", "created"): |
| 564 |
0 |
if not self.main_config.RO.ng: |
| 565 |
0 |
task = asyncio.ensure_future(self.vim.create(params, order_id)) |
| 566 |
0 |
self.lcm_tasks.register( |
| 567 |
|
"vim_account", vim_id, order_id, "vim_create", task |
| 568 |
|
) |
| 569 |
0 |
return |
| 570 |
0 |
elif command == "delete" or command == "deleted": |
| 571 |
0 |
self.lcm_tasks.cancel(topic, vim_id) |
| 572 |
0 |
task = asyncio.ensure_future(self.vim.delete(params, order_id)) |
| 573 |
0 |
self.lcm_tasks.register( |
| 574 |
|
"vim_account", vim_id, order_id, "vim_delete", task |
| 575 |
|
) |
| 576 |
0 |
return |
| 577 |
0 |
elif command == "show": |
| 578 |
0 |
print("not implemented show with vim_account") |
| 579 |
0 |
sys.stdout.flush() |
| 580 |
0 |
return |
| 581 |
0 |
elif command in ("edit", "edited"): |
| 582 |
0 |
if not self.main_config.RO.ng: |
| 583 |
0 |
task = asyncio.ensure_future(self.vim.edit(params, order_id)) |
| 584 |
0 |
self.lcm_tasks.register( |
| 585 |
|
"vim_account", vim_id, order_id, "vim_edit", task |
| 586 |
|
) |
| 587 |
0 |
return |
| 588 |
0 |
elif command == "deleted": |
| 589 |
0 |
return # TODO cleaning of task just in case should be done |
| 590 |
1 |
elif topic == "wim_account": |
| 591 |
0 |
wim_id = params["_id"] |
| 592 |
0 |
if command in ("create", "created"): |
| 593 |
0 |
if not self.main_config.RO.ng: |
| 594 |
0 |
task = asyncio.ensure_future(self.wim.create(params, order_id)) |
| 595 |
0 |
self.lcm_tasks.register( |
| 596 |
|
"wim_account", wim_id, order_id, "wim_create", task |
| 597 |
|
) |
| 598 |
0 |
return |
| 599 |
0 |
elif command == "delete" or command == "deleted": |
| 600 |
0 |
self.lcm_tasks.cancel(topic, wim_id) |
| 601 |
0 |
task = asyncio.ensure_future(self.wim.delete(params, order_id)) |
| 602 |
0 |
self.lcm_tasks.register( |
| 603 |
|
"wim_account", wim_id, order_id, "wim_delete", task |
| 604 |
|
) |
| 605 |
0 |
return |
| 606 |
0 |
elif command == "show": |
| 607 |
0 |
print("not implemented show with wim_account") |
| 608 |
0 |
sys.stdout.flush() |
| 609 |
0 |
return |
| 610 |
0 |
elif command in ("edit", "edited"): |
| 611 |
0 |
task = asyncio.ensure_future(self.wim.edit(params, order_id)) |
| 612 |
0 |
self.lcm_tasks.register( |
| 613 |
|
"wim_account", wim_id, order_id, "wim_edit", task |
| 614 |
|
) |
| 615 |
0 |
return |
| 616 |
0 |
elif command == "deleted": |
| 617 |
0 |
return # TODO cleaning of task just in case should be done |
| 618 |
1 |
elif topic == "sdn": |
| 619 |
0 |
_sdn_id = params["_id"] |
| 620 |
0 |
if command in ("create", "created"): |
| 621 |
0 |
if not self.main_config.RO.ng: |
| 622 |
0 |
task = asyncio.ensure_future(self.sdn.create(params, order_id)) |
| 623 |
0 |
self.lcm_tasks.register( |
| 624 |
|
"sdn", _sdn_id, order_id, "sdn_create", task |
| 625 |
|
) |
| 626 |
0 |
return |
| 627 |
0 |
elif command == "delete" or command == "deleted": |
| 628 |
0 |
self.lcm_tasks.cancel(topic, _sdn_id) |
| 629 |
0 |
task = asyncio.ensure_future(self.sdn.delete(params, order_id)) |
| 630 |
0 |
self.lcm_tasks.register("sdn", _sdn_id, order_id, "sdn_delete", task) |
| 631 |
0 |
return |
| 632 |
0 |
elif command in ("edit", "edited"): |
| 633 |
0 |
task = asyncio.ensure_future(self.sdn.edit(params, order_id)) |
| 634 |
0 |
self.lcm_tasks.register("sdn", _sdn_id, order_id, "sdn_edit", task) |
| 635 |
0 |
return |
| 636 |
0 |
elif command == "deleted": |
| 637 |
0 |
return # TODO cleaning of task just in case should be done |
| 638 |
1 |
self.logger.critical("unknown topic {} and command '{}'".format(topic, command)) |
| 639 |
|
|
| 640 |
1 |
async def kafka_read(self): |
| 641 |
0 |
self.logger.debug( |
| 642 |
|
"Task kafka_read Enter with worker_id={}".format(self.worker_id) |
| 643 |
|
) |
| 644 |
|
# future = asyncio.Future() |
| 645 |
0 |
self.consecutive_errors = 0 |
| 646 |
0 |
self.first_start = True |
| 647 |
0 |
while self.consecutive_errors < 10: |
| 648 |
0 |
try: |
| 649 |
0 |
topics = ( |
| 650 |
|
"ns", |
| 651 |
|
"vim_account", |
| 652 |
|
"wim_account", |
| 653 |
|
"sdn", |
| 654 |
|
"nsi", |
| 655 |
|
"k8scluster", |
| 656 |
|
"vca", |
| 657 |
|
"k8srepo", |
| 658 |
|
"pla", |
| 659 |
|
) |
| 660 |
0 |
topics_admin = ("admin",) |
| 661 |
0 |
await asyncio.gather( |
| 662 |
|
self.msg.aioread( |
| 663 |
|
topics, self.loop, self.kafka_read_callback, from_beginning=True |
| 664 |
|
), |
| 665 |
|
self.msg_admin.aioread( |
| 666 |
|
topics_admin, |
| 667 |
|
self.loop, |
| 668 |
|
self.kafka_read_callback, |
| 669 |
|
group_id=False, |
| 670 |
|
), |
| 671 |
|
) |
| 672 |
|
|
| 673 |
0 |
except LcmExceptionExit: |
| 674 |
0 |
self.logger.debug("Bye!") |
| 675 |
0 |
break |
| 676 |
0 |
except Exception as e: |
| 677 |
|
# if not first_start is the first time after starting. So leave more time and wait |
| 678 |
|
# to allow kafka starts |
| 679 |
0 |
if self.consecutive_errors == 8 if not self.first_start else 30: |
| 680 |
0 |
self.logger.error( |
| 681 |
|
"Task kafka_read task exit error too many errors. Exception: {}".format( |
| 682 |
|
e |
| 683 |
|
) |
| 684 |
|
) |
| 685 |
0 |
raise |
| 686 |
0 |
self.consecutive_errors += 1 |
| 687 |
0 |
self.logger.error( |
| 688 |
|
"Task kafka_read retrying after Exception {}".format(e) |
| 689 |
|
) |
| 690 |
0 |
wait_time = 2 if not self.first_start else 5 |
| 691 |
0 |
await asyncio.sleep(wait_time, loop=self.loop) |
| 692 |
|
|
| 693 |
|
# self.logger.debug("Task kafka_read terminating") |
| 694 |
0 |
self.logger.debug("Task kafka_read exit") |
| 695 |
|
|
| 696 |
1 |
def start(self): |
| 697 |
|
# check RO version |
| 698 |
0 |
self.loop.run_until_complete(self.check_RO_version()) |
| 699 |
|
|
| 700 |
0 |
self.ns = ns.NsLcm(self.msg, self.lcm_tasks, self.main_config, self.loop) |
| 701 |
|
# TODO: modify the rest of classes to use the LcmCfg object instead of dicts |
| 702 |
0 |
self.netslice = netslice.NetsliceLcm( |
| 703 |
|
self.msg, self.lcm_tasks, self.main_config.to_dict(), self.loop, self.ns |
| 704 |
|
) |
| 705 |
0 |
self.vim = vim_sdn.VimLcm( |
| 706 |
|
self.msg, self.lcm_tasks, self.main_config.to_dict(), self.loop |
| 707 |
|
) |
| 708 |
0 |
self.wim = vim_sdn.WimLcm( |
| 709 |
|
self.msg, self.lcm_tasks, self.main_config.to_dict(), self.loop |
| 710 |
|
) |
| 711 |
0 |
self.sdn = vim_sdn.SdnLcm( |
| 712 |
|
self.msg, self.lcm_tasks, self.main_config.to_dict(), self.loop |
| 713 |
|
) |
| 714 |
0 |
self.k8scluster = vim_sdn.K8sClusterLcm( |
| 715 |
|
self.msg, self.lcm_tasks, self.main_config.to_dict(), self.loop |
| 716 |
|
) |
| 717 |
0 |
self.vca = vim_sdn.VcaLcm( |
| 718 |
|
self.msg, self.lcm_tasks, self.main_config.to_dict(), self.loop |
| 719 |
|
) |
| 720 |
0 |
self.k8srepo = vim_sdn.K8sRepoLcm( |
| 721 |
|
self.msg, self.lcm_tasks, self.main_config.to_dict(), self.loop |
| 722 |
|
) |
| 723 |
|
|
| 724 |
0 |
self.loop.run_until_complete( |
| 725 |
|
asyncio.gather(self.kafka_read(), self.kafka_ping()) |
| 726 |
|
) |
| 727 |
|
|
| 728 |
|
# TODO |
| 729 |
|
# self.logger.debug("Terminating cancelling creation tasks") |
| 730 |
|
# self.lcm_tasks.cancel("ALL", "create") |
| 731 |
|
# timeout = 200 |
| 732 |
|
# while self.is_pending_tasks(): |
| 733 |
|
# self.logger.debug("Task kafka_read terminating. Waiting for tasks termination") |
| 734 |
|
# await asyncio.sleep(2, loop=self.loop) |
| 735 |
|
# timeout -= 2 |
| 736 |
|
# if not timeout: |
| 737 |
|
# self.lcm_tasks.cancel("ALL", "ALL") |
| 738 |
0 |
self.loop.close() |
| 739 |
0 |
self.loop = None |
| 740 |
0 |
if self.db: |
| 741 |
0 |
self.db.db_disconnect() |
| 742 |
0 |
if self.msg: |
| 743 |
0 |
self.msg.disconnect() |
| 744 |
0 |
if self.msg_admin: |
| 745 |
0 |
self.msg_admin.disconnect() |
| 746 |
0 |
if self.fs: |
| 747 |
0 |
self.fs.fs_disconnect() |
| 748 |
|
|
| 749 |
1 |
def read_config_file(self, config_file): |
| 750 |
1 |
try: |
| 751 |
1 |
with open(config_file) as f: |
| 752 |
1 |
return yaml.safe_load(f) |
| 753 |
0 |
except Exception as e: |
| 754 |
0 |
self.logger.critical("At config file '{}': {}".format(config_file, e)) |
| 755 |
0 |
exit(1) |
| 756 |
|
|
| 757 |
1 |
@staticmethod |
| 758 |
1 |
def get_process_id(): |
| 759 |
|
""" |
| 760 |
|
Obtain a unique ID for this process. If running from inside docker, it will get docker ID. If not it |
| 761 |
|
will provide a random one |
| 762 |
|
:return: Obtained ID |
| 763 |
|
""" |
| 764 |
|
# Try getting docker id. If fails, get pid |
| 765 |
1 |
try: |
| 766 |
1 |
with open("/proc/self/cgroup", "r") as f: |
| 767 |
1 |
text_id_ = f.readline() |
| 768 |
1 |
_, _, text_id = text_id_.rpartition("/") |
| 769 |
1 |
text_id = text_id.replace("\n", "")[:12] |
| 770 |
1 |
if text_id: |
| 771 |
0 |
return text_id |
| 772 |
0 |
except Exception: |
| 773 |
0 |
pass |
| 774 |
|
# Return a random id |
| 775 |
1 |
return "".join(random_choice("0123456789abcdef") for _ in range(12)) |
| 776 |
|
|
| 777 |
|
|
| 778 |
1 |
def usage(): |
| 779 |
0 |
print( |
| 780 |
|
"""Usage: {} [options] |
| 781 |
|
-c|--config [configuration_file]: loads the configuration file (default: ./lcm.cfg) |
| 782 |
|
--health-check: do not run lcm, but inspect kafka bus to determine if lcm is healthy |
| 783 |
|
-h|--help: shows this help |
| 784 |
|
""".format( |
| 785 |
|
sys.argv[0] |
| 786 |
|
) |
| 787 |
|
) |
| 788 |
|
# --log-socket-host HOST: send logs to this host") |
| 789 |
|
# --log-socket-port PORT: send logs using this port (default: 9022)") |
| 790 |
|
|
| 791 |
|
|
| 792 |
1 |
if __name__ == "__main__": |
| 793 |
0 |
try: |
| 794 |
|
# print("SYS.PATH='{}'".format(sys.path)) |
| 795 |
|
# load parameters and configuration |
| 796 |
|
# -h |
| 797 |
|
# -c value |
| 798 |
|
# --config value |
| 799 |
|
# --help |
| 800 |
|
# --health-check |
| 801 |
0 |
opts, args = getopt.getopt( |
| 802 |
|
sys.argv[1:], "hc:", ["config=", "help", "health-check"] |
| 803 |
|
) |
| 804 |
|
# TODO add "log-socket-host=", "log-socket-port=", "log-file=" |
| 805 |
0 |
config_file = None |
| 806 |
0 |
for o, a in opts: |
| 807 |
0 |
if o in ("-h", "--help"): |
| 808 |
0 |
usage() |
| 809 |
0 |
sys.exit() |
| 810 |
0 |
elif o in ("-c", "--config"): |
| 811 |
0 |
config_file = a |
| 812 |
0 |
elif o == "--health-check": |
| 813 |
0 |
from osm_lcm.lcm_hc import health_check |
| 814 |
|
|
| 815 |
0 |
health_check(config_file, Lcm.ping_interval_pace) |
| 816 |
|
# elif o == "--log-socket-port": |
| 817 |
|
# log_socket_port = a |
| 818 |
|
# elif o == "--log-socket-host": |
| 819 |
|
# log_socket_host = a |
| 820 |
|
# elif o == "--log-file": |
| 821 |
|
# log_file = a |
| 822 |
|
else: |
| 823 |
0 |
assert False, "Unhandled option" |
| 824 |
|
|
| 825 |
0 |
if config_file: |
| 826 |
0 |
if not path.isfile(config_file): |
| 827 |
0 |
print( |
| 828 |
|
"configuration file '{}' does not exist".format(config_file), |
| 829 |
|
file=sys.stderr, |
| 830 |
|
) |
| 831 |
0 |
exit(1) |
| 832 |
|
else: |
| 833 |
0 |
for config_file in ( |
| 834 |
|
__file__[: __file__.rfind(".")] + ".cfg", |
| 835 |
|
"./lcm.cfg", |
| 836 |
|
"/etc/osm/lcm.cfg", |
| 837 |
|
): |
| 838 |
0 |
if path.isfile(config_file): |
| 839 |
0 |
break |
| 840 |
|
else: |
| 841 |
0 |
print( |
| 842 |
|
"No configuration file 'lcm.cfg' found neither at local folder nor at /etc/osm/", |
| 843 |
|
file=sys.stderr, |
| 844 |
|
) |
| 845 |
0 |
exit(1) |
| 846 |
0 |
lcm = Lcm(config_file) |
| 847 |
0 |
lcm.start() |
| 848 |
0 |
except (LcmException, getopt.GetoptError) as e: |
| 849 |
0 |
print(str(e), file=sys.stderr) |
| 850 |
|
# usage() |
| 851 |
0 |
exit(1) |
![[Jenkins]](/jenkins/static/c06f6c43/images/jenkins-header-logo-v2.svg){kind=logo}
