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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775
|
# Copyright DataStax, Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import logging
import struct
import sys
import traceback
from cassandra import cqltypes
from cassandra import ConsistencyLevel, Unavailable, OperationTimedOut, ReadTimeout, ReadFailure, \
WriteTimeout, WriteFailure
from cassandra.cluster import NoHostAvailable, ExecutionProfile, EXEC_PROFILE_DEFAULT
from cassandra.concurrent import execute_concurrent_with_args
from cassandra.metadata import murmur3
from cassandra.policies import (
RoundRobinPolicy, DCAwareRoundRobinPolicy,
TokenAwarePolicy, WhiteListRoundRobinPolicy,
HostFilterPolicy
)
from cassandra.query import SimpleStatement
from tests.integration import use_singledc, use_multidc, remove_cluster, TestCluster, greaterthanorequalcass40, notdse
from tests.integration.long.utils import (wait_for_up, create_schema,
CoordinatorStats, force_stop,
wait_for_down, decommission, start,
bootstrap, stop, IP_FORMAT)
import unittest
log = logging.getLogger(__name__)
class LoadBalancingPolicyTests(unittest.TestCase):
def setUp(self):
remove_cluster() # clear ahead of test so it doesn't use one left in unknown state
self.coordinator_stats = CoordinatorStats()
self.prepared = None
self.probe_cluster = None
def tearDown(self):
if self.probe_cluster:
self.probe_cluster.shutdown()
@classmethod
def teardown_class(cls):
remove_cluster()
def _connect_probe_cluster(self):
if not self.probe_cluster:
# distinct cluster so we can see the status of nodes ignored by the LBP being tested
self.probe_cluster = TestCluster(
schema_metadata_enabled=False,
token_metadata_enabled=False,
execution_profiles={EXEC_PROFILE_DEFAULT: ExecutionProfile(load_balancing_policy=RoundRobinPolicy())}
)
self.probe_session = self.probe_cluster.connect()
def _wait_for_nodes_up(self, nodes, cluster=None):
log.debug('entered: _wait_for_nodes_up(nodes={ns}, '
'cluster={cs})'.format(ns=nodes,
cs=cluster))
if not cluster:
log.debug('connecting to cluster')
self._connect_probe_cluster()
cluster = self.probe_cluster
for n in nodes:
wait_for_up(cluster, n)
def _wait_for_nodes_down(self, nodes, cluster=None):
log.debug('entered: _wait_for_nodes_down(nodes={ns}, '
'cluster={cs})'.format(ns=nodes,
cs=cluster))
if not cluster:
self._connect_probe_cluster()
cluster = self.probe_cluster
for n in nodes:
wait_for_down(cluster, n)
def _cluster_session_with_lbp(self, lbp):
# create a cluster with no delay on events
cluster = TestCluster(topology_event_refresh_window=0, status_event_refresh_window=0,
execution_profiles={EXEC_PROFILE_DEFAULT: ExecutionProfile(load_balancing_policy=lbp)})
session = cluster.connect()
return cluster, session
def _insert(self, session, keyspace, count=12,
consistency_level=ConsistencyLevel.ONE):
log.debug('entered _insert('
'session={session}, keyspace={keyspace}, '
'count={count}, consistency_level={consistency_level}'
')'.format(session=session, keyspace=keyspace, count=count,
consistency_level=consistency_level))
session.execute('USE %s' % keyspace)
ss = SimpleStatement('INSERT INTO cf(k, i) VALUES (0, 0)', consistency_level=consistency_level)
tries = 0
while tries < 100:
try:
execute_concurrent_with_args(session, ss, [None] * count)
log.debug('Completed _insert on try #{}'.format(tries + 1))
return
except (OperationTimedOut, WriteTimeout, WriteFailure):
ex_type, ex, tb = sys.exc_info()
log.warning("{0}: {1} Backtrace: {2}".format(ex_type.__name__, ex, traceback.extract_tb(tb)))
del tb
tries += 1
raise RuntimeError("Failed to execute query after 100 attempts: {0}".format(ss))
def _query(self, session, keyspace, count=12,
consistency_level=ConsistencyLevel.ONE, use_prepared=False):
log.debug('entered _query('
'session={session}, keyspace={keyspace}, '
'count={count}, consistency_level={consistency_level}, '
'use_prepared={use_prepared}'
')'.format(session=session, keyspace=keyspace, count=count,
consistency_level=consistency_level,
use_prepared=use_prepared))
if use_prepared:
query_string = 'SELECT * FROM %s.cf WHERE k = ?' % keyspace
if not self.prepared or self.prepared.query_string != query_string:
self.prepared = session.prepare(query_string)
self.prepared.consistency_level = consistency_level
for i in range(count):
tries = 0
while True:
if tries > 100:
raise RuntimeError("Failed to execute query after 100 attempts: {0}".format(self.prepared))
try:
self.coordinator_stats.add_coordinator(session.execute_async(self.prepared.bind((0,))))
break
except (OperationTimedOut, ReadTimeout, ReadFailure):
ex_type, ex, tb = sys.exc_info()
log.warning("{0}: {1} Backtrace: {2}".format(ex_type.__name__, ex, traceback.extract_tb(tb)))
del tb
tries += 1
else:
routing_key = struct.pack('>i', 0)
for i in range(count):
ss = SimpleStatement('SELECT * FROM %s.cf WHERE k = 0' % keyspace,
consistency_level=consistency_level,
routing_key=routing_key)
tries = 0
while True:
if tries > 100:
raise RuntimeError("Failed to execute query after 100 attempts: {0}".format(ss))
try:
self.coordinator_stats.add_coordinator(session.execute_async(ss))
break
except (OperationTimedOut, ReadTimeout, ReadFailure):
ex_type, ex, tb = sys.exc_info()
log.warning("{0}: {1} Backtrace: {2}".format(ex_type.__name__, ex, traceback.extract_tb(tb)))
del tb
tries += 1
def test_token_aware_is_used_by_default(self):
"""
Test for default load balancing policy
test_token_aware_is_used_by_default tests that the default loadbalancing policy is policies.TokenAwarePolicy.
It creates a simple Cluster and verifies that the default loadbalancing policy is TokenAwarePolicy if the
murmur3 C extension is found. Otherwise, the default loadbalancing policy is DCAwareRoundRobinPolicy.
@since 2.6.0
@jira_ticket PYTHON-160
@expected_result TokenAwarePolicy should be the default loadbalancing policy.
@test_category load_balancing:token_aware
"""
cluster = TestCluster()
self.addCleanup(cluster.shutdown)
if murmur3 is not None:
self.assertTrue(isinstance(cluster.profile_manager.default.load_balancing_policy, TokenAwarePolicy))
else:
self.assertTrue(isinstance(cluster.profile_manager.default.load_balancing_policy, DCAwareRoundRobinPolicy))
def test_roundrobin(self):
use_singledc()
keyspace = 'test_roundrobin'
cluster, session = self._cluster_session_with_lbp(RoundRobinPolicy())
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 4), cluster)
create_schema(cluster, session, keyspace, replication_factor=3)
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 4)
self.coordinator_stats.assert_query_count_equals(self, 2, 4)
self.coordinator_stats.assert_query_count_equals(self, 3, 4)
force_stop(3)
self._wait_for_nodes_down([3], cluster)
self.coordinator_stats.reset_counts()
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 6)
self.coordinator_stats.assert_query_count_equals(self, 2, 6)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
decommission(1)
start(3)
self._wait_for_nodes_down([1], cluster)
self._wait_for_nodes_up([3], cluster)
self.coordinator_stats.reset_counts()
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 6)
self.coordinator_stats.assert_query_count_equals(self, 3, 6)
def test_roundrobin_two_dcs(self):
use_multidc([2, 2])
keyspace = 'test_roundrobin_two_dcs'
cluster, session = self._cluster_session_with_lbp(RoundRobinPolicy())
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 5), cluster)
create_schema(cluster, session, keyspace, replication_strategy=[2, 2])
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 3)
self.coordinator_stats.assert_query_count_equals(self, 2, 3)
self.coordinator_stats.assert_query_count_equals(self, 3, 3)
self.coordinator_stats.assert_query_count_equals(self, 4, 3)
force_stop(1)
bootstrap(5, 'dc3')
# reset control connection
self._insert(session, keyspace, count=1000)
self._wait_for_nodes_up([5], cluster)
self.coordinator_stats.reset_counts()
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 3)
self.coordinator_stats.assert_query_count_equals(self, 3, 3)
self.coordinator_stats.assert_query_count_equals(self, 4, 3)
self.coordinator_stats.assert_query_count_equals(self, 5, 3)
def test_roundrobin_two_dcs_2(self):
use_multidc([2, 2])
keyspace = 'test_roundrobin_two_dcs_2'
cluster, session = self._cluster_session_with_lbp(RoundRobinPolicy())
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 5), cluster)
create_schema(cluster, session, keyspace, replication_strategy=[2, 2])
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 3)
self.coordinator_stats.assert_query_count_equals(self, 2, 3)
self.coordinator_stats.assert_query_count_equals(self, 3, 3)
self.coordinator_stats.assert_query_count_equals(self, 4, 3)
force_stop(1)
bootstrap(5, 'dc1')
# reset control connection
self._insert(session, keyspace, count=1000)
self._wait_for_nodes_up([5], cluster)
self.coordinator_stats.reset_counts()
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 3)
self.coordinator_stats.assert_query_count_equals(self, 3, 3)
self.coordinator_stats.assert_query_count_equals(self, 4, 3)
self.coordinator_stats.assert_query_count_equals(self, 5, 3)
def test_dc_aware_roundrobin_two_dcs(self):
use_multidc([3, 2])
keyspace = 'test_dc_aware_roundrobin_two_dcs'
cluster, session = self._cluster_session_with_lbp(DCAwareRoundRobinPolicy('dc1'))
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 6))
create_schema(cluster, session, keyspace, replication_strategy=[2, 2])
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 4)
self.coordinator_stats.assert_query_count_equals(self, 2, 4)
self.coordinator_stats.assert_query_count_equals(self, 3, 4)
self.coordinator_stats.assert_query_count_equals(self, 4, 0)
self.coordinator_stats.assert_query_count_equals(self, 5, 0)
def test_dc_aware_roundrobin_two_dcs_2(self):
use_multidc([3, 2])
keyspace = 'test_dc_aware_roundrobin_two_dcs_2'
cluster, session = self._cluster_session_with_lbp(DCAwareRoundRobinPolicy('dc2'))
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 6))
create_schema(cluster, session, keyspace, replication_strategy=[2, 2])
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.assert_query_count_equals(self, 4, 6)
self.coordinator_stats.assert_query_count_equals(self, 5, 6)
def test_dc_aware_roundrobin_one_remote_host(self):
use_multidc([2, 2])
keyspace = 'test_dc_aware_roundrobin_one_remote_host'
cluster, session = self._cluster_session_with_lbp(DCAwareRoundRobinPolicy('dc2', used_hosts_per_remote_dc=1))
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 5))
create_schema(cluster, session, keyspace, replication_strategy=[2, 2])
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
self.coordinator_stats.assert_query_count_equals(self, 3, 6)
self.coordinator_stats.assert_query_count_equals(self, 4, 6)
self.coordinator_stats.reset_counts()
bootstrap(5, 'dc1')
self._wait_for_nodes_up([5])
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
self.coordinator_stats.assert_query_count_equals(self, 3, 6)
self.coordinator_stats.assert_query_count_equals(self, 4, 6)
self.coordinator_stats.assert_query_count_equals(self, 5, 0)
self.coordinator_stats.reset_counts()
decommission(3)
decommission(4)
self._wait_for_nodes_down([3, 4])
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.assert_query_count_equals(self, 4, 0)
responses = set()
for node in [1, 2, 5]:
responses.add(self.coordinator_stats.get_query_count(node))
self.assertEqual(set([0, 0, 12]), responses)
self.coordinator_stats.reset_counts()
decommission(5)
self._wait_for_nodes_down([5])
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.assert_query_count_equals(self, 4, 0)
self.coordinator_stats.assert_query_count_equals(self, 5, 0)
responses = set()
for node in [1, 2]:
responses.add(self.coordinator_stats.get_query_count(node))
self.assertEqual(set([0, 12]), responses)
self.coordinator_stats.reset_counts()
decommission(1)
self._wait_for_nodes_down([1])
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 12)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.assert_query_count_equals(self, 4, 0)
self.coordinator_stats.assert_query_count_equals(self, 5, 0)
self.coordinator_stats.reset_counts()
force_stop(2)
try:
self._query(session, keyspace)
self.fail()
except NoHostAvailable:
pass
def test_token_aware(self):
keyspace = 'test_token_aware'
self.token_aware(keyspace)
def test_token_aware_prepared(self):
keyspace = 'test_token_aware_prepared'
self.token_aware(keyspace, True)
def token_aware(self, keyspace, use_prepared=False):
use_singledc()
cluster, session = self._cluster_session_with_lbp(TokenAwarePolicy(RoundRobinPolicy()))
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 4), cluster)
create_schema(cluster, session, keyspace, replication_factor=1)
self._insert(session, keyspace)
self._query(session, keyspace, use_prepared=use_prepared)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 12)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.reset_counts()
self._query(session, keyspace, use_prepared=use_prepared)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 12)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.reset_counts()
force_stop(2)
self._wait_for_nodes_down([2], cluster)
try:
self._query(session, keyspace, use_prepared=use_prepared)
self.fail()
except Unavailable as e:
self.assertEqual(e.consistency, 1)
self.assertEqual(e.required_replicas, 1)
self.assertEqual(e.alive_replicas, 0)
self.coordinator_stats.reset_counts()
start(2)
self._wait_for_nodes_up([2], cluster)
self._query(session, keyspace, use_prepared=use_prepared)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 12)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.reset_counts()
stop(2)
self._wait_for_nodes_down([2], cluster)
try:
self._query(session, keyspace, use_prepared=use_prepared)
self.fail()
except Unavailable:
pass
self.coordinator_stats.reset_counts()
start(2)
self._wait_for_nodes_up([2], cluster)
decommission(2)
self._wait_for_nodes_down([2], cluster)
self._query(session, keyspace, use_prepared=use_prepared)
results = set([
self.coordinator_stats.get_query_count(1),
self.coordinator_stats.get_query_count(3)
])
self.assertEqual(results, set([0, 12]))
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
def test_token_aware_composite_key(self):
use_singledc()
keyspace = 'test_token_aware_composite_key'
table = 'composite'
cluster, session = self._cluster_session_with_lbp(TokenAwarePolicy(RoundRobinPolicy()))
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 4), cluster)
create_schema(cluster, session, keyspace, replication_factor=2)
session.execute('CREATE TABLE %s ('
'k1 int, '
'k2 int, '
'i int, '
'PRIMARY KEY ((k1, k2)))' % table)
prepared = session.prepare('INSERT INTO %s '
'(k1, k2, i) '
'VALUES '
'(?, ?, ?)' % table)
bound = prepared.bind((1, 2, 3))
result = session.execute(bound)
self.assertIn(result.response_future.attempted_hosts[0],
cluster.metadata.get_replicas(keyspace, bound.routing_key))
# There could be race condition with querying a node
# which doesn't yet have the data so we query one of
# the replicas
results = session.execute(SimpleStatement('SELECT * FROM %s WHERE k1 = 1 AND k2 = 2' % table,
routing_key=bound.routing_key))
self.assertIn(results.response_future.attempted_hosts[0],
cluster.metadata.get_replicas(keyspace, bound.routing_key))
self.assertTrue(results[0].i)
def test_token_aware_with_rf_2(self, use_prepared=False):
use_singledc()
keyspace = 'test_token_aware_with_rf_2'
cluster, session = self._cluster_session_with_lbp(TokenAwarePolicy(RoundRobinPolicy()))
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 4), cluster)
create_schema(cluster, session, keyspace, replication_factor=2)
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 12)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
self.coordinator_stats.reset_counts()
stop(2)
self._wait_for_nodes_down([2], cluster)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
self.coordinator_stats.assert_query_count_equals(self, 3, 12)
def test_token_aware_with_local_table(self):
use_singledc()
cluster, session = self._cluster_session_with_lbp(TokenAwarePolicy(RoundRobinPolicy()))
self.addCleanup(cluster.shutdown)
self._wait_for_nodes_up(range(1, 4), cluster)
p = session.prepare("SELECT * FROM system.local WHERE key=?")
# this would blow up prior to 61b4fad
r = session.execute(p, ('local',))
self.assertEqual(r[0].key, 'local')
def test_token_aware_with_shuffle_rf2(self):
"""
Test to validate the hosts are shuffled when the `shuffle_replicas` is truthy
@since 3.8
@jira_ticket PYTHON-676
@expected_result the request are spread across the replicas,
when one of them is down, the requests target the available one
@test_category policy
"""
keyspace = 'test_token_aware_with_rf_2'
cluster, session = self._set_up_shuffle_test(keyspace, replication_factor=2)
self.addCleanup(cluster.shutdown)
self._check_query_order_changes(session=session, keyspace=keyspace)
# check TokenAwarePolicy still return the remaining replicas when one goes down
self.coordinator_stats.reset_counts()
stop(2)
self._wait_for_nodes_down([2], cluster)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
self.coordinator_stats.assert_query_count_equals(self, 3, 12)
def test_token_aware_with_shuffle_rf3(self):
"""
Test to validate the hosts are shuffled when the `shuffle_replicas` is truthy
@since 3.8
@jira_ticket PYTHON-676
@expected_result the request are spread across the replicas,
when one of them is down, the requests target the other available ones
@test_category policy
"""
keyspace = 'test_token_aware_with_rf_3'
cluster, session = self._set_up_shuffle_test(keyspace, replication_factor=3)
self.addCleanup(cluster.shutdown)
self._check_query_order_changes(session=session, keyspace=keyspace)
# check TokenAwarePolicy still return the remaining replicas when one goes down
self.coordinator_stats.reset_counts()
stop(1)
self._wait_for_nodes_down([1], cluster)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
query_count_two = self.coordinator_stats.get_query_count(2)
query_count_three = self.coordinator_stats.get_query_count(3)
self.assertEqual(query_count_two + query_count_three, 12)
self.coordinator_stats.reset_counts()
stop(2)
self._wait_for_nodes_down([2], cluster)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
self.coordinator_stats.assert_query_count_equals(self, 3, 12)
@notdse
@greaterthanorequalcass40
def test_token_aware_with_transient_replication(self):
"""
Test to validate that the token aware policy doesn't route any request to a transient node.
@since 3.23
@jira_ticket PYTHON-1207
@expected_result the requests are spread across the 2 full replicas and
no other nodes are queried by the coordinator.
@test_category policy
"""
# We can test this with a single dc when CASSANDRA-15670 is fixed
use_multidc([3, 3])
cluster, session = self._cluster_session_with_lbp(
TokenAwarePolicy(DCAwareRoundRobinPolicy(), shuffle_replicas=True)
)
self.addCleanup(cluster.shutdown)
session.execute("CREATE KEYSPACE test_tr WITH replication = {'class': 'NetworkTopologyStrategy', 'dc1': '3/1', 'dc2': '3/1'};")
session.execute("CREATE TABLE test_tr.users (id int PRIMARY KEY, username text) WITH read_repair ='NONE';")
for i in range(100):
session.execute("INSERT INTO test_tr.users (id, username) VALUES (%d, 'user');" % (i,))
query = session.prepare("SELECT * FROM test_tr.users WHERE id = ?")
for i in range(100):
f = session.execute_async(query, (i,), trace=True)
full_dc1_replicas = [h for h in cluster.metadata.get_replicas('test_tr', cqltypes.Int32Type.serialize(i, cluster.protocol_version))
if h.datacenter == 'dc1']
self.assertEqual(len(full_dc1_replicas), 2)
f.result()
trace_hosts = [cluster.metadata.get_host(e.source) for e in f.get_query_trace().events]
for h in f.attempted_hosts:
self.assertIn(h, full_dc1_replicas)
for h in trace_hosts:
self.assertIn(h, full_dc1_replicas)
def _set_up_shuffle_test(self, keyspace, replication_factor):
use_singledc()
cluster, session = self._cluster_session_with_lbp(
TokenAwarePolicy(RoundRobinPolicy(), shuffle_replicas=True)
)
self._wait_for_nodes_up(range(1, 4), cluster)
create_schema(cluster, session, keyspace, replication_factor=replication_factor)
return cluster, session
def _check_query_order_changes(self, session, keyspace):
LIMIT_TRIES, tried, query_counts = 20, 0, set()
while len(query_counts) <= 1:
tried += 1
if tried >= LIMIT_TRIES:
raise Exception("After {0} tries shuffle returned the same output".format(LIMIT_TRIES))
self._insert(session, keyspace)
self._query(session, keyspace)
loop_qcs = (self.coordinator_stats.get_query_count(1),
self.coordinator_stats.get_query_count(2),
self.coordinator_stats.get_query_count(3))
query_counts.add(loop_qcs)
self.assertEqual(sum(loop_qcs), 12)
# end the loop if we get more than one query ordering
self.coordinator_stats.reset_counts()
def test_white_list(self):
use_singledc()
keyspace = 'test_white_list'
cluster = TestCluster(
contact_points=('127.0.0.2',), topology_event_refresh_window=0, status_event_refresh_window=0,
execution_profiles={
EXEC_PROFILE_DEFAULT: ExecutionProfile(
load_balancing_policy=WhiteListRoundRobinPolicy((IP_FORMAT % 2,))
)
}
)
self.addCleanup(cluster.shutdown)
session = cluster.connect()
self._wait_for_nodes_up([1, 2, 3])
create_schema(cluster, session, keyspace)
self._insert(session, keyspace)
self._query(session, keyspace)
self.coordinator_stats.assert_query_count_equals(self, 1, 0)
self.coordinator_stats.assert_query_count_equals(self, 2, 12)
self.coordinator_stats.assert_query_count_equals(self, 3, 0)
# white list policy should not allow reconnecting to ignored hosts
force_stop(3)
self._wait_for_nodes_down([3])
self.assertFalse(cluster.metadata.get_host(IP_FORMAT % 3).is_currently_reconnecting())
self.coordinator_stats.reset_counts()
force_stop(2)
self._wait_for_nodes_down([2])
try:
self._query(session, keyspace)
self.fail()
except NoHostAvailable:
pass
def test_black_list_with_host_filter_policy(self):
"""
Test to validate removing certain hosts from the query plan with
HostFilterPolicy
@since 3.8
@jira_ticket PYTHON-961
@expected_result the excluded hosts are ignored
@test_category policy
"""
use_singledc()
keyspace = 'test_black_list_with_hfp'
ignored_address = (IP_FORMAT % 2)
hfp = HostFilterPolicy(
child_policy=RoundRobinPolicy(),
predicate=lambda host: host.address != ignored_address
)
cluster = TestCluster(
contact_points=(IP_FORMAT % 1,),
topology_event_refresh_window=0,
status_event_refresh_window=0,
execution_profiles={EXEC_PROFILE_DEFAULT: ExecutionProfile(load_balancing_policy=hfp)}
)
self.addCleanup(cluster.shutdown)
session = cluster.connect()
self._wait_for_nodes_up([1, 2, 3])
self.assertNotIn(ignored_address, [h.address for h in hfp.make_query_plan()])
create_schema(cluster, session, keyspace)
self._insert(session, keyspace)
self._query(session, keyspace)
# RoundRobin doesn't provide a gurantee on the order of the hosts
# so we will have that for 127.0.0.1 and 127.0.0.3 the count for one
# will be 4 and for the other 8
first_node_count = self.coordinator_stats.get_query_count(1)
third_node_count = self.coordinator_stats.get_query_count(3)
self.assertEqual(first_node_count + third_node_count, 12)
self.assertTrue(first_node_count == 8 or first_node_count == 4)
self.coordinator_stats.assert_query_count_equals(self, 2, 0)
# policy should not allow reconnecting to ignored host
force_stop(2)
self._wait_for_nodes_down([2])
self.assertFalse(cluster.metadata.get_host(ignored_address).is_currently_reconnecting())
|