-
Notifications
You must be signed in to change notification settings - Fork 302
/
rule.py
3253 lines (2810 loc) · 178 KB
/
rule.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
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
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# -*- coding: utf-8 -*-
# Copyright 2012-2020 CERN
#
# 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.
#
# Authors:
# - Vincent Garonne <vincent.garonne@cern.ch>, 2012-2018
# - Mario Lassnig <mario.lassnig@cern.ch>, 2013-2019
# - Martin Barisits <martin.barisits@cern.ch>, 2013-2020
# - Cedric Serfon <cedric.serfon@cern.ch>, 2014-2020
# - David Cameron <david.cameron@cern.ch>, 2014
# - Joaquín Bogado <jbogado@linti.unlp.edu.ar>, 2014-2018
# - Thomas Beermann <thomas.beermann@cern.ch>, 2014-2015
# - Hannes Hansen <hannes.jakob.hansen@cern.ch>, 2018-2019
# - Robert Illingworth <illingwo@fnal.gov>, 2019
# - Andrew Lister <andrew.lister@stfc.ac.uk>, 2019
# - Dimitrios Christidis <dimitrios.christidis@cern.ch>, 2019
# - Brandon White <bjwhite@fnal.gov>, 2019
# - Luc Goossens <luc.goossens@cern.ch>, 2020
# - Eli Chadwick <eli.chadwick@stfc.ac.uk>, 2020
# - Patrick Austin <patrick.austin@stfc.ac.uk>, 2020
# - Benedikt Ziemons <benedikt.ziemons@cern.ch>, 2020
# - Eric Vaandering <ewv@fnal.gov>, 2020
from __future__ import division
import json
import logging
import sys
try:
from ConfigParser import NoOptionError
except ImportError:
from configparser import NoOptionError
from copy import deepcopy
from datetime import datetime, timedelta
from re import match
from string import Template
from dogpile.cache import make_region
from dogpile.cache.api import NO_VALUE
from six import string_types
from sqlalchemy.exc import IntegrityError, StatementError
from sqlalchemy.orm.exc import NoResultFound
from sqlalchemy.sql import func
from sqlalchemy.sql.expression import and_, or_, text, true, null, tuple_
from rucio.core.account import has_account_attribute
from rucio.core.config import get as core_config_get
import rucio.core.did
import rucio.core.lock # import get_replica_locks, get_files_and_replica_locks_of_dataset
import rucio.core.replica # import get_and_lock_file_replicas, get_and_lock_file_replicas_for_dataset
from rucio.common.policy import policy_filter, get_scratchdisk_lifetime
from rucio.common.config import config_get
from rucio.common.exception import (InvalidRSEExpression, InvalidReplicationRule, InsufficientAccountLimit,
DataIdentifierNotFound, RuleNotFound, InputValidationError, RSEOverQuota,
ReplicationRuleCreationTemporaryFailed, InsufficientTargetRSEs, RucioException,
InvalidRuleWeight, StagingAreaRuleRequiresLifetime, DuplicateRule,
InvalidObject, RSEBlacklisted, RSEWriteBlocked, RuleReplaceFailed, RequestNotFound,
ManualRuleApprovalBlocked, UnsupportedOperation, UndefinedPolicy)
from rucio.common.schema import validate_schema
from rucio.common.types import InternalScope, InternalAccount
from rucio.common.utils import str_to_date, sizefmt, chunks
from rucio.core import account_counter, rse_counter, request as request_core
from rucio.core.account import get_account
from rucio.core.lifetime_exception import define_eol
from rucio.core.message import add_message
from rucio.core.monitor import record_timer_block
from rucio.core.rse import get_rse_name, list_rse_attributes, get_rse, get_rse_usage
from rucio.core.rse_expression_parser import parse_expression
from rucio.core.rse_selector import RSESelector
from rucio.core.rule_grouping import apply_rule_grouping, repair_stuck_locks_and_apply_rule_grouping, create_transfer_dict, apply_rule
from rucio.db.sqla import models, filter_thread_work
from rucio.db.sqla.constants import (LockState, ReplicaState, RuleState, RuleGrouping,
DIDAvailability, DIDReEvaluation, DIDType,
RequestType, RuleNotification, OBSOLETE, RSEType)
from rucio.db.sqla.session import read_session, transactional_session, stream_session
from rucio.extensions.forecast import T3CModel
# module variable to activate use of new rule algorithm (rucio.core.rule_grouping.apply_rule)
USE_NEW_RULE_ALGORITHM = False
logging.basicConfig(stream=sys.stdout,
level=getattr(logging,
config_get('common', 'loglevel',
raise_exception=False,
default='DEBUG').upper()),
format='%(asctime)s\t%(process)d\t%(levelname)s\t%(message)s')
REGION = make_region().configure('dogpile.cache.memcached',
expiration_time=3600,
arguments={'url': config_get('cache', 'url', False, '127.0.0.1:11211'),
'distributed_lock': True})
@transactional_session
def add_rule(dids, account, copies, rse_expression, grouping, weight, lifetime, locked, subscription_id,
source_replica_expression=None, activity='User Subscriptions', notify=None, purge_replicas=False,
ignore_availability=False, comment=None, ask_approval=False, asynchronous=False, ignore_account_limit=False,
priority=3, split_container=False, meta=None, session=None):
"""
Adds a replication rule for every did in dids
:param dids: List of data identifiers.
:param account: Account issuing the rule.
:param copies: The number of replicas.
:param rse_expression: RSE expression which gets resolved into a list of rses.
:param grouping: ALL - All files will be replicated to the same RSE.
DATASET - All files in the same dataset will be replicated to the same RSE.
NONE - Files will be completely spread over all allowed RSEs without any grouping considerations at all.
:param weight: Weighting scheme to be used.
:param lifetime: The lifetime of the replication rule in seconds.
:param locked: If the rule is locked.
:param subscription_id: The subscription_id, if the rule is created by a subscription.
:param source_replica_expression: Only use replicas as source from this RSEs.
:param activity: Activity to be passed on to the conveyor.
:param notify: Notification setting of the rule ('Y', 'N', 'C'; None = 'N').
:param purge_replicas: Purge setting if a replica should be directly deleted after the rule is deleted.
:param ignore_availability: Option to ignore the availability of RSEs.
:param comment: Comment about the rule.
:param ask_approval: Ask for approval for this rule.
:param asynchronous: Create replication rule asynchronously by the judge-injector.
:param ignore_account_limit: Ignore quota and create the rule outside of the account limits.
:param priority: Priority of the rule and the transfers which should be submitted.
:param split_container: Should a container rule be split into individual dataset rules.
:param meta: Dictionary with metadata from the WFMS.
:param session: The database session in use.
:returns: A list of created replication rule ids.
:raises: InvalidReplicationRule, InsufficientAccountLimit, InvalidRSEExpression, DataIdentifierNotFound, ReplicationRuleCreationTemporaryFailed, InvalidRuleWeight,
StagingAreaRuleRequiresLifetime, DuplicateRule, RSEBlacklisted, ScratchDiskLifetimeConflict, ManualRuleApprovalBlocked, RSEOverQuota
"""
rule_ids = []
grouping = {'ALL': RuleGrouping.ALL, 'NONE': RuleGrouping.NONE}.get(grouping, RuleGrouping.DATASET)
if USE_NEW_RULE_ALGORITHM:
use_new_rule_algorithm = True
else:
use_new_rule_algorithm = core_config_get('rules', 'use_new_rule_algorithm', default=False, session=session)
with record_timer_block('rule.add_rule'):
# 1. Resolve the rse_expression into a list of RSE-ids
with record_timer_block('rule.add_rule.parse_rse_expression'):
vo = account.vo
if ignore_availability:
rses = parse_expression(rse_expression, filter={'vo': vo}, session=session)
else:
rses = parse_expression(rse_expression, filter={'vo': vo, 'availability_write': True}, session=session)
if lifetime is None: # Check if one of the rses is a staging area
if [rse for rse in rses if rse.get('staging_area', False)]:
raise StagingAreaRuleRequiresLifetime('Rules for a staging area must include a lifetime')
# Check SCRATCHDISK Policy
try:
lifetime = get_scratch_policy(account, rses, lifetime, session=session)
except UndefinedPolicy:
pass
# Auto-lock rules for TAPE rses
if not locked and lifetime is None:
if [rse for rse in rses if rse.get('rse_type', RSEType.DISK) == RSEType.TAPE]:
locked = True
# Block manual approval if RSE does not allow it
if ask_approval:
for rse in rses:
if list_rse_attributes(rse_id=rse['id'], session=session).get('block_manual_approval', False):
raise ManualRuleApprovalBlocked()
if source_replica_expression:
source_rses = parse_expression(source_replica_expression, filter={'vo': vo}, session=session)
else:
source_rses = []
# 2. Create the rse selector
with record_timer_block('rule.add_rule.create_rse_selector'):
rseselector = RSESelector(account=account, rses=rses, weight=weight, copies=copies, ignore_account_limit=ask_approval or ignore_account_limit, session=session)
expires_at = datetime.utcnow() + timedelta(seconds=lifetime) if lifetime is not None else None
notify = {'Y': RuleNotification.YES, 'C': RuleNotification.CLOSE, 'P': RuleNotification.PROGRESS}.get(notify, RuleNotification.NO)
for elem in dids:
# 3. Get the did
with record_timer_block('rule.add_rule.get_did'):
try:
did = session.query(models.DataIdentifier).filter(models.DataIdentifier.scope == elem['scope'],
models.DataIdentifier.name == elem['name']).one()
except NoResultFound:
raise DataIdentifierNotFound('Data identifier %s:%s is not valid.' % (elem['scope'], elem['name']))
except TypeError as error:
raise InvalidObject(error.args)
# 3.1 If the did is a constituent, relay the rule to the archive
if did.did_type == DIDType.FILE and did.constituent:
# Check if a single replica of this DID exists; Do not put rule on file if there are only replicas on TAPE
replica_cnt = session.query(models.RSEFileAssociation).join(models.RSE, models.RSEFileAssociation.rse_id == models.RSE.id)\
.filter(models.RSEFileAssociation.scope == did.scope,
models.RSEFileAssociation.name == did.name,
models.RSEFileAssociation.state == ReplicaState.AVAILABLE,
models.RSE.rse_type != RSEType.TAPE).count()
if replica_cnt == 0: # Put the rule on the archive
archive = session.query(models.ConstituentAssociation).join(models.RSEFileAssociation,
and_(models.ConstituentAssociation.scope == models.RSEFileAssociation.scope,
models.ConstituentAssociation.name == models.RSEFileAssociation.name))\
.filter(models.ConstituentAssociation.child_scope == did.scope,
models.ConstituentAssociation.child_name == did.name).first()
if archive is not None:
elem['name'] = archive.name
elem['scope'] = archive.scope
try:
did = session.query(models.DataIdentifier).filter(models.DataIdentifier.scope == elem['scope'],
models.DataIdentifier.name == elem['name']).one()
except NoResultFound:
raise DataIdentifierNotFound('Data identifier %s:%s is not valid.' % (elem['scope'], elem['name']))
except TypeError as error:
raise InvalidObject(error.args)
else: # Put the rule on the constituent directly
pass
# 3.2 Get the lifetime
eol_at = define_eol(elem['scope'], elem['name'], rses, session=session)
# 4. Create the replication rule
with record_timer_block('rule.add_rule.create_rule'):
if meta is not None:
try:
meta = json.dumps(meta)
except Exception:
meta = None
new_rule = models.ReplicationRule(account=account,
name=elem['name'],
scope=elem['scope'],
did_type=did.did_type,
copies=copies,
rse_expression=rse_expression,
locked=locked,
grouping=grouping,
expires_at=expires_at,
weight=weight,
source_replica_expression=source_replica_expression,
activity=activity,
subscription_id=subscription_id,
notification=notify,
purge_replicas=purge_replicas,
ignore_availability=ignore_availability,
comments=comment,
ignore_account_limit=ignore_account_limit,
priority=priority,
split_container=split_container,
meta=meta,
eol_at=eol_at)
try:
new_rule.save(session=session)
except IntegrityError as error:
if match('.*ORA-00001.*', str(error.args[0])) \
or match('.*IntegrityError.*UNIQUE constraint failed.*', str(error.args[0])) \
or match('.*1062.*Duplicate entry.*for key.*', str(error.args[0])) \
or match('.*IntegrityError.*duplicate key value violates unique constraint.*', error.args[0]) \
or match('.*UniqueViolation.*duplicate key value violates unique constraint.*', error.args[0]) \
or match('.*IntegrityError.*columns? .*not unique.*', error.args[0]):
raise DuplicateRule(error.args[0])
raise InvalidReplicationRule(error.args[0])
rule_ids.append(new_rule.id)
if ask_approval:
new_rule.state = RuleState.WAITING_APPROVAL
# Block manual approval for multi-rse rules
if len(rses) > 1:
raise InvalidReplicationRule('Ask approval is not allowed for rules with multiple RSEs')
if len(rses) == 1 and not did.is_open and did.bytes is not None and did.length is not None:
# This rule can be considered for auto-approval:
rse_attr = list_rse_attributes(rse_id=rses[0]['id'], session=session)
auto_approve = False
if 'auto_approve_bytes' in rse_attr and 'auto_approve_files' in rse_attr:
if did.bytes < int(rse_attr.get('auto_approve_bytes')) and did.length < int(rse_attr.get('auto_approve_bytes')):
auto_approve = True
elif did.bytes < int(rse_attr.get('auto_approve_bytes', -1)):
auto_approve = True
elif did.length < int(rse_attr.get('auto_approve_files', -1)):
auto_approve = True
if auto_approve:
logging.debug("Auto approving rule %s", str(new_rule.id))
logging.debug("Created rule %s for injection", str(new_rule.id))
approve_rule(rule_id=new_rule.id, notify_approvers=False, session=session)
continue
logging.debug("Created rule %s in waiting for approval", str(new_rule.id))
__create_rule_approval_email(rule=new_rule, session=session)
continue
# Force ASYNC mode for large rules
if did.length is not None and (did.length * copies) >= 10000:
asynchronous = True
logging.debug("Forced injection of rule %s", str(new_rule.id))
if asynchronous:
# TODO: asynchronous mode only available for closed dids (on the whole tree?)
new_rule.state = RuleState.INJECT
logging.debug("Created rule %s for injection", str(new_rule.id))
continue
# If Split Container is chosen, the rule will be processed ASYNC
if split_container and did.did_type == DIDType.CONTAINER:
new_rule.state = RuleState.INJECT
logging.debug("Created rule %s for injection due to Split Container mode", str(new_rule.id))
continue
if use_new_rule_algorithm:
# 5. Apply the rule
with record_timer_block('rule.add_rule.apply_rule'):
try:
apply_rule(did, new_rule, [x['id'] for x in rses], [x['id'] for x in source_rses], rseselector, session=session)
except IntegrityError as error:
raise ReplicationRuleCreationTemporaryFailed(error.args[0])
else:
# 5. Resolve the did to its contents
with record_timer_block('rule.add_rule.resolve_dids_to_locks_replicas'):
# Get all Replicas, not only the ones interesting for the rse_expression
datasetfiles, locks, replicas, source_replicas = __resolve_did_to_locks_and_replicas(did=did,
nowait=False,
restrict_rses=[rse['id'] for rse in rses],
source_rses=[rse['id'] for rse in source_rses],
session=session)
sumfiles = sum([len(x['files']) for x in datasetfiles])
if sumfiles > 30000:
logging.warning('Rule %s for %s:%s involves %d files', str(new_rule.id), new_rule.scope, new_rule.name, sumfiles)
# 6. Apply the replication rule to create locks, replicas and transfers
with record_timer_block('rule.add_rule.create_locks_replicas_transfers'):
try:
__create_locks_replicas_transfers(datasetfiles=datasetfiles,
locks=locks,
replicas=replicas,
source_replicas=source_replicas,
rseselector=rseselector,
rule=new_rule,
preferred_rse_ids=[],
source_rses=[rse['id'] for rse in source_rses],
session=session)
except IntegrityError as error:
raise ReplicationRuleCreationTemporaryFailed(error.args[0])
if new_rule.locks_stuck_cnt > 0:
new_rule.state = RuleState.STUCK
new_rule.error = 'MissingSourceReplica'
if new_rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=new_rule.id).update({'state': LockState.STUCK})
elif new_rule.locks_replicating_cnt == 0:
new_rule.state = RuleState.OK
if new_rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=new_rule.id).update({'state': LockState.OK})
session.flush()
if new_rule.notification == RuleNotification.YES:
generate_email_for_rule_ok_notification(rule=new_rule, session=session)
generate_rule_notifications(rule=new_rule, replicating_locks_before=0, session=session)
else:
new_rule.state = RuleState.REPLICATING
if new_rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=new_rule.id).update({'state': LockState.REPLICATING})
# Add rule to History
insert_rule_history(rule=new_rule, recent=True, longterm=True, session=session)
if use_new_rule_algorithm:
logging.info("Created rule %s [%d/%d/%d] with new algorithm for did %s:%s in state %s", str(new_rule.id), new_rule.locks_ok_cnt, new_rule.locks_replicating_cnt, new_rule.locks_stuck_cnt, new_rule.scope, new_rule.name, str(new_rule.state))
else:
logging.info("Created rule %s [%d/%d/%d] for did %s:%s in state %s", str(new_rule.id), new_rule.locks_ok_cnt, new_rule.locks_replicating_cnt, new_rule.locks_stuck_cnt, new_rule.scope, new_rule.name, str(new_rule.state))
return rule_ids
@transactional_session
def add_rules(dids, rules, session=None):
"""
Adds a list of replication rules to every did in dids
:params dids: List of data identifiers.
:param rules: List of dictionaries defining replication rules.
{account, copies, rse_expression, grouping, weight, lifetime, locked, subscription_id, source_replica_expression, activity, notifiy, purge_replicas}
:param session: The database session in use.
:returns: Dictionary (scope, name) with list of created rule ids
:raises: InvalidReplicationRule, InsufficientAccountLimit, InvalidRSEExpression, DataIdentifierNotFound, ReplicationRuleCreationTemporaryFailed, InvalidRuleWeight,
StagingAreaRuleRequiresLifetime, DuplicateRule, RSEBlacklisted, ScratchDiskLifetimeConflict, ManualRuleApprovalBlocked
"""
with record_timer_block('rule.add_rules'):
rule_ids = {}
# 1. Fetch the RSEs from the RSE expression to restrict further queries just on these RSEs
restrict_rses = []
all_source_rses = []
with record_timer_block('rule.add_rules.parse_rse_expressions'):
for rule in rules:
vo = rule['account'].vo
if rule.get('ignore_availability'):
restrict_rses.extend(parse_expression(rule['rse_expression'], filter={'vo': vo}, session=session))
else:
restrict_rses.extend(parse_expression(rule['rse_expression'], filter={'vo': vo, 'availability_write': True}, session=session))
restrict_rses = list(set([rse['id'] for rse in restrict_rses]))
for rule in rules:
if rule.get('source_replica_expression'):
vo = rule['account'].vo
all_source_rses.extend(parse_expression(rule.get('source_replica_expression'), filter={'vo': vo}, session=session))
all_source_rses = list(set([rse['id'] for rse in all_source_rses]))
for elem in dids:
# 2. Get the did
with record_timer_block('rule.add_rules.get_did'):
try:
did = session.query(models.DataIdentifier).filter(
models.DataIdentifier.scope == elem['scope'],
models.DataIdentifier.name == elem['name']).one()
except NoResultFound:
raise DataIdentifierNotFound('Data identifier %s:%s is not valid.' % (elem['scope'], elem['name']))
except TypeError as error:
raise InvalidObject(error.args)
# 2.1 If the did is a constituent, relay the rule to the archive
if did.did_type == DIDType.FILE and did.constituent: # Check if a single replica of this DID exists
replica_cnt = session.query(models.RSEFileAssociation).join(models.RSE, models.RSEFileAssociation.rse_id == models.RSE.id)\
.filter(models.RSEFileAssociation.scope == did.scope,
models.RSEFileAssociation.name == did.name,
models.RSEFileAssociation.state == ReplicaState.AVAILABLE,
models.RSE.rse_type != RSEType.TAPE).count()
if replica_cnt == 0: # Put the rule on the archive
archive = session.query(models.ConstituentAssociation).join(models.RSEFileAssociation,
and_(models.ConstituentAssociation.scope == models.RSEFileAssociation.scope,
models.ConstituentAssociation.name == models.RSEFileAssociation.name))\
.filter(models.ConstituentAssociation.child_scope == did.scope,
models.ConstituentAssociation.child_name == did.name).first()
if archive is not None:
elem['name'] = archive.name
elem['scope'] = archive.scope
try:
did = session.query(models.DataIdentifier).filter(models.DataIdentifier.scope == elem['scope'],
models.DataIdentifier.name == elem['name']).one()
except NoResultFound:
raise DataIdentifierNotFound('Data identifier %s:%s is not valid.' % (elem['scope'], elem['name']))
except TypeError as error:
raise InvalidObject(error.args)
else: # Put the rule on the constituent directly
pass
rule_ids[(elem['scope'], elem['name'])] = []
# 3. Resolve the did into its contents
with record_timer_block('rule.add_rules.resolve_dids_to_locks_replicas'):
# Get all Replicas, not only the ones interesting for the rse_expression
datasetfiles, locks, replicas, source_replicas = __resolve_did_to_locks_and_replicas(did=did,
nowait=False,
restrict_rses=restrict_rses,
source_rses=all_source_rses,
session=session)
for rule in rules:
with record_timer_block('rule.add_rules.add_rule'):
# 4. Resolve the rse_expression into a list of RSE-ids
vo = rule['account'].vo
if rule.get('ignore_availability'):
rses = parse_expression(rule['rse_expression'], filter={'vo': vo}, session=session)
else:
rses = parse_expression(rule['rse_expression'], filter={'vo': vo, 'availability_write': True}, session=session)
if rule.get('lifetime', None) is None: # Check if one of the rses is a staging area
if [rse for rse in rses if rse.get('staging_area', False)]:
raise StagingAreaRuleRequiresLifetime()
# Check SCRATCHDISK Policy
try:
lifetime = get_scratch_policy(rule.get('account'), rses, rule.get('lifetime', None), session=session)
except UndefinedPolicy:
lifetime = rule.get('lifetime', None)
rule['lifetime'] = lifetime
# 4.5 Get the lifetime
eol_at = define_eol(did.scope, did.name, rses, session=session)
# Auto-lock rules for TAPE rses
if not rule.get('locked', False) and rule.get('lifetime', None) is None:
if [rse for rse in rses if rse.get('rse_type', RSEType.DISK) == RSEType.TAPE]:
rule['locked'] = True
# Block manual approval if RSE does not allow it
if rule.get('ask_approval', False):
for rse in rses:
if list_rse_attributes(rse_id=rse['id'], session=session).get('block_manual_approval', False):
raise ManualRuleApprovalBlocked()
if rule.get('source_replica_expression'):
source_rses = parse_expression(rule.get('source_replica_expression'), filter={'vo': vo}, session=session)
else:
source_rses = []
# 5. Create the RSE selector
with record_timer_block('rule.add_rules.create_rse_selector'):
rseselector = RSESelector(account=rule['account'], rses=rses, weight=rule.get('weight'), copies=rule['copies'], ignore_account_limit=rule.get('ask_approval', False), session=session)
# 4. Create the replication rule
with record_timer_block('rule.add_rules.create_rule'):
grouping = {'ALL': RuleGrouping.ALL, 'NONE': RuleGrouping.NONE}.get(rule.get('grouping'), RuleGrouping.DATASET)
expires_at = datetime.utcnow() + timedelta(seconds=rule.get('lifetime')) if rule.get('lifetime') is not None else None
notify = {'Y': RuleNotification.YES, 'C': RuleNotification.CLOSE, 'P': RuleNotification.PROGRESS}.get(rule.get('notify'), RuleNotification.NO)
if rule.get('meta') is not None:
try:
meta = json.dumps(rule.get('meta'))
except Exception:
meta = None
else:
meta = None
new_rule = models.ReplicationRule(account=rule['account'],
name=did.name,
scope=did.scope,
did_type=did.did_type,
copies=rule['copies'],
rse_expression=rule['rse_expression'],
locked=rule.get('locked'),
grouping=grouping,
expires_at=expires_at,
weight=rule.get('weight'),
source_replica_expression=rule.get('source_replica_expression'),
activity=rule.get('activity'),
subscription_id=rule.get('subscription_id'),
notification=notify,
purge_replicas=rule.get('purge_replicas', False),
ignore_availability=rule.get('ignore_availability', False),
comments=rule.get('comment', None),
priority=rule.get('priority', 3),
split_container=rule.get('split_container', False),
meta=meta,
eol_at=eol_at)
try:
new_rule.save(session=session)
except IntegrityError as error:
if match('.*ORA-00001.*', str(error.args[0])):
raise DuplicateRule(error.args[0])
elif str(error.args[0]) == '(IntegrityError) UNIQUE constraint failed: rules.scope, rules.name, rules.account, rules.rse_expression, rules.copies':
raise DuplicateRule(error.args[0])
raise InvalidReplicationRule(error.args[0])
rule_ids[(did.scope, did.name)].append(new_rule.id)
if rule.get('ask_approval', False):
new_rule.state = RuleState.WAITING_APPROVAL
# Block manual approval for multi-rse rules
if len(rses) > 1:
raise InvalidReplicationRule('Ask approval is not allowed for rules with multiple RSEs')
if len(rses) == 1 and not did.is_open and did.bytes is not None and did.length is not None:
# This rule can be considered for auto-approval:
rse_attr = list_rse_attributes(rse_id=rses[0]['id'], session=session)
auto_approve = False
if 'auto_approve_bytes' in rse_attr and 'auto_approve_files' in rse_attr:
if did.bytes < int(rse_attr.get('auto_approve_bytes')) and did.length < int(rse_attr.get('auto_approve_bytes')):
auto_approve = True
elif did.bytes < int(rse_attr.get('auto_approve_bytes', -1)):
auto_approve = True
elif did.length < int(rse_attr.get('auto_approve_files', -1)):
auto_approve = True
if auto_approve:
logging.debug("Auto approving rule %s", str(new_rule.id))
logging.debug("Created rule %s for injection", str(new_rule.id))
approve_rule(rule_id=new_rule.id, notify_approvers=False, session=session)
continue
logging.debug("Created rule %s in waiting for approval", str(new_rule.id))
__create_rule_approval_email(rule=new_rule, session=session)
continue
if rule.get('asynchronous', False):
new_rule.state = RuleState.INJECT
logging.debug("Created rule %s for injection", str(new_rule.id))
continue
if rule.get('split_container', False) and did.did_type == DIDType.CONTAINER:
new_rule.state = RuleState.INJECT
logging.debug("Created rule %s for injection due to Split Container mode", str(new_rule.id))
continue
# 5. Apply the replication rule to create locks, replicas and transfers
with record_timer_block('rule.add_rules.create_locks_replicas_transfers'):
try:
__create_locks_replicas_transfers(datasetfiles=datasetfiles,
locks=locks,
replicas=replicas,
source_replicas=source_replicas,
rseselector=rseselector,
rule=new_rule,
preferred_rse_ids=[],
source_rses=[rse['id'] for rse in source_rses],
session=session)
except IntegrityError as error:
raise ReplicationRuleCreationTemporaryFailed(error.args[0])
if new_rule.locks_stuck_cnt > 0:
new_rule.state = RuleState.STUCK
new_rule.error = 'MissingSourceReplica'
if new_rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=new_rule.id).update({'state': LockState.STUCK})
elif new_rule.locks_replicating_cnt == 0:
new_rule.state = RuleState.OK
if new_rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=new_rule.id).update({'state': LockState.OK})
session.flush()
if new_rule.notification == RuleNotification.YES:
generate_email_for_rule_ok_notification(rule=new_rule, session=session)
generate_rule_notifications(rule=new_rule, replicating_locks_before=0, session=session)
else:
new_rule.state = RuleState.REPLICATING
if new_rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=new_rule.id).update({'state': LockState.REPLICATING})
# Add rule to History
insert_rule_history(rule=new_rule, recent=True, longterm=True, session=session)
logging.info("Created rule %s [%d/%d/%d] in state %s", str(new_rule.id), new_rule.locks_ok_cnt, new_rule.locks_replicating_cnt, new_rule.locks_stuck_cnt, str(new_rule.state))
return rule_ids
@transactional_session
def inject_rule(rule_id, session=None):
"""
Inject a replication rule.
:param rule_id: The id of the rule to inject.
:param new_owner: The new owner of the rule.
:param session: The database session in use.
:raises: InvalidReplicationRule, InsufficientAccountLimit, InvalidRSEExpression, DataId, RSEOverQuota
"""
if USE_NEW_RULE_ALGORITHM:
use_new_rule_algorithm = True
else:
use_new_rule_algorithm = core_config_get('rules', 'use_new_rule_algorithm', default=False, session=session)
try:
rule = session.query(models.ReplicationRule).filter(models.ReplicationRule.id == rule_id).with_for_update(nowait=True).one()
except NoResultFound:
raise RuleNotFound('No rule with the id %s found' % (rule_id))
# Check if rule will expire in the next 5 minutes:
if rule.child_rule_id is None and rule.expires_at is not None and rule.expires_at < datetime.utcnow() + timedelta(seconds=300):
logging.info('Rule %s expiring soon, skipping', str(rule.id))
return
# Special R2D2 container handling
if (rule.did_type == DIDType.CONTAINER and '.r2d2_request.' in rule.name) or (rule.split_container and rule.did_type == DIDType.CONTAINER):
logging.debug("Creating dataset rules for Split Container rule %s", str(rule.id))
# Get all child datasets and put rules on them
dids = [{'scope': dataset['scope'], 'name': dataset['name']} for dataset in rucio.core.did.list_child_datasets(scope=rule.scope, name=rule.name, session=session)]
# Remove duplicates from the list of dictionaries
dids = [dict(t) for t in {tuple(d.items()) for d in dids}]
# Remove dids which already have a similar rule
dids = [did for did in dids if session.query(models.ReplicationRule).filter_by(scope=did['scope'], name=did['name'], account=rule.account, rse_expression=rule.rse_expression).count() == 0]
if rule.expires_at:
lifetime = (rule.expires_at - datetime.utcnow()).days * 24 * 3600 + (rule.expires_at - datetime.utcnow()).seconds
else:
lifetime = None
notify = {RuleNotification.YES: 'Y', RuleNotification.CLOSE: 'C', RuleNotification.PROGRESS: 'P'}.get(rule.notification, 'N')
add_rule(dids=dids,
account=rule.account,
copies=rule.copies,
rse_expression=rule.rse_expression,
grouping='DATASET',
weight=None,
lifetime=lifetime,
locked=False,
subscription_id=None,
activity=rule.activity,
notify=notify,
comment=rule.comments,
asynchronous=True,
ignore_availability=rule.ignore_availability,
ignore_account_limit=True,
priority=rule.priority,
split_container=rule.split_container,
session=session)
rule.delete(session=session)
return
# 1. Resolve the rse_expression into a list of RSE-ids
with record_timer_block('rule.add_rule.parse_rse_expression'):
vo = rule['account'].vo
if rule.ignore_availability:
rses = parse_expression(rule.rse_expression, filter={'vo': vo}, session=session)
else:
rses = parse_expression(rule.rse_expression, filter={'vo': vo, 'availability_write': True}, session=session)
if rule.source_replica_expression:
source_rses = parse_expression(rule.source_replica_expression, filter={'vo': vo}, session=session)
else:
source_rses = []
# 2. Create the rse selector
with record_timer_block('rule.add_rule.create_rse_selector'):
rseselector = RSESelector(account=rule['account'], rses=rses, weight=rule.weight, copies=rule.copies, ignore_account_limit=rule.ignore_account_limit, session=session)
# 3. Get the did
with record_timer_block('rule.add_rule.get_did'):
try:
did = session.query(models.DataIdentifier).filter(models.DataIdentifier.scope == rule.scope,
models.DataIdentifier.name == rule.name).one()
except NoResultFound:
raise DataIdentifierNotFound('Data identifier %s:%s is not valid.' % (rule.scope, rule.name))
except TypeError as error:
raise InvalidObject(error.args)
if use_new_rule_algorithm:
# 4. Apply the rule
with record_timer_block('rule.add_rule.apply_rule'):
try:
apply_rule(did, rule, [x['id'] for x in rses], [x['id'] for x in source_rses], rseselector, session=session)
except IntegrityError as error:
raise ReplicationRuleCreationTemporaryFailed(error.args[0])
else:
# 5. Resolve the did to its contents
with record_timer_block('rule.add_rule.resolve_dids_to_locks_replicas'):
# Get all Replicas, not only the ones interesting for the rse_expression
datasetfiles, locks, replicas, source_replicas = __resolve_did_to_locks_and_replicas(did=did,
nowait=True,
restrict_rses=[rse['id'] for rse in rses],
source_rses=[rse['id'] for rse in source_rses],
session=session)
# 6. Apply the replication rule to create locks, replicas and transfers
with record_timer_block('rule.add_rule.create_locks_replicas_transfers'):
try:
__create_locks_replicas_transfers(datasetfiles=datasetfiles,
locks=locks,
replicas=replicas,
source_replicas=source_replicas,
rseselector=rseselector,
rule=rule,
preferred_rse_ids=[],
source_rses=[rse['id'] for rse in source_rses],
session=session)
except IntegrityError as error:
raise ReplicationRuleCreationTemporaryFailed(error.args[0])
if rule.locks_stuck_cnt > 0:
rule.state = RuleState.STUCK
rule.error = 'MissingSourceReplica'
if rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=rule.id).update({'state': LockState.STUCK})
elif rule.locks_replicating_cnt == 0:
rule.state = RuleState.OK
if rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=rule.id).update({'state': LockState.OK})
session.flush()
if rule.notification == RuleNotification.YES:
generate_email_for_rule_ok_notification(rule=rule, session=session)
generate_rule_notifications(rule=rule, replicating_locks_before=0, session=session)
# Try to release potential parent rules
release_parent_rule(child_rule_id=rule.id, session=session)
else:
rule.state = RuleState.REPLICATING
if rule.grouping != RuleGrouping.NONE:
session.query(models.DatasetLock).filter_by(rule_id=rule.id).update({'state': LockState.REPLICATING})
# Add rule to History
insert_rule_history(rule=rule, recent=True, longterm=True, session=session)
if use_new_rule_algorithm:
logging.info("Created rule %s [%d/%d/%d] with new algorithm for did %s:%s in state %s", str(rule.id), rule.locks_ok_cnt, rule.locks_replicating_cnt, rule.locks_stuck_cnt, rule.scope, rule.name, str(rule.state))
else:
logging.info("Created rule %s [%d/%d/%d] for did %s:%s in state %s", str(rule.id), rule.locks_ok_cnt, rule.locks_replicating_cnt, rule.locks_stuck_cnt, rule.scope, rule.name, str(rule.state))
@stream_session
def list_rules(filters={}, session=None):
"""
List replication rules.
:param filters: dictionary of attributes by which the results should be filtered.
:param session: The database session in use.
:raises: RucioException
"""
query = session.query(models.ReplicationRule)
if filters:
for (key, value) in filters.items():
if key in ['account', 'scope']:
if '*' in value.internal:
value = value.internal.replace('*', '%')
query = query.filter(getattr(models.ReplicationRule, key).like(value))
continue
# else fall through
elif key == 'created_before':
query = query.filter(models.ReplicationRule.created_at <= str_to_date(value))
continue
elif key == 'created_after':
query = query.filter(models.ReplicationRule.created_at >= str_to_date(value))
continue
elif key == 'updated_before':
query = query.filter(models.ReplicationRule.updated_at <= str_to_date(value))
continue
elif key == 'updated_after':
query = query.filter(models.ReplicationRule.updated_at >= str_to_date(value))
continue
elif key == 'state':
if isinstance(value, string_types):
value = RuleState.from_string(value)
else:
try:
value = RuleState.from_sym(value)
except ValueError:
pass
elif key == 'did_type' and isinstance(value, string_types):
value = DIDType.from_string(value)
elif key == 'grouping' and isinstance(value, string_types):
value = RuleGrouping.from_string(value)
query = query.filter(getattr(models.ReplicationRule, key) == value)
try:
for rule in query.yield_per(5):
d = {}
for column in rule.__table__.columns:
d[column.name] = getattr(rule, column.name)
yield d
except StatementError:
raise RucioException('Badly formatted input (IDs?)')
@stream_session
def list_rule_history(rule_id, session=None):
"""
List the rule history of a rule.
:param rule_id: The id of the rule.
:param session: The database session in use.
:raises: RucioException
"""
query = session.query(models.ReplicationRuleHistoryRecent.updated_at,
models.ReplicationRuleHistoryRecent.state,
models.ReplicationRuleHistoryRecent.locks_ok_cnt,
models.ReplicationRuleHistoryRecent.locks_stuck_cnt,
models.ReplicationRuleHistoryRecent.locks_replicating_cnt).filter_by(id=rule_id).order_by(models.ReplicationRuleHistoryRecent.updated_at)
try:
for rule in query.yield_per(5):
yield {'updated_at': rule[0], 'state': rule[1], 'locks_ok_cnt': rule[2], 'locks_stuck_cnt': rule[3], 'locks_replicating_cnt': rule[4]}
except StatementError:
raise RucioException('Badly formatted input (IDs?)')
@stream_session
def list_rule_full_history(scope, name, session=None):
"""
List the rule history of a DID.
:param scope: The scope of the DID.
:param name: The name of the DID.
:param session: The database session in use.
:raises: RucioException
"""
query = session.query(models.ReplicationRuleHistory.id,
models.ReplicationRuleHistory.created_at,
models.ReplicationRuleHistory.updated_at,
models.ReplicationRuleHistory.rse_expression,
models.ReplicationRuleHistory.state,
models.ReplicationRuleHistory.account,
models.ReplicationRuleHistory.locks_ok_cnt,
models.ReplicationRuleHistory.locks_stuck_cnt,
models.ReplicationRuleHistory.locks_replicating_cnt).\
with_hint(models.ReplicationRuleHistory, "INDEX(RULES_HISTORY_SCOPENAME_IDX)", 'oracle').\
filter(models.ReplicationRuleHistory.scope == scope, models.ReplicationRuleHistory.name == name).\
order_by(models.ReplicationRuleHistory.created_at, models.ReplicationRuleHistory.updated_at)
for rule in query.yield_per(5):
yield {'rule_id': rule[0], 'created_at': rule[1], 'updated_at': rule[2], 'rse_expression': rule[3], 'state': rule[4],
'account': rule[5], 'locks_ok_cnt': rule[6], 'locks_stuck_cnt': rule[7], 'locks_replicating_cnt': rule[8]}
@stream_session
def list_associated_rules_for_file(scope, name, session=None):
"""
List replication rules a file is affected from.
:param scope: Scope of the file.
:param name: Name of the file.
:param session: The database session in use.
:raises: RucioException
"""
rucio.core.did.get_did(scope=scope, name=name, session=session) # Check if the did acually exists
query = session.query(models.ReplicationRule).\
with_hint(models.ReplicaLock, "INDEX(LOCKS LOCKS_PK)", 'oracle').\
join(models.ReplicaLock, models.ReplicationRule.id == models.ReplicaLock.rule_id).\
filter(models.ReplicaLock.scope == scope, models.ReplicaLock.name == name).distinct()
try:
for rule in query.yield_per(5):
d = {}
for column in rule.__table__.columns:
d[column.name] = getattr(rule, column.name)
yield d
except StatementError:
raise RucioException('Badly formatted input (IDs?)')
@transactional_session
def delete_rule(rule_id, purge_replicas=None, soft=False, delete_parent=False, nowait=False, session=None,
ignore_rule_lock=False):
"""
Delete a replication rule.
:param rule_id: The rule to delete.
:param purge_replicas: Purge the replicas immediately.
:param soft: Only perform a soft deletion.
:param delete_parent: Delete rules even if they have a child_rule_id set.
:param nowait: Nowait parameter for the FOR UPDATE statement.
:param session: The database session in use.
:param ignore_rule_lock: Ignore any locks on the rule
:raises: RuleNotFound if no Rule can be found.
:raises: UnsupportedOperation if the Rule is locked.
"""
with record_timer_block('rule.delete_rule'):
try:
rule = session.query(models.ReplicationRule)\
.filter(models.ReplicationRule.id == rule_id)\
.with_for_update(nowait=nowait).one()
except NoResultFound:
raise RuleNotFound('No rule with the id %s found' % rule_id)
if rule.locked and not ignore_rule_lock:
raise UnsupportedOperation('The replication rule is locked and has to be unlocked before it can be deleted.')
if rule.child_rule_id is not None and not delete_parent:
raise UnsupportedOperation('The replication rule has a child rule and thus cannot be deleted.')
if purge_replicas is not None:
rule.purge_replicas = purge_replicas
if soft:
if rule.expires_at:
rule.expires_at = min(datetime.utcnow() + timedelta(seconds=3600), rule.expires_at)
else:
rule.expires_at = datetime.utcnow() + timedelta(seconds=3600)
if rule.child_rule_id is not None and delete_parent:
rule.child_rule_id = None
insert_rule_history(rule=rule, recent=True, longterm=False, session=session)
return
locks = session.query(models.ReplicaLock)\
.filter(models.ReplicaLock.rule_id == rule_id)\
.with_for_update(nowait=nowait).yield_per(100)
# Remove locks, set tombstone if applicable
transfers_to_delete = [] # [{'scope': , 'name':, 'rse_id':}]
account_counter_decreases = {} # {'rse_id': [file_size, file_size, file_size]}
for lock in locks:
if __delete_lock_and_update_replica(lock=lock, purge_replicas=rule.purge_replicas,
nowait=nowait, session=session):
transfers_to_delete.append({'scope': lock.scope, 'name': lock.name, 'rse_id': lock.rse_id})
if lock.rse_id not in account_counter_decreases:
account_counter_decreases[lock.rse_id] = []
account_counter_decreases[lock.rse_id].append(lock.bytes)
# Delete the DatasetLocks
session.query(models.DatasetLock)\
.filter(models.DatasetLock.rule_id == rule_id)\
.delete(synchronize_session=False)
# Decrease account_counters
for rse_id in account_counter_decreases.keys():
account_counter.decrease(rse_id=rse_id, account=rule.account, files=len(account_counter_decreases[rse_id]),
bytes=sum(account_counter_decreases[rse_id]), session=session)
# Try to release potential parent rules
release_parent_rule(child_rule_id=rule.id, remove_parent_expiration=True, session=session)
# Insert history
insert_rule_history(rule=rule, recent=False, longterm=True, session=session)