-
Notifications
You must be signed in to change notification settings - Fork 123
/
mqbcfg.xsd
815 lines (755 loc) · 38.6 KB
/
mqbcfg.xsd
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
<?xml version='1.0' encoding='UTF-8'?>
<schema xmlns = 'http://www.w3.org/2001/XMLSchema'
xmlns:bdem = 'http://bloomberg.com/schemas/bdem'
xmlns:tns = 'http://bloomberg.com/schemas/mqbcfg'
targetNamespace = 'http://bloomberg.com/schemas/mqbcfg'
bdem:package = 'mqbcfg'
elementFormDefault = 'qualified'>
<complexType name='Configuration'>
<sequence>
<element name='taskConfig' type='tns:TaskConfig'/>
<element name='appConfig' type='tns:AppConfig'/>
</sequence>
</complexType>
<complexType name='TaskConfig'>
<sequence>
<element name='allocatorType' type='tns:AllocatorType'/>
<element name='allocationLimit' type='unsignedLong'/>
<element name='logController' type='tns:LogController'/>
</sequence>
</complexType>
<simpleType name='AllocatorType'>
<restriction base='string' bdem:preserveEnumOrder='1'>
<enumeration value='NEWDELETE' bdem:id='0'/>
<enumeration value='COUNTING' bdem:id='1'/>
<enumeration value='STACKTRACETEST' bdem:id='2'/>
</restriction>
</simpleType>
<complexType name='SyslogConfig'>
<sequence>
<element name='enabled' type='boolean' default='false'/>
<element name='appName' type='string'/>
<element name='logFormat' type='string'/>
<element name='verbosity' type='string'/>
</sequence>
</complexType>
<complexType name='LogController'>
<sequence>
<element name='fileName' type='string'/>
<element name='fileMaxAgeDays' type='int'/>
<element name='rotationBytes' type='int'/>
<element name='logfileFormat' type='string'/>
<element name='consoleFormat' type='string'/>
<element name='loggingVerbosity' type='string'/>
<element name='bslsLogSeverityThreshold' type='string' default='ERROR'/>
<element name='consoleSeverityThreshold' type='string'/>
<element name='categories' type='string' maxOccurs='unbounded'/>
<!-- format for 'categories': 'categoryExpression:severity:color' -->
<element name='syslog' type='tns:SyslogConfig'/>
</sequence>
</complexType>
<complexType name='AppConfig'>
<annotation>
<documentation>
Top level typ for the broker's configuration.
brokerInstanceName...: name of the broker instance
brokerVersion........: version of the broker
configVersion........: version of the bmqbrkr.cfg config
etcDir...............: directory containing the json config files
hostName.............: name of the current host
hostTags.............: tags of the current host
hostDataCenter.......: datacenter the current host resides in
isRunningOnDev.......: true if running on dev
logsObserverMaxSize..: maximum number of log records to keep
latencyMonitorDomain.: common part of all latemon domains
dispatcherConfig.....: configuration for the dispatcher
stats................: configuration for the stats
networkInterfaces....: configuration for the network interfaces
bmqconfConfig........: configuration for bmqconf
plugins..............: configuration for the plugins
msgPropertiesSupport.: information about if/how to advertise support for v2 message properties
configureStream......: send new ConfigureStream instead of old ConfigureQueue
advertiseSubscriptions.: temporarily control use of ConfigureStream in SDK/>
</documentation>
</annotation>
<sequence>
<element name='brokerInstanceName' type='string'/>
<element name='brokerVersion' type='int'/>
<element name='configVersion' type='int'/>
<element name='etcDir' type='string'/>
<element name='hostName' type='string'/>
<element name='hostTags' type='string'/>
<element name='hostDataCenter' type='string'/>
<element name='isRunningOnDev' type='boolean'/>
<element name='logsObserverMaxSize' type='int'/>
<element name='latencyMonitorDomain' type='string'
default='bmq.sys.latemon.latency'/>
<element name='dispatcherConfig' type='tns:DispatcherConfig'/>
<element name='stats' type='tns:StatsConfig'/>
<element name='networkInterfaces' type='tns:NetworkInterfaces'/>
<element name='bmqconfConfig' type='tns:BmqconfConfig'/>
<element name='plugins' type='tns:Plugins'/>
<element name='messagePropertiesV2' type='tns:MessagePropertiesV2'/>
<element name='configureStream' type='boolean' default='false'/>
<element name='advertiseSubscriptions' type='boolean' default='false'/>
</sequence>
</complexType>
<complexType name='DispatcherProcessorParameters'>
<sequence>
<element name='queueSize' type='int'/>
<element name='queueSizeLowWatermark' type='int'/>
<element name='queueSizeHighWatermark' type='int'/>
</sequence>
</complexType>
<complexType name='DispatcherProcessorConfig'>
<sequence>
<element name='numProcessors' type='int'/>
<element name='processorConfig' type='tns:DispatcherProcessorParameters'/>
</sequence>
</complexType>
<complexType name='DispatcherConfig'>
<sequence>
<element name='sessions' type='tns:DispatcherProcessorConfig'/>
<element name='queues' type='tns:DispatcherProcessorConfig'/>
<element name='clusters' type='tns:DispatcherProcessorConfig'/>
</sequence>
</complexType>
<complexType name='StatsConfig'>
<sequence>
<element name='snapshotInterval' type='int' default='1'/> <!-- 0 to disable -->
<element name='appIdTagDomains' type='string' maxOccurs='unbounded'/>
<element name='plugins' type='tns:StatPluginConfig' maxOccurs='unbounded'/>
<element name='printer' type='tns:StatsPrinterConfig'/>
</sequence>
</complexType>
<complexType name="StatPluginConfig">
<sequence>
<element name='name' type='string' default=''/>
<element name='queueSize' type='int' default='10000'/>
<element name='queueHighWatermark' type='int' default='5000'/>
<element name='queueLowWatermark' type='int' default='1000'/>
<element name='publishInterval' type='int' default='30'/> <!-- 0 to disable -->
<element name='namespacePrefix' type='string' default=''/>
<element name='hosts' type='string' maxOccurs='unbounded'/>
<element name='instanceId' type='string' default=''/>
<element name='prometheusSpecific' type='tns:StatPluginConfigPrometheus' minOccurs='0'/>
</sequence>
</complexType>
<complexType name="StatPluginConfigPrometheus">
<sequence>
<element name='mode' type='tns:ExportMode' default='E_PULL'/>
<element name='host' type='string' default='localhost'/>
<element name='port' type='int' default='8080'/>
</sequence>
</complexType>
<simpleType name='ExportMode' bdem:preserveEnumOrder='1'>
<restriction base='string'>
<enumeration value='E_PUSH' bdem:id='0'/>
<enumeration value='E_PULL' bdem:id='1'/>
</restriction>
</simpleType>
<complexType name="StatsPrinterConfig">
<sequence>
<element name='printInterval' type='int' default='60'/> <!-- 0 to disable -->
<element name='file' type='string'/>
<element name='maxAgeDays' type='int'/>
<element name='rotateBytes' type='int' default='268435456'/> <!-- 256 * 1024 * 1024 -->
<element name='rotateDays' type='int' default='1'/>
</sequence>
</complexType>
<complexType name='NetworkInterfaces'>
<sequence>
<element name='heartbeats' type='tns:Heartbeat'/>
<element name='tcpInterface' type='tns:TcpInterfaceConfig' minOccurs='0'/>
</sequence>
</complexType>
<complexType name='MessagePropertiesV2'>
<annotation>
<documentation>
This complex type captures information which can be used to tell a
broker if it should advertise support for message properties v2 format
(also knownn as extended or 'EX' message properties at some places).
Additionally, broker can be configured to advertise this feature only
to those C++ and Java clients which match a certain minimum SDK version.
</documentation>
</annotation>
<sequence>
<element name='advertiseV2Support' type='boolean' default='true'/>
<element name='minCppSdkVersion' type='int' default='11207'/>
<element name='minJavaSdkVersion' type='int' default='10'/>
</sequence>
</complexType>
<complexType name='Heartbeat'>
<annotation>
<documentation>
The following parameters define, for the various connection types,
after how many missed heartbeats the connection should be proactively
resetted. Note that a value of 0 means that smart-heartbeat is
entirely disabled for this kind of connection (i.e., it will not
periodically emit heatbeats in case no traffic is received, and will
therefore not quickly detect stale remote peer). Each value is in
multiple of the 'NetworkInterfaces/TCPInterfaceConfig/heartIntervalMs'.
client............:
The channel represents a client connected to the broker
downstreamBroker..:
The channel represents a downstream broker connected to this
broker, i.e. a proxy.
upstreamBroker....:
The channel represents an upstream broker connection from this
broker, i.e. a cluster proxy connection.
clusterPeer.......:
The channel represents a connection with a peer node in the cluster
this broker is part of.
</documentation>
</annotation>
<sequence>
<element name='client' type='int' default='0'/>
<element name='downstreamBroker' type='int' default='0'/>
<element name='upstreamBroker' type='int' default='0'/>
<element name='clusterPeer' type='int' default='0'/>
</sequence>
</complexType>
<complexType name='TcpInterfaceConfig'>
<annotation>
<documentation>
lowWatermark.........:
highWatermark........:
Watermarks used for channels with a client or proxy.
nodeLowWatermark.....:
nodeHighWatermark....:
Reduced watermarks for communication between cluster nodes where
BlazingMQ maintains its own cache.
heartbeatIntervalMs..:
How often (in milliseconds) to check if the channel received data,
and emit heartbeat. 0 to globally disable.
useNtf...............:
Use the new NTF based TCP transport library instead of
the existing one based on BTE
</documentation>
</annotation>
<sequence>
<element name='name' type='string'/>
<element name='port' type='int'/>
<element name='ioThreads' type='int'/>
<element name='maxConnections' type='int' default='10000'/>
<element name='lowWatermark' type='long'/>
<element name='highWatermark' type='long'/>
<element name='nodeLowWatermark' type='long' default='1024'/>
<element name='nodeHighWatermark' type='long' default='2048'/>
<element name='heartbeatIntervalMs' type='int' default='3000'/>
<element name='useNtf' type='boolean' default='false'/>
</sequence>
</complexType>
<complexType name='BmqconfConfig'>
<sequence>
<element name='cacheTTLSeconds' type='int'/>
</sequence>
</complexType>
<complexType name='Plugins'>
<sequence>
<element name='libraries' type='string' maxOccurs='unbounded'/>
<element name='enabled' type='string' maxOccurs='unbounded'/>
</sequence>
</complexType>
<!-- ========================================================================
DOMAIN_RESOLUTION
========================================================================= -->
<complexType name='ResolvedDomain'>
<annotation>
<documentation>
Top level type representing the information retrieved when resolving a
domain.
resolvedName.: Resolved name of the domain
clusterName..: Name of the cluster where this domain exists
</documentation>
</annotation>
<sequence>
<element name='resolvedName' type='string'/>
<element name='clusterName' type='string'/>
</sequence>
</complexType>
<!-- ========================================================================
CLUSTER
========================================================================= -->
<complexType name='ClustersDefinition'>
<annotation>
<documentation>
Top level type representing the configuration for all clusters.
myClusters.................:
definition of the clusters the current machine is part of (if any);
empty means this broker does not belong to any cluster
myReverseClusters..........:
name of the clusters (if any) the current machine is expected to
receive inbound connections about and therefore should pro-actively
create a proxy cluster at startup
myVirtualClusters..........:
information about all the virtual clusters the current machine is
considered to belong to (if any)
clusters...................: array of cluster definition
reversedClusterConnections.:
cluster and associated remote connections that should be
established
</documentation>
</annotation>
<sequence>
<element name='myClusters' type='tns:ClusterDefinition' maxOccurs='unbounded'/>
<element name='myReverseClusters' type='string' maxOccurs='unbounded'/>
<element name='myVirtualClusters' type='tns:VirtualClusterInformation' maxOccurs='unbounded'/>
<element name='proxyClusters' type='tns:ClusterProxyDefinition' maxOccurs='unbounded'/>
<element name='reversedClusterConnections' type='tns:ReversedClusterConnection' maxOccurs='unbounded'/>
</sequence>
</complexType>
<complexType name='VirtualClusterInformation'>
<annotation>
<documentation>
Type representing the information about the current node with regards
to virtual cluster.
name.............: name of the cluster
selfNodeId.......: id of the current node in that virtual cluster
</documentation>
</annotation>
<sequence>
<element name='name' type='string'/>
<element name='selfNodeId' type='int'/>
</sequence>
</complexType>
<complexType name='ClusterDefinition'>
<annotation>
<documentation>
Type representing the configuration for a cluster.
name..................: name of the cluster
nodes.................: list of nodes in the cluster
partitionConfig.......: configuration for the storage
masterAssignment......: algorithm to use for partition's master assignment
elector...............: configuration for leader election amongst the nodes
queueOperations.......: configuration for queue operations on the cluster
clusterAttributes.....: attributes specific to this cluster
clusterMonitorConfig..: configuration for cluster state monitor
messageThrottleConfig.: configuration for message throttling intervals and
thresholds.
</documentation>
</annotation>
<sequence>
<element name='name' type='string'/>
<element name='nodes' type='tns:ClusterNode' maxOccurs='unbounded'/>
<element name='partitionConfig' type='tns:PartitionConfig'/>
<element name='masterAssignment' type='tns:MasterAssignmentAlgorithm'/>
<element name='elector' type='tns:ElectorConfig'/>
<element name='queueOperations' type='tns:QueueOperationsConfig'/>
<element name='clusterAttributes' type='tns:ClusterAttributes'/>
<element name='clusterMonitorConfig' type='tns:ClusterMonitorConfig'/>
<element name='messageThrottleConfig' type='tns:MessageThrottleConfig'/>
</sequence>
</complexType>
<complexType name='ClusterProxyDefinition'>
<annotation>
<documentation>
Type representing the configuration for a cluster proxy.
name..................: name of the cluster
nodes.................: list of nodes in the cluster
queueOperations.......: configuration for queue operations with the cluster
clusterMonitorConfig..: configuration for cluster state monitor
messageThrottleConfig.: configuration for message throttling intervals and
thresholds.
</documentation>
</annotation>
<sequence>
<element name='name' type='string'/>
<element name='nodes' type='tns:ClusterNode' maxOccurs='unbounded'/>
<element name='queueOperations' type='tns:QueueOperationsConfig'/>
<element name='clusterMonitorConfig' type='tns:ClusterMonitorConfig'/>
<element name='messageThrottleConfig' type='tns:MessageThrottleConfig'/>
</sequence>
</complexType>
<simpleType name='MasterAssignmentAlgorithm'>
<annotation>
<documentation>
Enumeration of the various algorithm's used for assigning a master to
a partition:
- E_LEADER_IS_MASTER_ALL: the leader is master for all partitions
- E_LEAST_ASSIGNED: the active node with the least number of
partitions assigned is used
</documentation>
</annotation>
<restriction base='string' bdem:preserveEnumOrder='1'>
<enumeration value='E_LEADER_IS_MASTER_ALL' bdem:id='0'/>
<enumeration value='E_LEAST_ASSIGNED' bdem:id='1'/>
</restriction>
</simpleType>
<complexType name='PartitionConfig'>
<annotation>
<documentation>
Type representing the configuration for the storage layer of a cluster.
numPartitions........: number of partitions at each node in the cluster
location.............: location of active files for a partition
archiveLocation......: location of archive files for a partition
maxDataFileSize......: maximum size of partitions' data file
maxJournalFileSize...: maximum size of partitions' journal file
maxQlistFileSize.....: maximum size of partitions' qlist file
preallocate..........: flag to indicate whether files should be
preallocated on disk
maxArchivedFileSets..: maximum number of archived file sets per
partition to keep
prefaultPages........: flag to indicate whether to populate (prefault)
page tables for a mapping.
flushAtShutdown......: flag to indicate whether broker should flush
storage files to disk at shutdown
syncConfig...........: configuration for storage synchronization and
recovery
</documentation>
</annotation>
<sequence>
<element name='numPartitions' type='int'/>
<element name='location' type='string'/>
<element name='archiveLocation' type='string'/>
<element name='maxDataFileSize' type='unsignedLong'/>
<element name='maxJournalFileSize' type='unsignedLong'/>
<element name='maxQlistFileSize' type='unsignedLong'/>
<element name='preallocate' type='boolean' default='false'/>
<element name='maxArchivedFileSets' type='int'/>
<element name='prefaultPages' type='boolean' default='false'/>
<element name='flushAtShutdown' type='boolean' default='true'/>
<element name='syncConfig' type='tns:StorageSyncConfig'/>
</sequence>
</complexType>
<complexType name='ElectorConfig'>
<annotation>
<documentation>
Type representing the configuration for leader election amongst a
cluster of nodes.
initialWaitTimeoutMs.......:
initial wait timeout, in milliseconds, of a follower for leader
heartbeat before initiating election, as per the *Raft* Algorithm.
Note that `initialWaitTimeoutMs` should be larger than
`maxRandomWaitTimeoutMs`
maxRandomWaitTimeoutMs.....:
maximum random wait timeout, in milliseconds, of a follower for
leader heartbeat before initiating election, as per the *Raft*
Algorithm
scoutingResultTimeoutMs....:
timeout, in milliseconds, of a follower for awaiting scouting
responses from all nodes after sending scouting request.
electionResultTimeoutMs....:
timeout, in milliseconds, of a candidate for awaiting quorum to be
reached after proposing election, as per the *Raft* Algorithm
heartbeatBroadcastPeriodMs.:
frequency, in milliseconds, in which the leader broadcasts a
heartbeat signal, as per the *Raft* Algorithm,
heartbeatCheckPeriodMs.....:
frequency, in milliseconds, in which a follower checks for
heartbeat signals from the leader, as per the *Raft* Algorithm
heartbeatMissCount.........:
the number of missed heartbeat signals required before a follower
marks the current leader as inactive, as per the *Raft* Algorithm
quorum.....................:
the minimum number of votes required for a candidate to transition
to the leader. If zero, dynamically set to half the number of nodes
plus one
leaderSyncDelayMs..........:
delay, in milliseconds, after a leader has been elected before
initiating leader sync, in order to give a chance to all nodes to
come up and declare themselves AVAILABLE. Note that this should be
done only in case of cluster of size > 1
</documentation>
</annotation>
<sequence>
<element name='initialWaitTimeoutMs' type='int' default='8000'/>
<element name='maxRandomWaitTimeoutMs' type='int' default='3000'/>
<element name='scoutingResultTimeoutMs' type='int' default='4000'/>
<element name='electionResultTimeoutMs' type='int' default='4000'/>
<element name='heartbeatBroadcastPeriodMs' type='int' default='2000'/>
<element name='heartbeatCheckPeriodMs' type='int' default='1000'/>
<element name='heartbeatMissCount' type='int' default='10'/>
<element name='quorum' type='int' default='0'/>
<element name='leaderSyncDelayMs' type='int' default='80000'/>
</sequence>
</complexType>
<complexType name='QueueOperationsConfig'>
<annotation>
<documentation>
Type representing the configuration for queue operations on a cluster.
openTimeoutMs..............:
timeout, in milliseconds, to use when opening a queue. An open
request requires some co-ordination among the nodes (queue
assignment request/response, followed by queue open
request/response, etc)
configureTimeoutMs.........:
timeout, in milliseconds, to use for a configure queue request.
Note that `configureTimeoutMs` must be less than or equal to
`closeTimeoutMs` to prevent out-of-order processing of closeQueue
(e.g. closeQueue sent after configureQueue but timeout response
processed first for the closeQueue)
closeTimeoutMs.............:
timeout, in milliseconds, to use for a close queue request
reopenTimeoutMs............:
timeout, in milliseconds, to use when sending a reopen-queue
request. Ideally, we should use same value as `openTimeoutMs`, but
we are using a very large value as a workaround: during network
outages, a proxy or a replica may failover to a new upstream node,
which itself may be out of sync or not yet ready. Eventually, the
reopen-queue requests sent during failover may timeout, and will
never be retried, leading to a 'permanent' failure (client consumer
app stops receiving messages; PUT messages from client producer app
starts getting NAK'd or buffered). Using such a large timeout
value helps in a situation when network outages or its
after-effects are fixed after a few hours).
reopenRetryIntervalMs......:
duration, in milliseconds, after which a retry attempt should be
made to reopen the queue
reopenMaxAttempts..........:
maximum number of attempts to reopen a queue when restoring the
state in a proxy upon getting a new active node notification
assignmentTimeoutMs........:
timeout, in milliseconds, to use for a queue assignment request
keepaliveDurationMs........:
duration, in milliseconds, to keep a queue alive after it has met
the criteria for deletion
consumptionMonitorPeriodMs.:
frequency, in milliseconds, in which the consumption monitor checks
queue consumption statistics on the cluster
stopTimeoutMs..............:
timeout, in milliseconds, to use in StopRequest between
deconfiguring and closing each affected queue. This is primarily
to give a chance for pending PUSH mesages to be CONFIRMed.
shutdownTimeoutMs..........:
timeout, in milliseconds, to use when node stops for shutdown or
maintenance mode. This timeout should be greater than the
'stopTimeoutMs'. This is to handle misbehaving downstream which may
not reply to stopRequest (otherwise, this timeout is not expected
to be reached).
ackWindowSize..............:
number of PUTs without ACK requested after which we request an ACK.
This is to remove pending broadcast PUTs.
</documentation>
</annotation>
<sequence>
<element name='openTimeoutMs' type='int' default='300000'/> <!-- 5 minutes -->
<element name='configureTimeoutMs' type='int' default='300000'/> <!-- 5 minutes -->
<element name='closeTimeoutMs' type='int' default='300000'/> <!-- 5 minutes -->
<element name='reopenTimeoutMs' type='int' default='43200000'/> <!-- 12 hours -->
<element name='reopenRetryIntervalMs' type='int' default='5000'/> <!-- 5 seconds -->
<element name='reopenMaxAttempts' type='int' default='10'/>
<element name='assignmentTimeoutMs' type='int' default='15000'/> <!-- 15 seconds -->
<element name='keepaliveDurationMs' type='int' default='1800000'/> <!-- 30 minutes -->
<element name='consumptionMonitorPeriodMs' type='int' default='30000'/> <!-- 30 seconds -->
<element name='stopTimeoutMs' type='int' default='10000'/> <!-- 10 seconds -->
<element name='shutdownTimeoutMs' type='int' default='20000'/> <!-- 20 seconds -->
<element name='ackWindowSize' type='int' default='500'/> <!-- 500 messages -->
</sequence>
</complexType>
<complexType name='ClusterMonitorConfig'>
<annotation>
<documentation>
Type representing the configuration for cluster state monitor.
maxTimeLeader......:
Time (in seconds) before alarming that the cluster's leader is not
'active'
maxTimeMaster......:
Time (in seconds) before alarming that a partition's master is not
'active'
maxTimeNode........:
Time (in seconds) before alarming that a node is not 'available'
maxTimeFailover..:
Time (in seconds) before alarming that failover hasn't completed
thresholdLeader....:
Time (in seconds) before first notifying observers that cluster's
leader is not 'active'. This time interval is smaller than
'maxTimeLeader' because observing components may attempt to heal
the cluster state before an alarm is raised.
thresholdMaster....:
Time (in seconds) before notifying observers that a partition's
master is not 'active'. This time interval is smaller than
'maxTimeMaster' because observing components may attempt to heal
the cluster state before an alarm is raised.
thresholdNode......:
Time (in seconds) before notifying observers that a node is not
'available'. This time interval is smaller than 'maxTimeNode'
because observing components may attempt to heal the cluster state
before an alarm is raised.
thresholdFailover..:
Time (in seconds) before notifying observers that failover has not
completed. This time interval is smaller than 'maxTimeFailover'
because observing components may attempt to fix the issue before an
alarm is raised.
</documentation>
</annotation>
<sequence>
<element name='maxTimeLeader' type='int' default='60'/> <!-- 1 minute -->
<element name='maxTimeMaster' type='int' default='120'/> <!-- 2 minutes -->
<element name='maxTimeNode' type='int' default='120'/> <!-- 2 minutes -->
<element name='maxTimeFailover' type='int' default='600'/> <!-- 10 minutes -->
<element name='thresholdLeader' type='int' default='30'/> <!-- 30 seconds -->
<element name='thresholdMaster' type='int' default='60'/> <!-- 1 minute -->
<element name='thresholdNode' type='int' default='60'/> <!-- 1 minute -->
<element name='thresholdFailover' type='int' default='300'/> <!-- 5 minutes -->
</sequence>
</complexType>
<complexType name='ClusterAttributes'>
<annotation>
<documentation>
Type representing the attributes specific to a cluster.
isCSLModeEnabled.: indicates if CSL is enabled for this cluster
isFSMWorkflow....: indicates if CSL FSM workflow is enabled for this
cluster. This flag *must* be false if
'isCSLModeEnabled' is false.
</documentation>
</annotation>
<sequence>
<element name='isCSLModeEnabled' type ='boolean' default='false'/>
<element name='isFSMWorkflow' type ='boolean' default='false'/>
</sequence>
</complexType>
<complexType name='MessageThrottleConfig'>
<annotation>
<documentation>
Configuration values for message throttling intervals and thresholds.
lowInterval...: time in milliseconds.
highInterval..: time in milliseconds.
lowThreshold..: indicates the rda counter value at which we start
throttlling for time equal to 'lowInterval'.
highThreshold.: indicates the rda counter value at which we start
throttlling for time equal to 'highInterval'.
Note: lowInterval should be less than/equal to highInterval,
lowThreshold should be less than highThreshold.
</documentation>
</annotation>
<sequence>
<element name='lowThreshold' type='unsignedInt' default='2'/>
<element name='highThreshold' type='unsignedInt' default='4'/>
<element name='lowInterval' type='unsignedInt' default='1000'/>
<element name='highInterval' type='unsignedInt' default='3000'/>
</sequence>
</complexType>
<complexType name='StorageSyncConfig'>
<annotation>
<documentation>
Type representing the configuration for storage synchronization and
recovery of a cluster.
startupRecoveryMaxDurationMs...:
maximum amount of time, in milliseconds, in which recovery for a
partition at node startup must complete. This interval captures
the time taken to receive storage-sync response, retry attempts for
failed storage-sync request, as well as the time taken by the peer
node to send partition files to the starting (i.e. requester) node
maxAttemptsStorageSync.........:
maximum number of attempts that a node makes for storage-sync
requests when it comes up (this value includes the 1st attempt)
storageSyncReqTimeoutMs........:
timeout, in milliseconds, for the storage-sync request. A bigger
value is recommended because peer node could be busy serving
storage-sync request for other partition(s) assigned to the same
partition-dispatcher thread, etc. This timeout does *not* capture
the time taken by the peer to send partition files
masterSyncMaxDurationMs........:
maximum amount of time, in milliseconds, in which master sync for a
partition must complete. This interval includes the time taken by
replica node (the one with advanced view of the partition) to send
the file chunks, as well as time taken to receive partition-sync
state and data responses
partitionSyncStateReqTimeoutMs.:
timeout, in milliseconds, for partition-sync-state-query request.
This request is sent by a new master node to all replica nodes to
query their view of the partition
partitionSyncDataReqTimeoutMs..:
timeout, in milliseconds, for partition-sync-data-query request.
This request is sent by a new master node to a replica node (which
has an advanced view of the partition) to initiate partition sync.
This duration does *not* capture the amount of time which replica
might take to send the partition file
startupWaitDurationMs..........:
duration, in milliseconds, for which recovery manager waits for a
sync point for a partition. If no sync point is received from a
peer for a partition during this time, it is assumed that there is
no master for that partition, and recovery manager randomly picks
up a node from all the available ones with send a sync request. If
no peers are available at this time, it is assumed that entire
cluster is coming up together, and proceeds with local recovery for
that partition. Note that this value should be less than the
duration for which a node waits if its elected a leader and there
are no AVAILABLE nodes. This is important so that if all nodes in
the cluster are starting, they have a chance to wait for
'startupWaitDurationMs' milliseconds, find out that none of the
partitions have any master, go ahead with local recovery and
declare themselves as AVAILABLE. This will give the new leader
node a chance to make each node a master for a given partition.
Moreover, this value should be greater than the duration for which
a peer waits before attempting to reconnect to the node in the
cluster, so that peer has a chance to connect to this node, get
notified (via ClusterObserver), and send sync point if its a master
for any partition
fileChunkSize..................:
chunk size, in bytes, to send in one go to the peer when serving a
storage sync request from it
partitionSyncEventSize.........:
maximum size, in bytes, of bmqp::EventType::PARTITION_SYNC before
we send it to the peer
</documentation>
</annotation>
<sequence>
<element name='startupRecoveryMaxDurationMs' type='int' default='1200000'/> <!-- 20 minutes -->
<element name='maxAttemptsStorageSync' type='int' default='3'/>
<element name='storageSyncReqTimeoutMs' type='int' default='300000'/> <!-- 5 minutes -->
<element name='masterSyncMaxDurationMs' type='int' default='600000'/> <!-- 10 minutes -->
<element name='partitionSyncStateReqTimeoutMs' type='int' default='120000'/> <!-- 2 minutes -->
<element name='partitionSyncDataReqTimeoutMs' type='int' default='120000'/> <!-- 2 minutes -->
<element name='startupWaitDurationMs' type='int' default='60000'/> <!-- 60 seconds -->
<element name='fileChunkSize' type='int' default='4194304'/> <!-- 4 MB -->
<element name='partitionSyncEventSize' type='int' default='4194304'/> <!-- 4 MB -->
</sequence>
</complexType>
<complexType name='ClusterNode'>
<annotation>
<documentation>
Type representing the configuration of a node in a cluster.
id.........: the unique ID of that node in the cluster; must be a > 0
value
name.......: name of this node
datacenter.: the datacenter of that node
transport..: the transport configuration for establishing connectivity
with the node
</documentation>
</annotation>
<sequence>
<element name='id' type='int'/>
<element name='name' type='string'/>
<element name='dataCenter' type='string'/>
<element name='transport' type='tns:ClusterNodeConnection'/>
</sequence>
</complexType>
<complexType name='ClusterNodeConnection'>
<annotation>
<documentation>
Choice of all the various transport mode available to establish
connectivity with a node.
tcp.: TCP connectivity
</documentation>
</annotation>
<choice>
<element name='tcp' type='tns:TcpClusterNodeConnection'/>
</choice>
</complexType>
<complexType name='TcpClusterNodeConnection'>
<annotation>
<documentation>
Configuration of a TCP based cluster node connectivity.
endpoint.: endpoint URI of the node address
</documentation>
</annotation>
<sequence>
<element name='endpoint' type='string'/>
</sequence>
</complexType>
<complexType name='ReversedClusterConnection'>
<annotation>
<documentation>
Type representing the configuration for remote cluster connections..
name.............: name of the cluster
connections......: list of connections to establish
</documentation>
</annotation>
<sequence>
<element name='name' type='string'/>
<element name='connections' type='tns:ClusterNodeConnection' maxOccurs='unbounded'/>
</sequence>
</complexType>
</schema>