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
|
// wiredtiger_kv_engine.h
/**
* Copyright (C) 2014 MongoDB Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License, version 3,
* as published by the Free Software Foundation.
*
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
* As a special exception, the copyright holders give permission to link the
* code of portions of this program with the OpenSSL library under certain
* conditions as described in each individual source file and distribute
* linked combinations including the program with the OpenSSL library. You
* must comply with the GNU Affero General Public License in all respects for
* all of the code used other than as permitted herein. If you modify file(s)
* with this exception, you may extend this exception to your version of the
* file(s), but you are not obligated to do so. If you do not wish to do so,
* delete this exception statement from your version. If you delete this
* exception statement from all source files in the program, then also delete
* it in the license file.
*/
#pragma once
#include <list>
#include <memory>
#include <string>
#include <wiredtiger.h>
#include "mongo/bson/ordering.h"
#include "mongo/bson/timestamp.h"
#include "mongo/db/storage/kv/kv_engine.h"
#include "mongo/db/storage/wiredtiger/wiredtiger_oplog_manager.h"
#include "mongo/db/storage/wiredtiger/wiredtiger_session_cache.h"
#include "mongo/stdx/functional.h"
#include "mongo/stdx/mutex.h"
#include "mongo/util/elapsed_tracker.h"
namespace mongo {
class ClockSource;
class JournalListener;
class WiredTigerRecordStore;
class WiredTigerSessionCache;
class WiredTigerSizeStorer;
class WiredTigerKVEngine final : public KVEngine {
public:
static const int kDefaultJournalDelayMillis;
WiredTigerKVEngine(const std::string& canonicalName,
const std::string& path,
ClockSource* cs,
const std::string& extraOpenOptions,
size_t cacheSizeGB,
bool durable,
bool ephemeral,
bool repair,
bool readOnly);
virtual ~WiredTigerKVEngine();
void setRecordStoreExtraOptions(const std::string& options);
void setSortedDataInterfaceExtraOptions(const std::string& options);
virtual bool supportsDocLocking() const;
virtual bool supportsDirectoryPerDB() const;
virtual bool isDurable() const {
return _durable;
}
virtual bool isEphemeral() const {
return _ephemeral;
}
virtual RecoveryUnit* newRecoveryUnit();
virtual Status createRecordStore(OperationContext* opCtx,
StringData ns,
StringData ident,
const CollectionOptions& options) {
return createGroupedRecordStore(opCtx, ns, ident, options, KVPrefix::kNotPrefixed);
}
virtual std::unique_ptr<RecordStore> getRecordStore(OperationContext* opCtx,
StringData ns,
StringData ident,
const CollectionOptions& options) {
return getGroupedRecordStore(opCtx, ns, ident, options, KVPrefix::kNotPrefixed);
}
virtual Status createSortedDataInterface(OperationContext* opCtx,
StringData ident,
const IndexDescriptor* desc) {
return createGroupedSortedDataInterface(opCtx, ident, desc, KVPrefix::kNotPrefixed);
}
virtual SortedDataInterface* getSortedDataInterface(OperationContext* opCtx,
StringData ident,
const IndexDescriptor* desc) {
return getGroupedSortedDataInterface(opCtx, ident, desc, KVPrefix::kNotPrefixed);
}
virtual Status createGroupedRecordStore(OperationContext* opCtx,
StringData ns,
StringData ident,
const CollectionOptions& options,
KVPrefix prefix);
virtual std::unique_ptr<RecordStore> getGroupedRecordStore(OperationContext* opCtx,
StringData ns,
StringData ident,
const CollectionOptions& options,
KVPrefix prefix);
virtual Status createGroupedSortedDataInterface(OperationContext* opCtx,
StringData ident,
const IndexDescriptor* desc,
KVPrefix prefix);
virtual SortedDataInterface* getGroupedSortedDataInterface(OperationContext* opCtx,
StringData ident,
const IndexDescriptor* desc,
KVPrefix prefix);
virtual Status dropIdent(OperationContext* opCtx, StringData ident);
virtual Status okToRename(OperationContext* opCtx,
StringData fromNS,
StringData toNS,
StringData ident,
const RecordStore* originalRecordStore) const;
virtual int flushAllFiles(OperationContext* opCtx, bool sync);
virtual Status beginBackup(OperationContext* opCtx);
virtual void endBackup(OperationContext* opCtx);
virtual int64_t getIdentSize(OperationContext* opCtx, StringData ident);
virtual Status repairIdent(OperationContext* opCtx, StringData ident);
virtual bool hasIdent(OperationContext* opCtx, StringData ident) const;
std::vector<std::string> getAllIdents(OperationContext* opCtx) const;
virtual void cleanShutdown();
SnapshotManager* getSnapshotManager() const final {
return &_sessionCache->snapshotManager();
}
void setJournalListener(JournalListener* jl) final;
virtual void setStableTimestamp(Timestamp stableTimestamp) override;
virtual void setInitialDataTimestamp(Timestamp initialDataTimestamp) override;
/**
* This method will force the oldest timestamp to the input value. Callers must be serialized
* along with `setStableTimestamp`
*/
void setOldestTimestamp(Timestamp oldestTimestamp);
virtual bool supportsRecoverToStableTimestamp() const override;
virtual StatusWith<Timestamp> recoverToStableTimestamp() override;
virtual boost::optional<Timestamp> getRecoveryTimestamp() const override;
bool supportsReadConcernSnapshot() const final;
// wiredtiger specific
// Calls WT_CONNECTION::reconfigure on the underlying WT_CONNECTION
// held by this class
int reconfigure(const char* str);
WT_CONNECTION* getConnection() {
return _conn;
}
void dropSomeQueuedIdents();
std::list<WiredTigerCachedCursor> filterCursorsWithQueuedDrops(
std::list<WiredTigerCachedCursor>* cache);
bool haveDropsQueued() const;
void syncSizeInfo(bool sync) const;
/*
* An oplog manager is always accessible, but this method will start the background thread to
* control oplog entry visibility for reads.
*
* On mongod, the background thread will be started when the first oplog record store is
* created, and stopped when the last oplog record store is destroyed, at shutdown time. For
* unit tests, the background thread may be started and stopped multiple times as tests create
* and destroy oplog record stores.
*/
void startOplogManager(OperationContext* opCtx,
const std::string& uri,
WiredTigerRecordStore* oplogRecordStore);
void haltOplogManager();
/*
* Always returns a non-nil pointer. However, the WiredTigerOplogManager may not have been
* initialized and its background refreshing thread may not be running.
*
* A caller that wants to get the oplog read timestamp, or call
* `waitForAllEarlierOplogWritesToBeVisible`, is advised to first see if the oplog manager is
* running with a call to `isRunning`.
*
* A caller that simply wants to call `triggerJournalFlush` may do so without concern.
*/
WiredTigerOplogManager* getOplogManager() const {
return _oplogManager.get();
}
/*
* This function is called when replication has completed a batch. In this function, we
* refresh our oplog visiblity read-at-timestamp value.
*/
void replicationBatchIsComplete() const override;
/**
* Sets the implementation for `initRsOplogBackgroundThread` (allowing tests to skip the
* background job, for example). Intended to be called from a MONGO_INITIALIZER and therefroe in
* a single threaded context.
*/
static void setInitRsOplogBackgroundThreadCallback(stdx::function<bool(StringData)> cb);
/**
* Initializes a background job to remove excess documents in the oplog collections.
* This applies to the capped collections in the local.oplog.* namespaces (specifically
* local.oplog.rs for replica sets and local.oplog.$main for master/slave replication).
* Returns true if a background job is running for the namespace.
*/
static bool initRsOplogBackgroundThread(StringData ns);
static void appendGlobalStats(BSONObjBuilder& b);
private:
class WiredTigerJournalFlusher;
class WiredTigerCheckpointThread;
Status _salvageIfNeeded(const char* uri);
void _checkIdentPath(StringData ident);
bool _hasUri(WT_SESSION* session, const std::string& uri) const;
std::string _uri(StringData ident) const;
void _setOldestTimestamp(Timestamp oldestTimestamp, bool force = false);
WT_CONNECTION* _conn;
WT_EVENT_HANDLER _eventHandler;
std::unique_ptr<WiredTigerSessionCache> _sessionCache;
ClockSource* const _clockSource;
// Mutex to protect use of _oplogManagerCount by this instance of KV engine.
mutable stdx::mutex _oplogManagerMutex;
std::size_t _oplogManagerCount = 0;
std::unique_ptr<WiredTigerOplogManager> _oplogManager;
std::string _canonicalName;
std::string _path;
std::string _wtOpenConfig;
std::unique_ptr<WiredTigerSizeStorer> _sizeStorer;
std::string _sizeStorerUri;
mutable ElapsedTracker _sizeStorerSyncTracker;
bool _durable;
bool _ephemeral;
bool _readOnly;
std::unique_ptr<WiredTigerJournalFlusher> _journalFlusher; // Depends on _sizeStorer
std::unique_ptr<WiredTigerCheckpointThread> _checkpointThread;
std::string _rsOptions;
std::string _indexOptions;
mutable stdx::mutex _dropAllQueuesMutex;
mutable stdx::mutex _identToDropMutex;
std::list<std::string> _identToDrop;
mutable Date_t _previousCheckedDropsQueued;
std::unique_ptr<WiredTigerSession> _backupSession;
};
}
|