summaryrefslogtreecommitdiff
path: root/src/mongo/db/exec/delete.cpp
blob: b034df535fc0f9944ea51e3cf31ae97b0fef9c08 (plain)
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
/**
 *    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.
 */

#define MONGO_LOG_DEFAULT_COMPONENT ::mongo::logger::LogComponent::kWrite

#include "mongo/platform/basic.h"

#include "mongo/db/exec/delete.h"

#include "mongo/db/catalog/collection.h"
#include "mongo/db/concurrency/write_conflict_exception.h"
#include "mongo/db/exec/scoped_timer.h"
#include "mongo/db/exec/working_set_common.h"
#include "mongo/db/service_context.h"
#include "mongo/db/op_observer.h"
#include "mongo/db/query/canonical_query.h"
#include "mongo/db/repl/replication_coordinator_global.h"
#include "mongo/util/log.h"
#include "mongo/util/scopeguard.h"

namespace mongo {

    using std::auto_ptr;
    using std::vector;

    // static
    const char* DeleteStage::kStageType = "DELETE";

    DeleteStage::DeleteStage(OperationContext* txn,
                             const DeleteStageParams& params,
                             WorkingSet* ws,
                             Collection* collection,
                             PlanStage* child)
        : _txn(txn),
          _params(params),
          _ws(ws),
          _collection(collection),
          _child(child),
          _idRetrying(WorkingSet::INVALID_ID),
          _idReturning(WorkingSet::INVALID_ID),
          _commonStats(kStageType) { }

    DeleteStage::~DeleteStage() {}

    bool DeleteStage::isEOF() {
        if (!_collection) {
            return true;
        }
        if (!_params.isMulti && _specificStats.docsDeleted > 0) {
            return true;
        }
        return _idRetrying == WorkingSet::INVALID_ID
            && _idReturning == WorkingSet::INVALID_ID
            && _child->isEOF();
    }

    PlanStage::StageState DeleteStage::work(WorkingSetID* out) {
        ++_commonStats.works;

        // Adds the amount of time taken by work() to executionTimeMillis.
        ScopedTimer timer(&_commonStats.executionTimeMillis);

        if (isEOF()) { return PlanStage::IS_EOF; }
        invariant(_collection); // If isEOF() returns false, we must have a collection.

        // It is possible that after a delete was executed, a WriteConflictException occurred
        // and prevented us from returning ADVANCED with the old version of the document.
        if (_idReturning != WorkingSet::INVALID_ID) {
            // We should only get here if we were trying to return something before.
            invariant(_params.returnDeleted);

            WorkingSetMember* member = _ws->get(_idReturning);
            invariant(member->state == WorkingSetMember::OWNED_OBJ);

            *out = _idReturning;
            _idReturning = WorkingSet::INVALID_ID;
            ++_commonStats.advanced;
            return PlanStage::ADVANCED;
         }

        // Either retry the last WSM we worked on or get a new one from our child.
        WorkingSetID id;
        StageState status;
        if (_idRetrying == WorkingSet::INVALID_ID) {
            status = _child->work(&id);
        }
        else {
            status = ADVANCED;
            id = _idRetrying;
            _idRetrying = WorkingSet::INVALID_ID;
        }

        if (PlanStage::ADVANCED == status) {
            WorkingSetMember* member = _ws->get(id);

            // We want to free this member when we return, unless we need to retry it.
            ScopeGuard memberFreer = MakeGuard(&WorkingSet::free, _ws, id);

            if (!member->hasLoc()) {
                // We expect to be here because of an invalidation causing a force-fetch, and
                // doc-locking storage engines do not issue invalidations.
                ++_specificStats.nInvalidateSkips;
                ++_commonStats.needTime;
                return PlanStage::NEED_TIME;
            }
            RecordId rloc = member->loc;
            // Deletes can't have projections. This means that covering analysis will always add
            // a fetch. We should always get fetched data, and never just key data.
            invariant(member->hasObj());

            try {
                // If the snapshot changed, then we have to make sure we have the latest copy of the
                // doc and that it still matches.
                std::unique_ptr<RecordCursor> cursor;
                if (_txn->recoveryUnit()->getSnapshotId() != member->obj.snapshotId()) {
                    cursor = _collection->getCursor(_txn);
                    if (!WorkingSetCommon::fetch(_txn, member, cursor)) {
                        // Doc is already deleted. Nothing more to do.
                        ++_commonStats.needTime;
                        return PlanStage::NEED_TIME;
                    }

                    // Make sure the re-fetched doc still matches the predicate.
                    if (_params.canonicalQuery &&
                        !_params.canonicalQuery->root()->matchesBSON(member->obj.value(), NULL)) {
                        // Doesn't match.
                        ++_commonStats.needTime;
                        return PlanStage::NEED_TIME;
                    }
                }

                // TODO: Do we want to buffer docs and delete them in a group rather than
                // saving/restoring state repeatedly?

                try {
                    _child->saveState();
                    if (supportsDocLocking()) {
                        // Doc-locking engines require this after saveState() since they don't use
                        // invalidations.
                        WorkingSetCommon::prepareForSnapshotChange(_ws);
                    }
                }
                catch ( const WriteConflictException& wce ) {
                    std::terminate();
                }

                if (_params.returnDeleted) {
                    // Save a copy of the document that is about to get deleted.
                    BSONObj deletedDoc = member->obj.value();
                    member->obj.setValue(deletedDoc.getOwned());
                    member->loc = RecordId();
                    member->state = WorkingSetMember::OWNED_OBJ;
                }

                // Do the write, unless this is an explain.
                if (!_params.isExplain) {
                    WriteUnitOfWork wunit(_txn);

                    const bool deleteCappedOK = false;
                    const bool deleteNoWarn = false;
                    BSONObj deletedId;

                    _collection->deleteDocument(_txn, rloc, deleteCappedOK, deleteNoWarn,
                                                _params.shouldCallLogOp ? &deletedId : NULL);

                    wunit.commit();
                }

                ++_specificStats.docsDeleted;
            }
            catch ( const WriteConflictException& wce ) {
                _idRetrying = id;
                memberFreer.Dismiss(); // Keep this member around so we can retry deleting it.
                *out = WorkingSet::INVALID_ID;
                _commonStats.needYield++;
                return NEED_YIELD;
            }

            //  As restoreState may restore (recreate) cursors, cursors are tied to the
            //  transaction in which they are created, and a WriteUnitOfWork is a
            //  transaction, make sure to restore the state outside of the WritUnitOfWork.
            try {
                _child->restoreState(_txn);
            }
            catch ( const WriteConflictException& wce ) {
                // Note we don't need to retry anything in this case since the delete already
                // was committed. However, we still need to return the deleted document
                // (if it was requested).
                if (_params.returnDeleted) {
                    // member->obj should refer to the deleted document.
                    invariant(member->state == WorkingSetMember::OWNED_OBJ);

                    _idReturning = id;
                    // Keep this member around so that we can return it on the next work() call.
                    memberFreer.Dismiss();
                }
                *out = WorkingSet::INVALID_ID;
                _commonStats.needYield++;
                return NEED_YIELD;
            }

            if (_params.returnDeleted) {
                // member->obj should refer to the deleted document.
                invariant(member->state == WorkingSetMember::OWNED_OBJ);

                memberFreer.Dismiss();  // Keep this member around so we can return it.
                *out = id;
                ++_commonStats.advanced;
                return PlanStage::ADVANCED;
            }

            ++_commonStats.needTime;
            return PlanStage::NEED_TIME;
        }
        else if (PlanStage::FAILURE == status || PlanStage::DEAD == status) {
            *out = id;
            // If a stage fails, it may create a status WSM to indicate why it failed, in which case
            // 'id' is valid.  If ID is invalid, we create our own error message.
            if (WorkingSet::INVALID_ID == id) {
                const std::string errmsg = "delete stage failed to read in results from child";
                *out = WorkingSetCommon::allocateStatusMember(_ws, Status(ErrorCodes::InternalError,
                                                                          errmsg));
            }
            return status;
        }
        else if (PlanStage::NEED_TIME == status) {
            ++_commonStats.needTime;
        }
        else if (PlanStage::NEED_YIELD == status) {
            *out = id;
            ++_commonStats.needYield;
        }

        return status;
    }

    void DeleteStage::saveState() {
        _txn = NULL;
        ++_commonStats.yields;
        _child->saveState();
    }

    void DeleteStage::restoreState(OperationContext* opCtx) {
        invariant(_txn == NULL);
        _txn = opCtx;
        ++_commonStats.unyields;
        _child->restoreState(opCtx);

        const NamespaceString& ns(_collection->ns());
        massert(28537,
                str::stream() << "Demoted from primary while removing from " << ns.ns(),
                !_params.shouldCallLogOp ||
                repl::getGlobalReplicationCoordinator()->canAcceptWritesFor(ns));
    }

    void DeleteStage::invalidate(OperationContext* txn, const RecordId& dl, InvalidationType type) {
        ++_commonStats.invalidates;
        _child->invalidate(txn, dl, type);
    }

    vector<PlanStage*> DeleteStage::getChildren() const {
        vector<PlanStage*> children;
        children.push_back(_child.get());
        return children;
    }

    PlanStageStats* DeleteStage::getStats() {
        _commonStats.isEOF = isEOF();
        auto_ptr<PlanStageStats> ret(new PlanStageStats(_commonStats, STAGE_DELETE));
        ret->specific.reset(new DeleteStats(_specificStats));
        ret->children.push_back(_child->getStats());
        return ret.release();
    }

    const CommonStats* DeleteStage::getCommonStats() const {
        return &_commonStats;
    }

    const SpecificStats* DeleteStage::getSpecificStats() const {
        return &_specificStats;
    }

    // static
    long long DeleteStage::getNumDeleted(PlanExecutor* exec) {
        invariant(exec->getRootStage()->isEOF());
        invariant(exec->getRootStage()->stageType() == STAGE_DELETE);
        DeleteStage* deleteStage = static_cast<DeleteStage*>(exec->getRootStage());
        const DeleteStats* deleteStats =
            static_cast<const DeleteStats*>(deleteStage->getSpecificStats());
        return deleteStats->docsDeleted;
    }

}  // namespace mongo