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
|
/**
* Copyright (C) 2018-present MongoDB, Inc.
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the Server Side Public License, version 1,
* as published by MongoDB, Inc.
*
* 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
* Server Side Public License for more details.
*
* You should have received a copy of the Server Side Public License
* along with this program. If not, see
* <http://www.mongodb.com/licensing/server-side-public-license>.
*
* 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 Server Side 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::kStorage
#include "mongo/platform/basic.h"
#include <algorithm>
#include "mongo/db/repair_database.h"
#include "mongo/base/status.h"
#include "mongo/base/string_data.h"
#include "mongo/bson/bson_validate.h"
#include "mongo/bson/bsonobjbuilder.h"
#include "mongo/db/background.h"
#include "mongo/db/catalog/collection.h"
#include "mongo/db/catalog/collection_catalog.h"
#include "mongo/db/catalog/collection_validation.h"
#include "mongo/db/catalog/database.h"
#include "mongo/db/catalog/database_holder.h"
#include "mongo/db/catalog/document_validation.h"
#include "mongo/db/catalog/index_key_validate.h"
#include "mongo/db/catalog/multi_index_block.h"
#include "mongo/db/concurrency/write_conflict_exception.h"
#include "mongo/db/db_raii.h"
#include "mongo/db/index/index_descriptor.h"
#include "mongo/db/index_builds_coordinator.h"
#include "mongo/db/logical_clock.h"
#include "mongo/db/namespace_string.h"
#include "mongo/db/query/query_knobs_gen.h"
#include "mongo/db/rebuild_indexes.h"
#include "mongo/db/storage/durable_catalog.h"
#include "mongo/db/storage/storage_engine.h"
#include "mongo/util/log.h"
#include "mongo/util/scopeguard.h"
namespace mongo {
Status rebuildIndexesForNamespace(OperationContext* opCtx,
const NamespaceString& nss,
StorageEngine* engine) {
opCtx->checkForInterrupt();
auto collection = CollectionCatalog::get(opCtx).lookupCollectionByNamespace(opCtx, nss);
auto swIndexNameObjs = getIndexNameObjs(opCtx, collection->getCatalogId());
if (!swIndexNameObjs.isOK())
return swIndexNameObjs.getStatus();
std::vector<BSONObj> indexSpecs = swIndexNameObjs.getValue().second;
Status status = rebuildIndexesOnCollection(opCtx, collection, indexSpecs, RepairData::kYes);
if (!status.isOK())
return status;
engine->flushAllFiles(opCtx, true);
return Status::OK();
}
namespace {
Status repairCollections(OperationContext* opCtx,
StorageEngine* engine,
const std::string& dbName) {
auto colls = CollectionCatalog::get(opCtx).getAllCollectionNamesFromDb(opCtx, dbName);
for (const auto& nss : colls) {
opCtx->checkForInterrupt();
log() << "Repairing collection " << nss;
auto collection = CollectionCatalog::get(opCtx).lookupCollectionByNamespace(opCtx, nss);
Status status = engine->repairRecordStore(opCtx, collection->getCatalogId(), nss);
// If data was modified during repairRecordStore, we know to rebuild indexes without needing
// to run an expensive collection validation.
if (status.code() == ErrorCodes::DataModifiedByRepair) {
Status status = rebuildIndexesForNamespace(opCtx, nss, engine);
if (!status.isOK()) {
return status;
}
continue;
} else if (!status.isOK()) {
return status;
}
// Run collection validation to avoid unecessarily rebuilding indexes on valid collections
// with consistent indexes. Initialize the collection prior to validation. Need to lookup
// from catalog again because the old collection object was invalidated by
// repairRecordStore.
collection = CollectionCatalog::get(opCtx).lookupCollectionByNamespace(opCtx, nss);
collection->init(opCtx);
ValidateResults validateResults;
BSONObjBuilder output;
// Set options to exclude FullRecordStoreValidation because we have already validated the
// underlying record store in the call to repairRecordStore above.
auto options = CollectionValidation::ValidateOptions::kFullIndexValidation;
const bool background = false;
status = CollectionValidation::validate(
opCtx, nss, options, background, &validateResults, &output);
if (!status.isOK()) {
return status;
}
log() << "Collection validation results: " << output.done();
if (!validateResults.valid) {
status = rebuildIndexesForNamespace(opCtx, nss, engine);
if (!status.isOK()) {
return status;
}
}
}
return Status::OK();
}
} // namespace
Status repairDatabase(OperationContext* opCtx, StorageEngine* engine, const std::string& dbName) {
DisableDocumentValidation validationDisabler(opCtx);
// We must hold some form of lock here
invariant(opCtx->lockState()->isW());
invariant(dbName.find('.') == std::string::npos);
log() << "repairDatabase " << dbName;
BackgroundOperation::assertNoBgOpInProgForDb(dbName);
opCtx->checkForInterrupt();
// Close the db and invalidate all current users and caches.
auto databaseHolder = DatabaseHolder::get(opCtx);
databaseHolder->close(opCtx, dbName);
// Reopening db is necessary for repairCollections.
auto db = databaseHolder->openDb(opCtx, dbName);
auto status = repairCollections(opCtx, engine, dbName);
if (!status.isOK()) {
severe() << "Failed to repair database " << dbName << ": " << status.reason();
}
try {
// Ensure that we don't trigger an exception when attempting to take locks.
UninterruptibleLockGuard noInterrupt(opCtx->lockState());
// Set the minimum snapshot for all Collections in this db. This ensures that readers
// using majority readConcern level can only use the collections after their repaired
// versions are in the committed view.
auto clusterTime = LogicalClock::getClusterTimeForReplicaSet(opCtx).asTimestamp();
for (auto collIt = db->begin(opCtx); collIt != db->end(opCtx); ++collIt) {
auto collection = *collIt;
if (collection) {
collection->setMinimumVisibleSnapshot(clusterTime);
}
}
// Restore oplog Collection pointer cache.
repl::acquireOplogCollectionForLogging(opCtx);
} catch (const ExceptionFor<ErrorCodes::MustDowngrade>&) {
// openDb can throw an exception with a MustDowngrade status if a collection does not
// have a UUID.
throw;
} catch (...) {
severe() << "Unexpected exception encountered while reopening database after repair.";
std::terminate(); // Logs additional info about the specific error.
}
return status;
}
} // namespace mongo
|
