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
|
// dbhash.cpp
/**
* Copyright (C) 2013-2014 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::kCommand
#include "mongo/platform/basic.h"
#include <boost/optional.hpp>
#include <map>
#include <string>
#include "mongo/db/catalog/collection.h"
#include "mongo/db/catalog/database.h"
#include "mongo/db/catalog/database_catalog_entry.h"
#include "mongo/db/catalog/index_catalog.h"
#include "mongo/db/commands.h"
#include "mongo/db/db_raii.h"
#include "mongo/db/exec/working_set_common.h"
#include "mongo/db/namespace_string.h"
#include "mongo/db/query/internal_plans.h"
#include "mongo/db/transaction_participant.h"
#include "mongo/stdx/mutex.h"
#include "mongo/util/log.h"
#include "mongo/util/md5.hpp"
#include "mongo/util/net/socket_utils.h"
#include "mongo/util/timer.h"
namespace mongo {
namespace {
class DBHashCmd : public ErrmsgCommandDeprecated {
public:
DBHashCmd() : ErrmsgCommandDeprecated("dbHash", "dbhash") {}
virtual bool supportsWriteConcern(const BSONObj& cmd) const override {
return false;
}
ReadWriteType getReadWriteType() const override {
return ReadWriteType::kRead;
}
bool supportsReadConcern(const std::string& dbName,
const BSONObj& cmdObj,
repl::ReadConcernLevel level) const override {
return level == repl::ReadConcernLevel::kLocalReadConcern ||
level == repl::ReadConcernLevel::kSnapshotReadConcern;
}
AllowedOnSecondary secondaryAllowed(ServiceContext*) const override {
return AllowedOnSecondary::kAlways;
}
virtual void addRequiredPrivileges(const std::string& dbname,
const BSONObj& cmdObj,
std::vector<Privilege>* out) const {
ActionSet actions;
actions.addAction(ActionType::dbHash);
out->push_back(Privilege(ResourcePattern::forDatabaseName(dbname), actions));
}
virtual bool errmsgRun(OperationContext* opCtx,
const std::string& dbname,
const BSONObj& cmdObj,
std::string& errmsg,
BSONObjBuilder& result) {
Timer timer;
std::set<std::string> desiredCollections;
if (cmdObj["collections"].type() == Array) {
BSONObjIterator i(cmdObj["collections"].Obj());
while (i.more()) {
BSONElement e = i.next();
if (e.type() != String) {
errmsg = "collections entries have to be strings";
return false;
}
desiredCollections.insert(e.String());
}
}
const std::string ns = parseNs(dbname, cmdObj);
uassert(ErrorCodes::InvalidNamespace,
str::stream() << "Invalid db name: " << ns,
NamespaceString::validDBName(ns, NamespaceString::DollarInDbNameBehavior::Allow));
// We lock the entire database in S-mode in order to ensure that the contents will not
// change for the snapshot.
auto lockMode = LockMode::MODE_S;
auto txnParticipant = TransactionParticipant::get(opCtx);
if (txnParticipant && txnParticipant->inMultiDocumentTransaction()) {
// However, if we are inside a multi-statement transaction, then we only need to lock
// the database in intent mode to ensure that none of the collections get dropped.
lockMode = getLockModeForQuery(opCtx);
}
AutoGetDb autoDb(opCtx, ns, lockMode);
Database* db = autoDb.getDb();
std::list<std::string> colls;
if (db) {
db->getDatabaseCatalogEntry()->getCollectionNamespaces(&colls);
colls.sort();
}
result.append("host", prettyHostName());
md5_state_t globalState;
md5_init(&globalState);
// A set of 'system' collections that are replicated, and therefore included in the db hash.
const std::set<StringData> replicatedSystemCollections{"system.backup_users",
"system.js",
"system.new_users",
"system.roles",
"system.users",
"system.version",
"system.views"};
BSONArrayBuilder cappedCollections;
BSONObjBuilder collectionsByUUID;
BSONObjBuilder bb(result.subobjStart("collections"));
for (const auto& collectionName : colls) {
NamespaceString collNss(collectionName);
if (collNss.size() - 1 <= dbname.size()) {
errmsg = str::stream() << "weird fullCollectionName [" << collNss.toString() << "]";
return false;
}
// Only include 'system' collections that are replicated.
bool isReplicatedSystemColl =
(replicatedSystemCollections.count(collNss.coll().toString()) > 0);
if (collNss.isSystem() && !isReplicatedSystemColl)
continue;
if (collNss.coll().startsWith("tmp.mr.")) {
// We skip any incremental map reduce collections as they also aren't replicated.
continue;
}
if (desiredCollections.size() > 0 &&
desiredCollections.count(collNss.coll().toString()) == 0)
continue;
// Don't include 'drop pending' collections.
if (collNss.isDropPendingNamespace())
continue;
if (Collection* collection = db->getCollection(opCtx, collectionName)) {
if (collection->isCapped()) {
cappedCollections.append(collNss.coll());
}
if (OptionalCollectionUUID uuid = collection->uuid()) {
uuid->appendToBuilder(&collectionsByUUID, collNss.coll());
}
}
// Compute the hash for this collection.
std::string hash = _hashCollection(opCtx, db, collNss.toString());
bb.append(collNss.coll(), hash);
md5_append(&globalState, (const md5_byte_t*)hash.c_str(), hash.size());
}
bb.done();
result.append("capped", BSONArray(cappedCollections.done()));
result.append("uuids", collectionsByUUID.done());
md5digest d;
md5_finish(&globalState, d);
std::string hash = digestToString(d);
result.append("md5", hash);
result.appendNumber("timeMillis", timer.millis());
return 1;
}
private:
std::string _hashCollection(OperationContext* opCtx,
Database* db,
const std::string& fullCollectionName) {
NamespaceString ns(fullCollectionName);
Collection* collection = db->getCollection(opCtx, ns);
if (!collection)
return "";
boost::optional<Lock::CollectionLock> collLock;
auto txnParticipant = TransactionParticipant::get(opCtx);
if (txnParticipant && txnParticipant->inMultiDocumentTransaction()) {
// When inside a multi-statement transaction, we are only holding the database lock in
// intent mode. We need to also acquire the collection lock in intent mode to ensure
// reading from the consistent snapshot doesn't overlap with any catalog operations on
// the collection.
invariant(
opCtx->lockState()->isDbLockedForMode(db->name(), getLockModeForQuery(opCtx)));
collLock.emplace(opCtx->lockState(), fullCollectionName, getLockModeForQuery(opCtx));
auto minSnapshot = collection->getMinimumVisibleSnapshot();
auto mySnapshot = opCtx->recoveryUnit()->getPointInTimeReadTimestamp();
invariant(mySnapshot);
uassert(ErrorCodes::SnapshotUnavailable,
str::stream() << "Unable to read from a snapshot due to pending collection"
" catalog changes; please retry the operation. Snapshot"
" timestamp is "
<< mySnapshot->toString()
<< ". Collection minimum timestamp is "
<< minSnapshot->toString(),
!minSnapshot || *mySnapshot >= *minSnapshot);
} else {
invariant(opCtx->lockState()->isDbLockedForMode(db->name(), MODE_S));
}
IndexDescriptor* desc = collection->getIndexCatalog()->findIdIndex(opCtx);
std::unique_ptr<PlanExecutor, PlanExecutor::Deleter> exec;
if (desc) {
exec = InternalPlanner::indexScan(opCtx,
collection,
desc,
BSONObj(),
BSONObj(),
BoundInclusion::kIncludeStartKeyOnly,
PlanExecutor::NO_YIELD,
InternalPlanner::FORWARD,
InternalPlanner::IXSCAN_FETCH);
} else if (collection->isCapped()) {
exec = InternalPlanner::collectionScan(
opCtx, fullCollectionName, collection, PlanExecutor::NO_YIELD);
} else {
log() << "can't find _id index for: " << fullCollectionName;
return "no _id _index";
}
md5_state_t st;
md5_init(&st);
long long n = 0;
PlanExecutor::ExecState state;
BSONObj c;
verify(NULL != exec.get());
while (PlanExecutor::ADVANCED == (state = exec->getNext(&c, NULL))) {
md5_append(&st, (const md5_byte_t*)c.objdata(), c.objsize());
n++;
}
if (PlanExecutor::IS_EOF != state) {
warning() << "error while hashing, db dropped? ns=" << fullCollectionName;
uasserted(34371,
"Plan executor error while running dbHash command: " +
WorkingSetCommon::toStatusString(c));
}
md5digest d;
md5_finish(&st, d);
std::string hash = digestToString(d);
return hash;
}
} dbhashCmd;
} // namespace
} // namespace mongo
|