summaryrefslogtreecommitdiff
path: root/src/mongo/s/commands/cluster_command_test_fixture.cpp
blob: 504fe0602e3b91d7d4fed61cb265bcfb26f65e47 (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
/**
 *    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::kDefault

#include "mongo/platform/basic.h"

#include "mongo/s/commands/cluster_command_test_fixture.h"

#include "mongo/db/commands.h"
#include "mongo/db/commands/test_commands_enabled.h"
#include "mongo/db/commands/txn_cmds_gen.h"
#include "mongo/db/keys_collection_client_sharded.h"
#include "mongo/db/keys_collection_manager.h"
#include "mongo/db/logical_clock.h"
#include "mongo/db/logical_session_cache_noop.h"
#include "mongo/db/logical_time_validator.h"
#include "mongo/s/cluster_last_error_info.h"
#include "mongo/util/fail_point_service.h"
#include "mongo/util/log.h"
#include "mongo/util/tick_source_mock.h"

namespace mongo {

void ClusterCommandTestFixture::setUp() {
    CatalogCacheTestFixture::setUp();
    CatalogCacheTestFixture::setupNShards(numShards);

    // Set up a logical clock with an initial time.
    auto logicalClock = std::make_unique<LogicalClock>(getServiceContext());
    logicalClock->setClusterTimeFromTrustedSource(kInMemoryLogicalTime);
    LogicalClock::set(getServiceContext(), std::move(logicalClock));

    auto keysCollectionClient = std::make_unique<KeysCollectionClientSharded>(
        Grid::get(operationContext())->catalogClient());

    auto keyManager = std::make_shared<KeysCollectionManager>(
        "dummy", std::move(keysCollectionClient), Seconds(KeysRotationIntervalSec));

    auto validator = std::make_unique<LogicalTimeValidator>(keyManager);
    LogicalTimeValidator::set(getServiceContext(), std::move(validator));

    LogicalSessionCache::set(getServiceContext(), std::make_unique<LogicalSessionCacheNoop>());

    // Set up a tick source for transaction metrics.
    auto tickSource = std::make_unique<TickSourceMock<Microseconds>>();
    tickSource->reset(1);
    getServiceContext()->setTickSource(std::move(tickSource));

    loadRoutingTableWithTwoChunksAndTwoShards(kNss);

    _staleVersionAndSnapshotRetriesBlock = std::make_unique<FailPointEnableBlock>(
        "enableStaleVersionAndSnapshotRetriesWithinTransactions");
}

BSONObj ClusterCommandTestFixture::_makeCmd(BSONObj cmdObj, bool includeAfterClusterTime) {
    BSONObjBuilder bob(cmdObj);
    // Each command runs in a new session.
    bob.append("lsid", makeLogicalSessionIdForTest().toBSON());
    bob.append("txnNumber", TxnNumber(1));
    bob.append("autocommit", false);
    bob.append("startTransaction", true);

    BSONObjBuilder readConcernBob = bob.subobjStart(repl::ReadConcernArgs::kReadConcernFieldName);
    readConcernBob.append("level", "snapshot");
    if (includeAfterClusterTime) {
        readConcernBob.append("afterClusterTime", kAfterClusterTime);
    }

    readConcernBob.doneFast();
    return bob.obj();
}

void ClusterCommandTestFixture::expectReturnsError(ErrorCodes::Error code) {
    onCommandForPoolExecutor([code](const executor::RemoteCommandRequest& request) {
        BSONObjBuilder resBob;
        CommandHelpers::appendCommandStatusNoThrow(resBob, Status(code, "dummy error"));
        return resBob.obj();
    });
}

DbResponse ClusterCommandTestFixture::runCommand(BSONObj cmd) {
    // Create a new client/operation context per command
    auto client = getServiceContext()->makeClient("ClusterCmdClient");
    auto opCtx = client->makeOperationContext();

    const auto opMsgRequest = OpMsgRequest::fromDBAndBody(kNss.db(), cmd);

    // Ensure the clusterGLE on the Client has not yet been initialized.
    ASSERT(!ClusterLastErrorInfo::get(client.get()));

    // Initialize the cluster last error info for the client with a new request.
    ClusterLastErrorInfo::get(client.get()) = std::make_shared<ClusterLastErrorInfo>();
    ASSERT(ClusterLastErrorInfo::get(client.get()));
    auto clusterGLE = ClusterLastErrorInfo::get(client.get());
    clusterGLE->newRequest();

    return Strategy::clientCommand(opCtx.get(), opMsgRequest.serialize());
}

void ClusterCommandTestFixture::runCommandSuccessful(BSONObj cmd, bool isTargeted) {
    auto future = launchAsync([&] {
        // Shouldn't throw.
        runCommand(cmd);
    });

    size_t numMocks = isTargeted ? 1 : numShards;
    for (size_t i = 0; i < numMocks; i++) {
        expectReturnsSuccess(i % numShards);
    }

    future.default_timed_get();
}

void ClusterCommandTestFixture::runTxnCommandOneError(BSONObj cmd,
                                                      ErrorCodes::Error code,
                                                      bool isTargeted) {
    auto future = launchAsync([&] {
        // Shouldn't throw.
        runCommand(cmd);
    });

    size_t numMocks = isTargeted ? 1 : numShards;
    for (size_t i = 0; i < numMocks; i++) {
        expectReturnsError(code);
    }

    // In a transaction, when the router encounters a retryable error it sends abortTransaction to
    // each pending participant shard before retrying.
    for (size_t i = 0; i < numMocks; i++) {
        expectAbortTransaction();
    }

    for (size_t i = 0; i < numMocks; i++) {
        expectReturnsSuccess(i % numShards);
    }

    future.default_timed_get();
}

void ClusterCommandTestFixture::runCommandInspectRequests(BSONObj cmd,
                                                          InspectionCallback cb,
                                                          bool isTargeted) {
    auto future = launchAsync([&] { runCommand(cmd); });

    size_t numMocks = isTargeted ? 1 : numShards;
    for (size_t i = 0; i < numMocks; i++) {
        expectInspectRequest(i % numShards, cb);
    }

    future.default_timed_get();
}

void ClusterCommandTestFixture::expectAbortTransaction() {
    onCommandForPoolExecutor([this](const executor::RemoteCommandRequest& request) {
        auto cmdName = request.cmdObj.firstElement().fieldNameStringData();
        ASSERT_EQ(cmdName, "abortTransaction");

        BSONObjBuilder bob;
        bob.append("ok", 1);
        appendTxnResponseMetadata(bob);
        return bob.obj();
    });
}

void ClusterCommandTestFixture::runTxnCommandMaxErrors(BSONObj cmd,
                                                       ErrorCodes::Error code,
                                                       bool isTargeted) {
    auto future = launchAsync([&] { runCommand(cmd); });

    size_t numTargetedShards = isTargeted ? 1 : numShards;
    for (size_t i = 0; i < kMaxNumStaleVersionRetries - 1; i++) {
        for (size_t j = 0; j < numTargetedShards; j++) {
            expectReturnsError(code);
        }

        // In a transaction, when the router encounters a retryable error it sends abortTransaction
        // to each pending participant shard before retrying.
        for (size_t j = 0; j < numTargetedShards; j++) {
            expectAbortTransaction();
        }
    }

    // The router should exhaust its retries here.
    for (size_t j = 0; j < numTargetedShards; j++) {
        expectReturnsError(code);
    }

    // In a transaction, each targeted shard is sent abortTransaction when the router exhausts its
    // retries.
    for (size_t i = 0; i < numTargetedShards; i++) {
        expectAbortTransaction();
    }

    future.default_timed_get();
}

void ClusterCommandTestFixture::testNoErrors(BSONObj targetedCmd, BSONObj scatterGatherCmd) {

    // Target one shard.
    runCommandSuccessful(_makeCmd(targetedCmd), true);

    // Target all shards.
    if (!scatterGatherCmd.isEmpty()) {
        runCommandSuccessful(_makeCmd(scatterGatherCmd), false);
    }
}

void ClusterCommandTestFixture::testRetryOnSnapshotError(BSONObj targetedCmd,
                                                         BSONObj scatterGatherCmd) {
    // Target one shard.
    runTxnCommandOneError(_makeCmd(targetedCmd), ErrorCodes::SnapshotUnavailable, true);
    runTxnCommandOneError(_makeCmd(targetedCmd), ErrorCodes::SnapshotTooOld, true);

    // Target all shards
    if (!scatterGatherCmd.isEmpty()) {
        runTxnCommandOneError(_makeCmd(scatterGatherCmd), ErrorCodes::SnapshotUnavailable, false);
        runTxnCommandOneError(_makeCmd(scatterGatherCmd), ErrorCodes::SnapshotTooOld, false);
    }
}

void ClusterCommandTestFixture::testMaxRetriesSnapshotErrors(BSONObj targetedCmd,
                                                             BSONObj scatterGatherCmd) {
    // Target one shard.
    runTxnCommandMaxErrors(_makeCmd(targetedCmd), ErrorCodes::SnapshotUnavailable, true);
    runTxnCommandMaxErrors(_makeCmd(targetedCmd), ErrorCodes::SnapshotTooOld, true);

    // Target all shards
    if (!scatterGatherCmd.isEmpty()) {
        runTxnCommandMaxErrors(_makeCmd(scatterGatherCmd), ErrorCodes::SnapshotUnavailable, false);
        runTxnCommandMaxErrors(_makeCmd(scatterGatherCmd), ErrorCodes::SnapshotTooOld, false);
    }
}

void ClusterCommandTestFixture::testAttachesAtClusterTimeForSnapshotReadConcern(
    BSONObj targetedCmd, BSONObj scatterGatherCmd) {

    auto containsAtClusterTime = [](const executor::RemoteCommandRequest& request) {
        ASSERT(!request.cmdObj["readConcern"]["atClusterTime"].eoo());
    };

    // Target one shard.
    runCommandInspectRequests(_makeCmd(targetedCmd), containsAtClusterTime, true);

    // Target all shards.
    if (!scatterGatherCmd.isEmpty()) {
        runCommandInspectRequests(_makeCmd(scatterGatherCmd), containsAtClusterTime, false);
    }
}

void ClusterCommandTestFixture::testSnapshotReadConcernWithAfterClusterTime(
    BSONObj targetedCmd, BSONObj scatterGatherCmd) {

    auto containsAtClusterTimeNoAfterClusterTime =
        [&](const executor::RemoteCommandRequest& request) {
            ASSERT(!request.cmdObj["readConcern"]["atClusterTime"].eoo());
            ASSERT(request.cmdObj["readConcern"]["afterClusterTime"].eoo());

            // The chosen atClusterTime should be greater than or equal to the request's
            // afterClusterTime.
            ASSERT_GTE(LogicalTime(request.cmdObj["readConcern"]["atClusterTime"].timestamp()),
                       LogicalTime(kAfterClusterTime));
        };

    // Target one shard.
    runCommandInspectRequests(
        _makeCmd(targetedCmd, true), containsAtClusterTimeNoAfterClusterTime, true);

    // Target all shards.
    if (!scatterGatherCmd.isEmpty()) {
        runCommandInspectRequests(
            _makeCmd(scatterGatherCmd, true), containsAtClusterTimeNoAfterClusterTime, false);
    }
}

void ClusterCommandTestFixture::appendTxnResponseMetadata(BSONObjBuilder& bob) {
    // Set readOnly to false to avoid opting in to the read-only optimization.
    TxnResponseMetadata txnResponseMetadata(false);
    txnResponseMetadata.serialize(&bob);
}

}  // namespace mongo